Included studies

Three hundred and sixty two studies of trauma-focused CBT for the treatment of PTSD in adults were identified for full-text review. Of these 362 studies, 88 RCTs (N=8450) were included. Many of these 88 RCTs were three- or four-armed trials and as such were included in more than one comparison. There were 17 comparisons for trauma-focused CBT.

For early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms, there was evidence for one relevant comparison: 2 RCTs (N=295) compared trauma-focused CBT with waitlist or no treatment (Bisson et al. 2004; Sijbrandij 2007).

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, 28 RCTs (N=2424) compared trauma-focused CBT with waitlist (Alghamdi et al. 2015; Blanchard 2002/Blanchard et al. 2003/Blanchard et al. 2003 [one study reported across three publications]; Bolton et al. 2014a; Buhmann et al. 2016; Chard, 2005; Cloitre et al. 2002; Difede et al. 2007b; Dunne et al. 2012; Ehlers et al. 2003; Ehlers et al. 2005; Ehlers et al. 2014; Falsetti et al. 2008; Fecteau & Nicki, 1999; Gersons et al. 2000; Hijazi et al. 2014; Hollifield et al. 2007; Jacob et al. 2014; Jung & Steil, 2013; Lindauer et al. 2005; Lindauer et al. 2008; McDonagh et al. 2005; Neuner et al. 2008; Pacella et al. 2012; Ruglass et al. 2017/ Hien 2011[one study reported across two publications]; van Emmerik et al. 2008; Weiss et al. 2015 (study 1); Weiss et al. 2015 (study 2); Zang et al. 2014). 36 RCTs (N=3257) compared trauma-focused CBT in addition to treatment as usual and/or medication, with treatment as usual or medication only (Akbarian et al. 2015; Asukai et al. 2010; Bass et al. 2013; Beck et al. 2009; Bohus et al. 2013; Brom et al. 1989; Buhmann et al. 2016; Coffey et al. 2016; Dorrepaal et al. 2012; Duffy et al. 2007; Foa et al. 2005; Foa et al. 2013b; Forbes et al. 2012; Hermenau et al. 2013; Hinton et al. 2005; Hinton et al. 2009; Kubany et al. 2003; Kubany et al. 2004; Maguen et al. 2017; Mills et al. 2012; Monson et al. 2006; Morath et al. 2014; Mueser et al. 2008; Neuner et al. 2004; Neuner et al. 2010; Pabst et al. 2014; Paunović, 2011; Popiel et al. 2015; Power et al. 2002; Resick et al. 2002; Rothbaum et al. 2005; Rothbaum et al. 2006; Ruglass et al. 2017/ Hien 2011[one study reported across two publications]; Sannibale et al. 2013; Stenmark et al. 2013; Wells et al. 2015). 6 RCTs (N=420) compared trauma-focused CBT (with or without additional treatment as usual) with eye movement desensitisation and reprocessing (EMDR; with or without additional treatment as usual) (Capezzani et al. 2013; Laugharne et al. 2016; Nijdam et al. 2012; Power et al. 2002; Rothbaum et al. 2005; Taylor et al. 2003). 4 RCTs (N=239) compared trauma-focused CBT (with or without additional treatment as usual) with non-trauma-focused CBT (with or without additional treatment as usual) (Cook et al. 2010; Foa et al. 1991; Hensel-Dittmann et al. 2011; Wells et al. 2015). 11 RCTs (N=903) compared trauma-focused CBT (with or without additional treatment as usual) with counselling (with or without additional treatment as usual) (Blanchard 2002/Blanchard et al. 2003/Blanchard et al. 2003 [one study reported across three publications]; Bryant et al. 2003a; Castillo et al. 2016; Cloitre et al. 2010; Cottraux et al. 2008; Ehlers et al. 2014; Foa et al. 1991; Katz et al. 2014; Nacasch et al. 2011; Neuner et al. 2004; Neuner et al. 2008). 7 RCTs (N=1152) compared trauma-focused CBT (with or without additional treatment as usual) with present-centred therapy (with or without additional treatment as usual) (Ghafoori et al. 2017; McDonagh et al. 2005; Rauch et al. 2015; Schnurr et al. 2003; Schnurr et al. 2007/Haug et al. 2004 [one study reported across two publications]; Sloan et al. 2016b/Sloan et al. unpublished [one study reported across two papers]; Surís et al. 2013). 1 RCT (N=110) compared trauma-focused CBT with interpersonal psychotherapy (IPT) (Markowitz et al. 2015a). 1 RCT (N=112) compared trauma-focused CBT (in addition to treatment as usual) with psychodynamic therapy (in addition to treatment as usual) (Brom et al. 1989). 2 RCTs (N=211) compared trauma-focused CBT (with or without additional treatment as usual) with self-help (without support; with or without additional treatment as usual) (Ehlers et al. 2003; Sloan et al. 2016a/2018 [one study reported across two publications]). 1 RCT (N=125) compared trauma-focused CBT with self-help with support (van Emmerik et al. 2008). 1 RCT (N=112) compared trauma-focused CBT (in addition to treatment as usual) with hypnotherapy (in addition to treatment as usual) (Brom et al. 1989). 1 RCT (N=690) compared trauma-focused CBT with a psychoeducational session (Chambers et al. 2014). 3 RCTs (N=194) compared trauma-focused CBT (with or without additional treatment as usual) with relaxation (with or without additional treatment as usual) (Hinton et al. 2011; Markowitz et al. 2015a; Taylor et al. 2003). 1 RCT (N=84) compared trauma-focused CBT with acupuncture (Hollifield et al. 2007). 3 RCTs (N=557) compared trauma-focused CBT with SSRIs (Buhmann et al. 2016; Echiverri-Cohen et al. 2016; Popiel et al. 2015). 1 RCT (N=280) compared combined trauma-focused CBT and SSRIs with waitlist (Buhmann et al. 2016).

Sub-analyses were possible, comparing effects by multiplicity of trauma, specific intervention, diagnostic status at baseline, and trauma type for the following delayed treatment comparisons: trauma-focused CBT versus waitlist; trauma-focused CBT in addition to treatment as usual or medication versus treatment as usual or medication only; trauma-focused CBT (with or without additional treatment as usual) versus EMDR (with or without additional treatment as usual).

Excluded studies

Two hundred and seventy four studies were reviewed at full text and excluded from this review. The most common reasons for exclusion were non-randomised group assignment, small sample size (less than 10 participants per arm), efficacy or safety data could not be extracted, comparison outside protocol (within-class comparison), subgroup or secondary analysis of an RCT already included and/or that is not relevant, and systematic review with no new useable data and any meta-analysis results not appropriate to extract.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

Forty-four studies of non-trauma-focused CBT for the treatment of PTSD in adults were identified for full-text review. Of these 44 studies, 13 RCTs (N=1316) were included. There were 5 comparisons for non-trauma-focused CBT. One RCT was included in two comparisons of non-trauma-focused CBT.

There were no studies for early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms.

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, 9 RCTs (N=737) compared non-trauma-focused CBT (alone or in addition to treatment as usual) with waitlist or treatment as usual (Davis & Wright 2007; Davis et al. 2011; Ford et al. 2011; Krakow et al. 2000; Margolies et al. 2013; McGovern et al. 2011; McGovern et al. 2015; Talbot et al. 2014; Zlotnick et al. 1997); 2 RCTs (N=413) compared non-trauma-focused CBT (alone or in addition to treatment as usual) with attention-placebo (alone or in addition to treatment as usual) (Hien et al. 2009; Nakamura et al. 2017); 1 RCT (N=111) compared non-trauma-focused CBT (in addition to treatment as usual) with a psychoeducational group (in addition to treatment as usual) (Dunn et al. 2007); 1 RCT (N=55) compared non-trauma-focused CBT with counselling (Foa et al. 1991); 1 RCT (N=146) compared non-trauma-focused CBT with present-centred therapy (Ford et al. 2011).

Comparisons with trauma-focused CBT are presented in the Trauma-focused CBT section above.

Sub-analyses were possible for the delayed treatment non-trauma-focused CBT (alone or in addition to TAU) versus waitlist or TAU comparison, comparing effects by multiplicity of trauma, specific intervention, diagnostic status at baseline, and trauma type.

Excluded studies

Thirty-one studies were reviewed at full text and excluded from this review. The most common reasons for exclusion were subgroup or secondary analysis of an RCT already included and/or that is not relevant, systematic review with no new useable data and any meta-analysis results not appropriate to extract, and intervention not targeted at PTSD symptoms.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

Four studies of present-centred therapy for the treatment of PTSD in adults were identified for full-text review. Of these 4 studies, all 4 RCTs (N=350) were included. There were 2 comparisons for present-centred therapy.

For early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms, 2 RCTs (N=130) compared present-centred therapy in addition to treatment as usual with treatment as usual-only (Johnson et al. 2011; Johnson et al. 2016).

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, 2 RCTs (N=220) compared present-centred therapy with waitlist (Ford et al. 2011; McDonagh et al. 2005).

Comparisons with trauma-focused CBT are presented in the Trauma-focused CBT section above.

Sub-analyses were not possible for present-centred therapy.

Excluded studies

No present-centred studies that were considered in full-text were excluded.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

Twenty-five studies of cognitive therapies for the treatment of PTSD in adults were identified for full-text review. Of these 25 studies, 4 RCTs (N=156) were included. There were 2 comparisons for cognitive therapies.

For early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms, no relevant RCTs were identified.

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, there was evidence for 2 relevant comparisons: 2 RCTs (N=52) compared metacognitive therapy (alone or in addition to TAU) with waitlist or TAU (Wells & Colbear 2012; Wells et al. 2015); 2 RCTs (N=104) compared reconsolidation of traumatic memories (RTM) intervention in addition to TAU with TAU-only (Gray et al. 2017; Tylee et al. 2017).

Comparisons with trauma-focused CBT are presented in the Trauma-focused CBT section above.

Sub-analyses were not possible for cognitive therapies.

Excluded studies

Twenty-one studies were reviewed at full text and excluded from this review. The most common reasons for exclusion were systematic review with no new useable data and any meta-analysis results not appropriate to extract, non-systematic review, intervention not targeted at PTSD symptoms, and non-randomised group assignment.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

Eleven studies of behavioural therapies for the treatment of PTSD in adults were identified for full-text review. Of these 11 studies, 2 RCTs (N=90) were included. There was 1 comparison for behavioural therapies.

For early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms, no relevant RCTs were identified.

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, both RCTs (N=90) compared single-session behavioural therapy with waitlist (Başoğlu et al. 2005; Başoğlu et al. 2007).

Sub-analyses were not possible for behavioural therapies.

Excluded studies

Nine studies were reviewed at full text and excluded from this review. The most common reason for exclusion was non-randomised group assignment.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

One study of problem solving for the treatment of PTSD in adults was identified for full-text review, and this RCT (N=309) was included in a single comparison for problem solving.

For early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms, the single included RCT (N=309) compared problem solving with supportive counselling (Sahler et al. 2013).

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, no relevant RCTs were identified.

Sub-analyses were not possible for problem solving interventions.

Excluded studies

There were no studies that met criteria for full-text review that were excluded.

Included studies

Fifty-three studies of eye movement desensitisation and reprocessing (EMDR) for the treatment of PTSD in adults were identified for full-text review. Of these 53 studies, 17 RCTs (N=1009) were included. Some of these 17 RCTs were three- or four-armed trials and as such were included in more than one comparison. There were 9 comparisons for EMDR.

For early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms, there was evidence for one relevant comparison: 1 RCT (N=39) compared EMDR with supportive counselling (Jarero et al. 2013).

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, 1 RCT (N=88) compared EMDR with pill placebo (Van der Kolk et al. 2007; this trial also included a fluoxetine arm). 10 RCTs (N=585) compared EMDR (alone or in addition to TAU) with waitlist or TAU (Acarturk et al. 2015; Acarturk et al. 2016; Aldahadha et al. 2012; Carlson et al. 1998; Edmond et al. 1999/ Edmond & Rubin 2004 [one study reported across two papers]; Himmerich et al. 2016; Jensen 1994; Power et al. 2002; Rothbaum et al. 2005; Yurtsever et al. 2018). 1 RCT (N=67) compared EMDR with supportive counselling (Scheck et al. 1998). 1 RCT (N=74) compared EMDR with non-trauma-focused CBT (Ter Heide et al. 2016). 1 RCT (N=59) compared EMDR with ‘other active psych intervention’ (Edmond et al. 1999/ Edmond & Rubin 2004 [one study reported across two papers]). 3 RCTs (N=145) compared EMDR (alone or in addition to TAU) with relaxation (alone or in addition to TAU) (Carletto et al. 2016; Carlson et al. 1998; Taylor et al. 2003). 1 RCT (N=46) compared EMDR with a combined somatic and cognitive therapy, emotional freedom technique (EFT) (Karatzias et al. 2011). Finally, 1 RCT (N=88) compared EMDR with fluoxetine (Van der Kolk et al. 2007).

Comparisons with trauma-focused CBT are presented in the Trauma-focused CBT section above.

Sub-analyses were possible for the delayed treatment EMDR (alone or in addition to TAU) versus waitlist or TAU comparison, comparing effects by multiplicity of trauma, diagnostic status at baseline, and trauma type.

Excluded studies

Thirty-six studies were reviewed at full text and excluded from this review. The most common reasons for exclusion were non-randomised group assignment, small sample size (N<10 per arm), efficacy or safety data could not be extracted, and systematic review with no new useable data and any meta-analysis results not appropriate to extract.

Studies not included in this review with reasons for their exclusions are provided in Appendix L.

Included studies

Seven studies of hypnotherapy for the treatment of PTSD in adults were identified for full-text review. Of these 7 studies, 3 RCTs (N=253) were included, and each involved a different comparison, so there were 3 comparisons for hypnotherapy.

For early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms, there were no relevant RCTs included.

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, 1 RCT (N=112) compared hypnotherapy in addition to TAU with TAU-only (Brom et al. 1989), 1 RCT (N=108) compared hypnotherapy followed by trauma-focused CBT with symptom monitoring followed by trauma-focused CBT (Galovski 2008/ Galovski et al. 2016 [protocol and paper]), and 1 RCT (N=33) compared hypnotherapy (in addition to TAU) with zolpidem (in addition to TAU) (Abramowitz et al. 2008).

Comparisons with trauma-focused CBT are presented in the Trauma-focused CBT section above.

Sub-analyses were not possible for hypnotherapy.

Excluded studies

Four studies were reviewed at full text and excluded from this review. The most common reason for exclusion was systematic review with no new useable data and any meta-analysis results not appropriate to extract.

Studies not included in this review with reasons for their exclusions are provided in Appendix L.

Included studies

Four studies of interpersonal psychotherapy (IPT) for the treatment of PTSD in adults were identified for full-text review. Of these 4 studies, 2 RCTs (N=158) were included, and each involved a different comparison, so there were 2 comparisons for IPT.

For early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms, there were no relevant RCTs included.

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, 1 RCT (N=48) compared IPT with waitlist (Krupnick et al. 2008), and 1 RCT (N=110) compared IPT with relaxation (Markowitz et al. 2015).

Comparisons with trauma-focused CBT are presented in the Trauma-focused CBT section above.

Sub-analyses were not possible for IPT.

Excluded studies

Two studies were reviewed at full text and excluded from this review due to small sample size (N<10 per arm) or subgroup/secondary analysis of RCT already included.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

Twelve studies of psychodynamic therapies for the treatment of PTSD in adults were identified for full-text review. Of these 12 studies, 2 RCTs (N=198) were included in 1 comparison for psychodynamic therapies.

For early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms, there were no relevant RCTs included.

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, 2 RCTs (N=198) compared psychodynamic therapy (alone or in addition to TAU) with waitlist (alone or in addition to TAU) (Brom et al. 1989; Steinert et al. 2017).

Comparisons with trauma-focused CBT are presented in the Trauma-focused CBT section above.

Sub-analyses were not possible for psychodynamic therapies.

Excluded studies

Ten studies were reviewed at full text and excluded from this review. The most common reasons for exclusion were non-randomised group assignment, nonsystematic review, and paper unavailable.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

Thirteen studies of counselling for the treatment of PTSD in adults were identified for full-text review. Of these 13 studies, 6 RCTs (N=842) were included in 1 comparison for counselling.

For early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms, there were no relevant RCTs included.

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, 6 RCTs (N=842) compared counselling (alone or in addition to TAU) with TAU or waitlist (Bass et al. 2016; Blanchard 2002/Blanchard et al. 2003/2004 [one study reported across three papers]; Ehlers et al. 2014; Neuner et al. 2004; Neuner et al. 2008; Yeomans et al. 2010).

Comparisons with trauma-focused CBT are presented in the Trauma-focused CBT section above.

Sub-analyses were not possible for counselling.

Excluded studies

Seven studies were reviewed at full text and excluded from this review. The most common reason for exclusion was that the comparison was outside protocol (within-class comparison).

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

Seven studies of combined somatic and cognitive therapies for the treatment of PTSD in adults were identified for full-text review. Of these 7 studies, 4 RCTs (N=544) were included in 1 comparison for combined somatic and cognitive therapies.

For early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms, there were no relevant RCTs included.

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, 4 RCTs (N=544) compared combined somatic and cognitive therapies (alone or in addition to TAU) with waitlist (alone or in addition to TAU) (Church et al. 2013/ Church 2014 [one study reported across two papers]; Connolly & Sakai 2011; Geronilla et al. 2016; Robson et al. 2016).

Sub-analyses were possible for this comparison, comparing effects by specific intervention and trauma type. Sub-analyses by multiplicity of trauma or diagnostic status at baseline were not possible as there is only one sub-group.

Excluded studies

Three studies were reviewed at full text and excluded from this review due to small sample size (N<10 per arm), non-randomised group assignment, or paper unavailable.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

One study of somatic experiencing for the treatment of PTSD in adults was identified for full-text review. This 1 RCT (N=63) was included in a single comparison: somatic experiencing in addition to TAU relative to TAU-only for the delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms (Brom et al. 2017).

Excluded studies

No somatic experiencing studies that were considered in full-text were excluded.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

Two studies of resilience-oriented treatment for the treatment of PTSD in adults were identified for full-text review. Of these 2 studies, 1 RCT (N=39) was included in 1 comparison for resilience-oriented treatment.

For early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms, there were no relevant RCTs included.

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, 1 RCT (N=39) compared resilience-oriented treatment with waitlist (Kent et al. 2011).

Sub-analyses were not possible for resilience-oriented treatment.

Excluded studies

One study was reviewed at full text and excluded from this review as efficacy or safety data cannot be extracted.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

Six studies of attention bias modification for the treatment of PTSD in adults were identified for full-text review. Of these 6 studies, 3 RCTs (N=200) were included in 1 comparison for attention bias modification.

For early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms, there were no relevant RCTs included.

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, 3 RCTs (N=200) compared attention bias modification with attention-placebo (Bar-Haim & Fruchter 2011/ Badura-Brack et al. 2015 study 1 [protocol and paper]; Bar-Haim & Fruchter 2011/ Badura-Brack et al. 2015 study 2 [protocol and paper]; Schoorl et al. 2013).

Sub-analyses were not possible for attention bias modification.

Excluded studies

Three studies were reviewed at full text and excluded from this review due to small sample size (N<10 per arm), subgroup/secondary analysis of RCT already included, or trials of soldiers on active service (population outside scope).

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

Nine studies of couple interventions for the treatment of PTSD in adults were identified for full-text review. Of these 9 studies, 2 RCTs (N=97) were included. There were 2 comparisons for couple interventions.

For early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms, no relevant RCTs were included.

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, 1 RCT (N=40) compared a couple intervention with waitlist (Monson & Vorstenbosch 2008/Monson et al. 2012 [protocol and paper]), and 1 RCT (N=57) compared a couple intervention with psychoeducational sessions (Sautter et al. 2015).

Sub-analyses were not possible for couple interventions.

Excluded studies

Seven studies were reviewed at full text and excluded from this review. The most common reason for exclusion was subgroup/secondary analysis of RCT already included and/or that is not relevant.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

Two studies of family interventions for the treatment of PTSD in adults were identified for full-text review. Of these 2 studies, both RCTs (N=221) were included. There were 2 comparisons for family interventions.

For early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms, no relevant RCTs were included.

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, 1 RCT (N=146) compared family therapy with waitlist (Kazak et al. 2004), and 1 RCT (N=75) compared child-parent psychotherapy (using play) with case management and individual treatment (for parent-only) (Lieberman et al. 2005/2006/Ghosh Ippen et al. 2011 [one study reported across three papers]).

Sub-analyses were not possible for family interventions.

Excluded studies

No family intervention studies that were considered in full-text were excluded.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

Seventeen studies of self-help with support for the treatment of PTSD in adults were identified for full-text review. Of these 17 studies, 9 RCTs (N=885) were included. There were 2 comparisons for self-help with support.

There were no studies for early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms.

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, 8 RCTs (N=798) compared self-help with support (alone or in addition to TAU) with waitlist or TAU (Ivarsson et al. 2014; Knaevelsrud & Maercker 2007; Knaevelsrud et al. 2015; Knaevelsrud et al. 2017; Lange et al. 2003; Lewis et al. 2017; van Dam et al. 2013; Van Emmerik et al. 2008), and 1 RCT (N=87) compared self-help with support with self-help without support (Littleton et al. 2016).

Comparisons with trauma-focused CBT are presented in the Trauma-focused CBT section above.

Sub-analyses were possible for the delayed treatment self-help with support (alone or in addition to TAU) versus waitlist or TAU, comparing effects by multiplicity of trauma, specific intervention, diagnostic status at baseline, trauma type, and baseline severity.

Excluded studies

Eight studies were reviewed at full text and excluded from this review. The most common reason for exclusion was that the comparison was outside protocol (within-class comparison).

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

Forty-two studies of self-help (without support) for the treatment of PTSD in adults were identified for full-text review. Of these 39 studies, 13 RCTs (N=904) were included. There were 2 comparisons for self-help (without support).

There were no studies for early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms.

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, 7 RCTs (N=462) compared self-help (without support) with waitlist (Ehlers et al. 2003; Hirai & Clum 2005; Kuhn et al. 2017; Miner et al. 2016; Sloan et al. 2012; Spence et al. 2011; Xu et al. 2016), and 6 RCTs (N=442) compared self-help (without support) with attention-placebo (Henderson et al. 2007; Meshberg-Cohen et al. 2014; Sloan & Marx 2004; Sloan et al. 2007; Sloan et al. 2011; Truijens & van Emmerik 2014).

Comparisons with trauma-focused CBT are presented in the Trauma-focused CBT section above.

Sub-analyses were possible for the delayed treatment self-help (without support) versus waitlist, or self-help (without support) versus attention-placebo, comparing effects by multiplicity of trauma, specific intervention, diagnostic status at baseline, trauma type, and baseline severity.

Excluded studies

Twenty-nine studies were reviewed at full text and excluded from this review. The most common reasons for exclusion were systematic review with no new useable data and any meta-analysis results not appropriate to extract, the comparison was outside the protocol (within-class comparison), or efficacy or safety data could not be extracted.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

The systematic search of economic literature identified 5 studies that assessed the cost effectiveness of psychological interventions for the treatment of adults with PTSD (Chatterton et al., 2016; Dunn et al., 2007; Le et al., 2014; Mihalopoulos et al., 2015; Tuerk et al., 2013); one of the studies (Le et al., 2014) was a comparison between a psychological and a pharmacological intervention. The search strategy for economic studies is provided in Appendix B.

Excluded studies

Four economic studies were reviewed at full text and excluded from this review. The reasons for exclusion were: assessment of a mixture of interventions (“optimal” versus “current” treatment), lack of reporting of results for each arm, >50% of population having psychosis, and military setting.

Studies not included in this review with reasons for their exclusion are provided in Appendix K.

Overview of economic modelling methods

A hybrid decision-analytic model consisting of a decision-tree followed by a three-state Markov model was constructed to evaluate the relative cost effectiveness of a range of interventions for the treatment of adults with PTSD in a community setting. The time horizon of the analysis was 3 years, consisting of the 6 months of the decision tree and another 2.5 years (10 x 3-month cycles) in the Markov component of the economic model. The range of interventions assessed in the economic analysis was determined by the availability of relevant clinical data included in the guideline systematic review of interventions for the treatment of adults with clinically important PTSD symptoms. Network meta-analysis (NMA) was employed for synthesis of the available efficacy data. The guideline economic analysis assessed psychological, pharmacological and combined psychological and pharmacological interventions that were connected to the network of evidence and were thus possible to include in the NMA. Based on the advice of the committee, only effective interventions that had been tested on at least 50 people across the RCTs included in the NMAs assessing efficacy at treatment endpoint were considered in the economic analysis, as this was deemed as the minimum evidence that would be adequate to support a practice recommendation. Interventions that belonged to the trauma-focused cognitive behavioural therapy (TF-CBT) class were not considered separately according to their type, as the description of the type of TF-CBT was not always clear in the publications, and in some studies the intervention included elements of more types of TF-CBT. However, based on reported resource use in each RCT included in the NMA, TF-CBT interventions were categorised according to their mode of delivery in individual, group and mixed (where the intervention was delivered by a combination of individual and group sessions). Each of these categories was further subdivided, as relevant, to those comprising fewer than 8 sessions, 8-12 sessions, and more than 12 sessions, and were considered separately in the NMA and the economic analysis, to reflect the different intervention costs and, potentially, different efficacy associated with each sub-category

Based on the available evidence, the following interventions were considered in the economic analysis of interventions for the treatment of adults with PTSD:

• Psychoeducation
• Counselling
• TF-CBT individual <8 sessions
• TF-CBT individual 8-12 sessions
• TF-CBT individual >12 sessions
• TF-CBT group 8-12 sessions
• non-TF-CBT
• Eye Movement Desensitisation Reprocessing (EMDR)
• Present-centred therapy
• Interpersonal psychotherapy
• Combined somatic and cognitive therapies
• Self-help with support
• Self-help without support
• Selective serotonin reuptake inhibitors (SSRIs)
• TF-CBT individual 8-12 sessions + SSRIs
• No treatment, reflected in the waitlist arms of RCTs included in the guideline systematic review and NMA.

According to the model structure, hypothetical cohorts of adults with PTSD were initiated on each of the treatment options assessed, including no treatment. Following a course of treatment, people in each cohort either remitted (that is, they did not meet criteria for a PTSD diagnosis) or did not remit. In the 3 months of follow-up after treatment completion, people who remitted could remain in remission, relapse to a PTSD state or die. Those who did not remit, could remain in the PTSD state, remit or die. After that point, people in each cohort, both those who remitted and those who did not remit, were entered into the Markov component of the economic model, in either the ‘PTSD’ or the ‘no PTSD’ health states, depending on their state at the end of the decision-tree. In each cycle of the Markov model, they could remain in the same health state or move between the two states of ‘PTSD’ and ‘no PTSD’ or move to the death state (absorbing state).

Efficacy data were derived from the guideline systematic review and NMAs; other clinical input parameters (baseline risk of remission, risk of relapse, probability of developing side effects from SSRIs, mortality) were derived from published literature and the committee’s expert opinion where evidence was lacking. The measure of outcome of the economic analysis was the number of QALYs gained. Utility data were selected after a systematic review of the literature. The perspective of the analysis was that of health and personal social care services. Resource use was based on published literature, national statistics and, where evidence was lacking, the committee’s expert opinion. National UK unit costs were used. The cost year was 2017. Model input parameters were synthesised in a probabilistic analysis. This approach allowed more comprehensive consideration of the uncertainty characterising the input parameters and captured the non-linearity characterising the economic model structure. Three probabilistic analyses were carried out:

• Analysis A (base-case) utilised efficacy data at treatment endpoint from a NMA of continuous data (changes in PTSD symptom scores), transformed to log-odds ratios of remission, and assumed no beneficial effect of interventions beyond treatment endpoint
• Analysis B utilised efficacy data at treatment endpoint from the NMA of continuous data (changes in PTSD symptom scores), transformed to log-odds ratios of remission, and efficacy data at 3 months post-treatment from the NMA of changes in PTSD symptom scores between baseline and 1-4 month follow-up, also transformed to log-odds ratios of remission
• Analysis C utilised efficacy data at treatment endpoint from the NMA of dichotomous remission data; the probability of remission of all active interventions at 3-6 months was assumed to equal that of no treatment, as dichotomous remission follow-up data were very limited.

A number of one-way deterministic sensitivity analyses were also carried out.

Results have been expressed in the form of Incremental Cost Effectiveness Ratios (ICERs) following the principles of incremental analysis. Net Monetary Benefits (NMBs) have also been estimated. Incremental mean costs and effects (QALYs) of each intervention versus no treatment have been presented in the form of cost effectiveness planes. Results of probabilistic analysis have been summarised in the form of cost effectiveness acceptability curves (CEACs), which express the probability of each intervention being cost effective at various cost effectiveness thresholds. Cost effectiveness acceptability frontiers (CEAFs) have also been plotted; these show the treatment option with the highest mean NMB over different cost effectiveness thresholds, and the probability that the option with the highest NMB is the most cost-effective among those assessed.

Overview of economic modelling results and conclusions

In the base-case analysis (which utilised continuous data at treatment endpoint and assumed no treatment effect beyond treatment endpoint), the order of interventions from the most to the least cost-effective for the treatment of PTSD in adults was: TF-CBT individual < 8 sessions, psychoeducation, EMDR, combined somatic and cognitive therapies, self-help with support, SSRI, self-help without support, TF-CBT individual 8-12 sessions, IPT, non-TF-CBT, present-centred therapy, TF-CBT group 8-12 sessions, combined TF-CBT individual 8-12 sessions + SSRI, no treatment, TF-CBT individual >12 sessions, and counselling. The probability of TF-CBT individual < 8 sessions being the most cost-effective treatment option was 0.28.

When a beneficial effect of up to 3 months post-treatment was assumed, there were no dramatic changes in the results; the ranking of combined somatic and cognitive therapies, self-help without support and IPT improved by one place, whereas EMDR and TF-CBT individual 8-12 sessions dropped one place in ranking. The order of interventions became TF-CBT individual < 8 sessions, psychoeducation, combined somatic and cognitive therapies, EMDR, self-help with support, self-help without support, SSRI, IPT, TF-CBT individual 8-12 sessions, non-TF-CBT, TF-CBT individual >12 sessions, present-centred therapy, TF-CBT group 8-12 sessions, TF-CBT individual 8-12 sessions + SSRI, counselling, and no treatment. The probability of TF-CBT individual < 8 sessions being the most cost-effective treatment option was 0.18.

When dichotomous remission data were used, there were more important changes in the results with non-TF-CBT becoming the most cost-effective intervention followed by EMDR, TF-CBT individual 8-12 sessions, IPT, SSRI, self-help without support, self-help with support, present-centred therapy, TF-CBT individual 8-12 sessions + SSRI, TF-CBT individual >12 sessions, counselling, TF-CBT group 8-12 sessions, and no treatment. The probability of non-TF-CBT being the most cost-effective treatment was 0.42.

Results of the economic analysis were robust to changes in input parameters tested in deterministic sensitivity analysis.

The guideline base-case economic analysis is based on the best quality efficacy data derived from NMA. However, the result for psychoeducation, which was found to be among the most cost-effective interventions, should be interpreted with great caution due to limitations in the evidence base and the considerably high uncertainty characterising its efficacy estimate. Moreover, the NMA that informed the base-case analysis was characterised by high between-study heterogeneity, as well as large effects and considerable uncertainty for some interventions, and this should be taken into account when interpreting the results of the analysis.

Results from the alternative scenarios explored in the other two probabilistic analyses (i.e. consideration of efficacy data derived from the NMAs of continuous 1-4 month follow-up data and of dichotomous remission data) should also be interpreted with caution due to the limitations characterising the respective evidence base and the NMAs that informed them (limited evidence base, evidence of inconsistency between direct and indirect evidence, high between-study heterogeneity, large effects and considerable uncertainty for some interventions).

Trauma-focused CBT for early treatment (1-3 months)

• Low to very low quality single-RCT evidence (N=152) suggests a statistically significant small-to-moderate delayed benefit (significant only at 10-month follow-up) of early treatment with trauma-focused CBT (initiated 1-3 months after trauma), relative to no treatment, on improving self-rated PTSD symptomatology and the rate of response, in adults with PTSD. Very low quality evidence from another single RCT (N=143) suggests a clinically important benefit, that just misses statistical significance, of trauma-focused CBT on the rate of remission at endpoint. However, this effect is not maintained at 4-month follow-up. Moderate to very low quality evidence from 1-2 RCTs (N=98-265) suggests neither clinically important nor statistically significant effects on clinician-rated PTSD symptomatology at endpoint, 4-month or 10-month follow-up. No clinically important and statistically significant effects were observed for anxiety symptoms, depression symptoms or discontinuation

Trauma-focused CBT for delayed treatment (>3 months)

• Very low quality evidence from 12-14 RCTs (N=618-632) suggests a large and statistically significant benefit of trauma-focused CBT, relative to waitlist, on improving PTSD symptomatology (self-rated and clinician-rated) in adults with PTSD over 3 months after trauma. Evidence from 1-2 RCTs (N=63-145) suggests benefits on self-rated PTSD symptomatology are maintained up to 1-year follow-up (with the exception of a non-significant effect at 3-months), and evidence from 4 RCTs (N=507) suggests benefits on clinician-rated PTSD symptomatology are maintained up to 3-5 month follow-up (longest follow-up). Very low quality evidence from 14 RCTs (N=628) suggests a clinically important and statistically significant benefit of trauma-focused CBT on the rate of remission, and evidence from 1-3 RCTs (N=166-175) suggests benefits are maintained at 3-6 month and 8-month follow-up. Low quality evidence from 3 RCTs (N=89-111) suggests a clinically important and statistically significant benefit of trauma-focused CBT on the rate of response (based on self-rated and clinician-rated measures) and single-RCT (N=57) evidence suggests this effect (self-rated) is maintained at 6-month follow-up. Very low quality evidence from 15-19 RCTs (N=760-972) suggests large and statistically significant benefits of trauma-focused CBT on anxiety and depression symptoms and evidence from 1-5 RCTs (N=82-550) suggests these benefits are maintained up to 1-year follow-up. In addition, there is low to very low quality evidence from 1-6 RCTs (N=46-339) for large and statistically significant benefits of trauma-focused CBT on dissociative symptoms, global functioning, functional impairment and relationship difficulties at endpoint, although the evidence for follow-up is more limited. However, in addition to the considerable evidence for benefit of trauma-focused CBT relative to waitlist for the delayed treatment of PTSD, there is low quality evidence from 26 RCTs (N=1834) for higher drop-out associated with trauma-focused CBT relative to waitlist. There is also very high heterogeneity observed across outcomes. Sub-analyses by specific intervention suggests some differential effects but within-subgroup heterogeneity remains high and benefits are observed across all interventions (although statistical significance varies). Sub-analyses by diagnostic status at baseline suggests larger effect sizes for those with a diagnosis at baseline but again within-subgroup heterogeneity is high. Finally, sub-analyses by trauma type suggests some differences with larger effects associated with some trauma types but these are difficult to disentangle as the larger effects are associated with the single smaller study subgroups.
• Very low quality evidence from 21-22 RCTs (N=1179-1640) suggests a large and statistically significant benefit of trauma-focused CBT (in addition to medication or TAU), relative to medication or TAU-only, on improving PTSD symptomatology (self-rated and clinician-rated) in adults with PTSD over 3 months after trauma. Low to very low quality evidence from 2-7 RCTs (N=94-648) suggests large and statistically significant benefits are maintained up to 6-month follow-up, and clinically important (but not statistically significant) benefits are maintained up to 1-year follow-up. Very low quality evidence from 12 RCTs (N=917) suggests a clinically important and statistically significant benefit of trauma-focused CBT (in addition to TAU or medication) on the rate of remission, and moderate to low quality evidence from 4 RCTs (N=324) suggests clinically important and statistically significant benefits are maintained at 6-month follow-up, and clinically important (but not statistically significant) benefits are observed at 1-3 month and 1-year follow-ups. Very low quality evidence from 4-5 RCTs (N=245-328) suggests a clinically important and statistically significant benefit of trauma-focused CBT (in addition to TAU or medication) on the rate of response (based on self-rated and clinician-rated measures) at endpoint, with some evidence that this benefit is maintained at 1-6 month follow-up. Very low quality evidence from 13-22 RCTs (N=647-1536) suggests moderate-to-large and statistically significant benefits of trauma-focused CBT (in addition to TAU or medication) on anxiety and depression symptoms, although evidence for effects at follow-ups are less consistent. In addition, low to very low quality evidence from 1-5 RCTs (N=59-295) suggests large and statistically significant benefits of trauma-focused CBT (in addition to TAU or medication) on dissociative symptoms, personality disorder symptoms, global functioning, functional impairment and relationship difficulties, and low quality evidence from 2 RCTs (N=89) suggests a small-to-moderate benefit on anger/aggression. Evidence for effects on substance misuse outcomes are more mixed but for at least some of these studies the comparator is standard substance misuse services. Moderate quality evidence from 35 RCTs (N=2764) suggests higher drop-out associated with trauma-focused CBT, however, although this effect is statistically significant it does not meet the threshold for clinical importance. Heterogeneity across outcomes is very high. Sub-analyses by specific intervention suggests some differential effects but within-subgroup heterogeneity remains high and benefits are observed across all interventions (although statistical significance varies). Sub-analyses by diagnostic status at baseline was non-significant. Sub-analyses by trauma type suggests some differences but again within-subgroup heterogeneity remains high.
• Very low quality evidence from 6 RCTs (N=277-321) suggests moderate to large benefits of trauma-focused CBT, relative to counselling, on improving self-rated and clinician-rated PTSD symptomatology at endpoint in adults with PTSD over 3 months after trauma. Low to very low quality evidence from 1-5 RCTs (N=39-434) suggests clinically important and statistically significant effects are maintained up to 2-year follow-up for clinician-rated PTSD symptomatology. Effects on self-rated PTSD symptomatology are not statistically significant at follow-up although a trend remains up to 2-year follow-up. Low quality evidence from 6 RCTs (N=320) suggests a clinically important and statistically significant benefit of trauma-focused CBT on remission at endpoint and low to very low quality evidence from 2-5 RCTs (N=70-472) suggests that this effect is maintained up to 1-year follow-up. Low quality evidence from 8 RCTs (N=358) suggests a large and statistically significant benefit of trauma-focused CBT on anxiety symptoms at endpoint that is maintained up to 2-year follow-up. Evidence from these same 8 RCTs also suggests a small-to-moderate but statistically significant benefit of trauma-focused CBT on depression symptoms at endpoint and 6-8 month follow-up but effects are neither clinically important nor statistically significant at 3-months, 1-year or 2-year follow-ups. Low to very low quality evidence from 1-3 RCTs (N=61-175) suggests a moderate to large benefit of trauma-focused CBT on quality of life at endpoint that is maintained up to 1-year follow-up. Low quality evidence from single-RCT (N=39-61) analyses also suggests large and statistically significant benefits of trauma-focused CBT on functional impairment (maintained up to 6-month follow-up [longest follow-up]) and global functioning (maintained up to 1-year but not 2-year follow-up), and delayed large benefits (significant at 3- and 6-month follow-up but not endpoint) on relationship difficulties. Very low quality single-RCT (N=28) evidence suggests no statistically significant difference between trauma-focused CBT and counselling for response. Low quality evidence from 11 RCTs (N=754) suggests a neither clinically important nor statistically significant difference between trauma-focused CBT and counselling for discontinuation.
• Low to very low quality evidence from 1-6 RCTs (N=86-970) suggests clinically important and statistically significant benefits of trauma-focused CBT (alone or in addition to TAU), relative to present-centred therapy (alone or in addition to TAU), on improving PTSD symptomatology (clinician-rated) at endpoint and up to 6-month follow-up, the rate of remission at 1-3 month follow-up (clinically important that just misses statistical significance at endpoint but non-significant at 6-month follow-up), and depression symptoms at 2-3 month, 4-month and 6-month follow-ups (non-significant at endpoint) in adults with PTSD over 3 months after trauma. The effect on self-rated PTSD symptomatology is also clinically important but just misses statistical significance (p=0.05). Moderate to very low quality evidence from 1-3 RCTs (N=34-680) suggests no statistically significant differences between trauma-focused CBT and present-centred therapy on clinician-rated response, dissociative symptoms, anxiety symptoms, anger, or quality of life, at endpoint or 3- or 6- month follow-up. Low quality evidence from 6 RCTs (N=931) suggests higher drop-out associated with trauma-focused CBT relative to present-centred therapy, however this effect is not statistically significant.
• Low quality single-RCT (N=40) evidence suggests a clinically important benefit, that just misses statistical significance (p=0.06), of trauma-focused CBT relative to interpersonal psychotherapy (IPT) on improving self-rated PTSD symptomatology in adults with PTSD over 3 months after trauma. However, evidence from the same RCT (N=37-78) suggests neither clinically important nor statistically significant differences between trauma-focused CBT and IPT on clinician-rated PTSD symptomatology, remission, response, functional impairment, or relationship difficulties. This study (N=39-63) did find evidence for clinically important and statistically significant benefits of trauma-focused CBT relative to IPT on depression symptoms and quality of life. There is some evidence from the same RCT for higher drop-out with trauma-focused CBT relative to IPT, however, this effect is not statistically significant.
• Moderate to very low quality evidence from 1-2 RCTs (N=53-182) suggests clinically important and statistically significant benefits of trauma-focused CBT (alone or in addition to TAU) relative to self-help without support (alone or in addition to TAU) on clinician-rated PTSD symptomatology, remission at 6-month follow-up (clinically important but not statistically significant at endpoint), response (at endpoint and 6-month follow-up), depression symptoms (at endpoint and 6-month follow-up), anxiety symptoms (at endpoint and 6-month follow-up), and functional impairment (at endpoint and 6-month follow-up) in adults with PTSD over 3 months after trauma. Very low quality evidence from both RCTs (N=182) suggests there may be higher drop-out associated with trauma-focused CBT relative to self-help without support, however, heterogeneity is very high and this effect is not statistically significant.
• Very low quality single-RCT (N=230-244) evidence suggests neither clinically important nor statistically significant differences between brief trauma-focused CBT and a psychoeducational session on self-rated PTSD symptomatology at endpoint, 3- and 6- month follow-up, in adults with PTSD over 3 months after trauma. Although, low quality evidence from this same RCT (N=336) does suggest a higher rate of discontinuation associated with trauma-focused CBT relative to psychoeducation.
• Low to very low quality evidence from 1-3 RCTs (N=24-84) suggests moderate to large and statistically significant benefits of trauma-focused CBT (alone or in addition to TAU) relative to relaxation (alone or in addition to TAU) on improving PTSD symptomatology (self-rated and clinician-rated) and anxiety symptoms at endpoint and 3-month follow-up, and functional impairment, quality of life, and relationship difficulties at endpoint (no follow-up data), in adults with PTSD over 3 months after trauma. Low to very low quality evidence from 1-2 RCTs (N=30-111) suggests clinically important but not statistically significant benefits of trauma-focused CBT on remission (at endpoint and 3-month follow-up), response, and dissociative symptoms at 3-month follow-up (non-significant effect at endpoint). Low to very low quality evidence from 1-3 RCTs (N=30-135) suggests non-significant differences between trauma-focused CBT and relaxation for depression symptoms at endpoint and 3-month follow-up, and discontinuation.
• Very low quality single-RCT (N=49-57) evidence suggests non-significant differences between trauma-focused CBT and acupuncture for self-rated PTSD symptomatology, remission, depression symptoms, anxiety symptoms, functional impairment, and discontinuation, in adults with PTSD over 3 months after trauma.
• Very low quality evidence from 2-3 RCTs (N=161-275) suggests small but statistically significant benefits of SSRIs relative to trauma-focused CBT on improving self-rated PTSD symptomatology and anxiety symptoms at endpoint in adults with PTSD over 3 months after trauma. Although, very low quality evidence from 1 of these RCTs (N=112) suggests effects are not maintained at 1-year follow-up. Conversely, low to very low quality evidence from 1-2 RCTs (N=49-171) suggests large and statistically significant benefits of trauma-focused CBT relative to SSRIs on clinician-rated PTSD symptomatology, remission, and dissociative symptoms at endpoint (no follow-up available). Very low quality evidence from 1-3 RCTs (N=112-275) suggests neither clinically important nor statistically significant differences between trauma-focused CBT and SSRIs on depression symptoms at endpoint and 1-year follow-up, functional impairment and quality of life at endpoint (no follow-up available), and discontinuation.

Combined trauma-focused CBT and medication for delayed treatment (>3 months)

• Very low quality single-RCT (N=103) evidence suggests neither clinically important nor statistically significant effects of combined trauma-focused CBT and sertraline relative to waitlist, on self-rated PTSD symptomatology or quality of life, in adults with PTSD over 3 months after trauma. However, evidence from this same RCT suggests a moderate and statistically significant benefit of combined trauma-focused CBT and sertraline relative to waitlist on improving anxiety and depression symptoms, and functional impairment. Low quality evidence (N=139) from this study also suggests a clinically important benefit, that just misses statistical significance, on discontinuation with less drop-out associated with combined trauma-focused CBT and sertraline treatment.

Non-trauma-focused CBT for delayed treatment (>3 months)

• Low to very low quality evidence from 4-5 RCTs (N=228-339) suggests a moderate to large and statistically significant benefit of non-trauma-focused CBT (alone or in addition to TAU), relative to waitlist or TAU, on improving PTSD symptomatology (self-rated and clinician-rated) at endpoint in adults with PTSD over 3 months after trauma. Low to very low quality evidence from 1-5 RCTs (N=33-263) also suggests clinically important and statistically significant benefits on dissociative symptoms and sleeping difficulties. However, very low quality evidence from 1-3 RCTs (N=53-194) suggests effects on the rate of remission are not statistically significant at endpoint, and neither clinically important nor statistically significant at 3-month follow-up. Low to very low quality evidence from 2-4 RCTs (N=199-234) also suggests neither clinically important nor statistically significant effects of non-trauma-focused CBT on depression symptoms, alcohol use or drug use, at endpoint. Low quality evidence from 9 RCTs (N=684) suggests neither a clinically important nor statistically significant effect of non-trauma-focused CBT on discontinuation.
• Low quality single-RCT (N=60) evidence suggests potential benefits of non-trauma-focused CBT, relative to attention-placebo, on self-reported PTSD symptomatology in adults with PTSD over 3 months after trauma. However, when data is considered together with a much larger RCT (N=353) effects are non-significant.
• Low to very low quality evidence from 1-2 RCTs (N=24-121) suggests neither clinically important nor statistically significant differences between trauma-focused CBT (alone or in addition to TAU) and non-trauma-focused CBT (alone or in addition to TAU) on self-rated PTSD symptomatology at 1-month, 3-month or 6-month follow-ups (no endpoint data available) or clinician-rated PTSD symptomatology at endpoint or 1-3 month follow-up, although there is low quality single-RCT (N=22) evidence for a large and statistically significant benefit of trauma-focused, relative to non-trauma-focused, CBT on clinician-rated PTSD symptomatology at 6-month follow-up. Low to very low quality evidence from 1-2 RCTs (24-121) suggests no statistically significant difference between trauma-focused and non-trauma-focused CBT on remission (at endpoint, or 1-3 month, 6-month or 1-year follow-ups), response, anxiety symptoms, depression symptoms (at endpoint, or 1-, 3-, or 6- month follow-ups), or sleeping difficulties (at 1-, 3-, or 6- month follow-ups). Low quality single-RCT (N=95) evidence suggests a moderate and statistically significant benefit of trauma-focused CBT, relative to non-trauma-focused CBT, on quality of life at 1-month follow-up (no endpoint data available), however, this effect is not maintained at 3- or 6- month follow-up. Low quality evidence from 3 RCTs (N=183) suggests higher drop-out associated with trauma-focused, relative to non-trauma-focused, CBT.
• Moderate quality single-RCT (N=66) evidence suggests a delayed and moderate benefit of non-trauma-focused CBT (in addition to TAU) relative to a psychoeducational group (in addition to TAU) on clinician-rated PTSD symptomatology at 1-year follow-up (non-significant effects at endpoint, 3-month and 6-month follow-up) in adults with PTSD over 3 months after trauma. Moderate quality evidence from this RCT (N=66-77) also suggests moderate to large benefits of non-trauma-focused CBT on depression symptoms at endpoint, and 3-month and 6-month follow-up, although these are not maintained at 1-year follow-up. However, low quality evidence from this same RCT (N=66-77) suggests neither clinically important nor statistically significant differences between non-trauma-focused CBT and a psychoeducational group on self-rated PTSD symptomatology at endpoint, or at 3-month, 6-month or 1-year follow-ups. Moderate quality evidence from this RCT (N=111) also suggests higher drop-out associated with non-trauma-focused CBT relative to a psychoeducational group.
• Very low quality single-RCT (N=25-31) evidence suggests large and statistically significant benefits of non-trauma-focused CBT relative to supportive counselling on improving clinician-rated PTSD symptomatology and the rate of response in adults with PTSD over 3 months after trauma. Evidence from the same RCT also suggests clinically important, but not statistically significant, benefits of non-trauma-focused CBT on remission, anxiety symptoms and depression symptoms. There was a non-significant difference between non-trauma-focused CBT and counselling for discontinuation.
• Very low quality single-RCT (N=101) evidence suggests non-significant differences between non-trauma-focused CBT and present-centred therapy for clinician-rated PTSD symptomatology, remission, depression symptoms and discontinuation, in adults with PTSD over 3 months after trauma.

Present-centred therapy for delayed treatment (>3 months)

• Very low quality evidence from 1-2 RCTs (N=45-143) suggests moderate to large and statistically significant benefits of present-centred therapy relative to waitlist on improving clinician-rated PTSD symptomatology, dissociative symptoms, anxiety symptoms and depression symptoms, in adults with PTSD over 3 months after trauma. Evidence from these same 2 RCTs also suggests a clinically important, but not statistically significant, benefit of present-centred therapy on remission. Very low quality evidence from 1 of these RCTs (N=45) suggests a neither clinically important nor statistically significant effect of present-centred therapy on anger or quality of life. Very low quality evidence from both RCTs suggests there may be higher drop-out associated with present-centred therapy, however, this effect is not statistically significant.
• Very low quality evidence from 2 RCTs (N=114-119) suggests a moderate and statistically significant benefit of present-centred therapy in addition to TAU relative to TAU-only on improving clinician-rated PTSD symptomatology at endpoint, and a large and statistically significant benefit on improving depression symptoms at endpoint and 3-month and 6-month follow-up, in adults with PTSD over 3 months after trauma. However, the effect on PTSD symptomatology was not maintained at 3-month or 6-month follow-up. Low to very low quality evidence from 1-2 of these RCTs (N=60-130) also found non-significant effects on response (at endpoint, and 3- and 6-month follow-up) and discontinuation.

Cognitive therapies for delayed treatment (>3 months)

• Low quality evidence from 1-2 RCTs (N=21-40) suggests large and statistically significant benefits of metacognitive therapy (alone or in addition to TAU) relative to waitlist or TAU on self-rated PTSD symptomatology, response, anxiety symptoms and depression symptoms, in adults with PTSD over 3 months after trauma. Low quality evidence from both RCTs (N=41) suggests higher drop-out may be associated with metacognitive therapy, however, this effect is not statistically significant.
• Low quality single-RCT (N=20) evidence suggests a large and statistically significant benefit of metacognitive therapy (in addition to TAU) relative to trauma-focused CBT (in addition to TAU) on improving self-rated PTSD symptomatology at endpoint, in adults with PTSD over 3 months after trauma. However, this effect is not maintained at 3-month follow-up. In addition, low to very low quality evidence from this same RCT suggests non-significant differences between metacognitive therapy and trauma-focused CBT on remission, response, anxiety symptoms, depression symptoms, and discontinuation.
• Low to very low quality evidence from 1-2 RCTs (N=30-104) suggests large and statistically significant benefits of reconsolidation of traumatic memories (RTM) intervention in addition to TAU relative to TAU-only on improving PTSD symptomatology (self-rated and clinician-rated), in adults with PTSD over 3 months after trauma. Moderate quality evidence from both RCTs (N=104) also suggests that the reconsolidation of traumatic memories (RTM) intervention may be associated with lower drop-out than TAU-alone, although this effect is not statistically significant.

Behavioural therapies for delayed treatment (>3 months)

• Very low quality evidence from 1-2 RCTs (N=59-90) suggests large and statistically significant benefits of single-session behavioural therapy relative to waitlist on improving PTSD symptomatology (self-rated and clinician-rated), the rate of response, functional impairment, and depression symptoms in adults with PTSD over 3 months after trauma. Discontinuation is only reported by 1 of these RCTs (N=31) and there was no drop-out in either arm.

Problem solving for delayed treatment (>3 months)

• Low quality single-RCT (N=309) evidence suggests non-significant differences between problem solving and supportive counselling on self-rated PTSD symptomatology (at endpoint and 3-month follow-up) and discontinuation, in adults with PTSD over 3 months after trauma.

Eye movement desensitisation and reprocessing (EMDR) for early treatment (1-3 months)

• Very low quality single-RCT (N=39) evidence suggests a large and statistically significant benefit of early treatment with EMDR (initiated 1-3 months after trauma), relative to supportive counselling, on improving clinician-rated PTSD symptomatology and this benefit is maintained up to 3-month follow-up (longest follow-up). No participants dropped out of this study.

Eye movement desensitisation and reprocessing (EMDR) for delayed treatment (>3 months)

• Low to very low quality single-RCT (N=55-58) evidence suggests no statistically significant effects of EMDR relative to pill placebo on clinician-rated PTSD symptomatology, remission, depression symptoms or discontinuation, in adults with PTSD over 3 months after trauma.
• Very low quality evidence from 10 RCTs (N=440) suggests a large and statistically significant benefit of EMDR (in addition to TAU or alone), relative to TAU or waitlist, on improving self-rated PTSD symptomatology at endpoint in adults with PTSD over 3 months after trauma. Very low quality evidence from 2 RCTs (N=145) suggests a trend for this benefit to be maintained at 1-month follow-up. Moderate to very low quality evidence from 1-2 RCTs (N=40-147) also suggests clinically important and statistically significant benefits of EMDR on clinician-rated PTSD symptomatology, remission at endpoint and 1-month follow-up, response, dissociative symptoms and functional impairment. In addition, very low quality evidence from 3-7 RCTs (N=113-326) suggests large and statistically significant benefits of EMDR on anxiety and depression symptoms at endpoint, and very low quality evidence from 2 RCTs (N=145) suggests effects on depression are maintained at 1-month follow-up. Very low quality evidence from 8 RCTs (N=419) suggests there may be higher drop-out associated with EMDR, however, this effect is not statistically significant.
• Low to very low quality evidence from 1-5 RCTs (N=30-230) suggests no statistically significant differences between EMDR and trauma-focused CBT on PTSD outcomes (self-rated and clinician-rated symptomatology, remission and response), although there is a trend in favour of EMDR for adults who had experienced single incident index trauma more than 3 months ago.
• Low quality single-RCT (N=57-60) evidence suggests moderate to large and statistically significant benefits of EMDR relative to supportive counselling on improving self-rated PTSD symptomatology, anxiety symptoms, and depression symptoms in adults with PTSD over 3 months after trauma. Evidence from this same RCT (N=67) suggests there may be higher drop-out associated with EMDR, however, this effect is not statistically significant.
• Very low quality single-RCT (N=61-74) evidence suggests non-significant differences between EMDR and non-trauma-focused CBT on PTSD symptomatology (clinician-rated and self-rated), anxiety symptoms and depression symptoms at endpoint and 3-month follow-up, and response and discontinuation at endpoint, in adults with PTSD over 3 months after trauma.
• Very low quality single-RCT (N=31-40) evidence suggests moderate to large and delayed benefits of EMDR relative to ‘other active psychological intervention’ on improving self-rated PTSD symptomatology at 3-month and 18-month follow-up (non-significant at endpoint), and depression symptoms at 3-month follow-up (non-significant at endpoint and 18-month follow-up), in adults with PTSD over 3 months after trauma. No participants discontinued this study in either arm.
• Low to very low quality evidence from 1-3 RCTs (N=30-88) suggests non-significant differences between EMDR (alone or in addition to TAU) and relaxation (alone or in addition to TAU) on PTSD symptomatology (self-rated and clinician-rated) and remission at endpoint, 3-month and 6-month follow-up, dissociative symptoms at endpoint and 3-month follow-up (longest follow-up), and anxiety symptoms, quality of life and discontinuation at endpoint, in adults with PTSD over 3 months after trauma. Low quality evidence from 2 of these RCTs (N=52) suggests a moderate and statistically significant benefit of EMDR relative to relaxation on improving depression symptoms at endpoint, however, this effect is not maintained at 3-6 month follow-up.
• Low to very low quality single-RCT (N=46) evidence suggests non-significant differences between EMDR and emotional freedom technique (EFT) on PTSD symptomatology (self-rated and clinician-rated), response (based on self-rated and clinician-rated measures), anxiety symptoms, depression symptoms and quality of life, at endpoint and 3-month follow-up, and discontinuation, in adults with PTSD over 3 months after trauma.
• Low quality single-RCT evidence (N=50) suggests a delayed, large and statistically significant benefit of EMDR relative to fluoxetine on improving clinician-rated PTSD symptomatology, remission and depression symptoms at 6-month follow-up (non-significant at endpoint) in adults with PTSD over 3 months after trauma. Low quality evidence from this same RCT (N=59) suggests EMDR may be associated with higher drop-out, however, this effect is not statistically significant.

Hypnotherapy for delayed treatment (>3 months)

• Low quality single-RCT (N=52) evidence suggests a large and statistically significant benefit of hypnotherapy in addition to TAU relative to TAU-only on improving self-rated PTSD symptomatology in adults with PTSD over 3 months after trauma. Evidence is not available for any other outcomes.
• Very low quality single-RCT (N=60) evidence suggests neither clinically important nor statistically significant differences between hypnotherapy (in addition to TAU) and trauma-focused CBT (in addition to TAU) on self-rated PTSD symptomatology at endpoint and 3-month follow-up, in adults with PTSD over 3 months after trauma.
• Very low quality single-RCT (N=54-108) evidence suggests non-significant differences between hypnotherapy followed by trauma-focused CBT and symptom monitoring followed by trauma-focused CBT on clinician-rated PTSD symptomatology and sleeping difficulties at endpoint and 3-month follow-up, and on discontinuation, in adults with PTSD over 3 months after trauma. Very low quality evidence from this same RCT (N=54) suggests a moderate and statistically significant benefit of hypnotherapy followed by trauma-focused CBT on depression symptoms, however, this effect is not maintained at 3-month follow-up.
• Low quality single-RCT (N=32) evidence suggests large and statistically significant benefits of hypnotherapy (in addition to TAU) relative to zolpidem (in addition to TAU) on improving self-rated PTSD symptomatology and depression symptoms at endpoint and 1-month follow-up, in adults with PTSD over 3 months after trauma. Very low quality evidence from this same RCT (N=33) suggests higher drop-out may be associated with zolpidem, however, absolute numbers are small and this effect is not statistically significant.

Interpersonal psychotherapy (IPT) for delayed treatment (>3 months)

• Very low quality single-RCT (N=48) evidence suggests large and statistically significant benefits of IPT relative to waitlist on improving clinician-rated PTSD symptomatology, remission and depression symptoms at endpoint, in adults with PTSD over 3 months after trauma. However, these effects are not maintained at 4-month follow-up. Low quality evidence from this same RCT suggests non-significant effects on discontinuation.
• Low quality single-RCT (N=36-72) evidence suggests moderate to large and statistically significant benefits of IPT relative to relaxation on improving self-rated PTSD symptomatology, the rate of response, functional impairment and relationship difficulties, in adults with PTSD over 3 months after trauma. However, low to very low quality evidence from this same RCT (N=38-72) suggests non-significant effects on clinician-rated PTSD symptomatology, remission, depression symptoms, quality of life and discontinuation.

Psychodynamic therapies for delayed treatment (>3 months)

• Low quality evidence from single-study analyses (N=52-84) suggests large and statistically significant benefits of psychodynamic therapy (alone or in addition to TAU) relative to waitlist (alone or in addition to TAU) on improving self-rated PTSD symptomatology, remission, anxiety and depression symptoms, in adults with PTSD over 3 months after trauma. Very low quality evidence from one of these RCTs (N=86) suggests non-significant effects on discontinuation.
• Very low quality single-RCT (N=60) evidence suggests non-significant differences between psychodynamic therapy (in addition to TAU) and trauma-focused CBT (in addition to TAU) on self-rated PTSD symptomatology in adults with PTSD over 3 months after trauma.

Counselling for delayed treatment (>3 months)

• Low quality evidence from 1-4 RCTs (N=60-249) suggests large and statistically significant benefits of counselling (alone or in addition to TAU) relative to TAU or waitlist, on improving PTSD symptomatology (self-rated and clinician-rated) and functional impairment at endpoint, in adults with PTSD over 3 months after trauma. Very low quality evidence from 2-RCT-analyses (N=190-234) suggests the effect on self-rated PTSD symptomatology is maintained at 8-12 month follow-up and clinically important but not statistically significant at 1-4 month follow-up. Very low quality single-RCT (N=24) evidence suggests the effect on clinician-rated PTSD symptomatology is not maintained at 1-year follow-up. Low quality evidence from 2 RCTs (N=102) suggests a clinically important and statistically significant benefit of counselling on remission at endpoint and very low quality evidence from 2 other RCTs (N=192) shows a trend for the same effect at 8-12 month follow-up. Low quality evidence from 2 RCTs (N=111) suggests moderate and statistically significant benefits of counselling on anxiety and depression symptoms at endpoint, and low quality evidence from another single RCT (N=209) suggests a trend for benefits to be observed at 1-month follow-up. However, low quality single-RCT (N=24-25) evidence suggests counselling may be associated with lower quality of life scores than treatment as usual at 4-month and 1-year follow-up for adults who had experienced multiple incident index trauma (non-significant effects at endpoint). Very low quality evidence from 1-6 RCTs (N=51-646) suggests non-significant effects on global functioning and discontinuation.

Combined somatic and cognitive therapies for delayed treatment (>3 months)

• Low to very low quality evidence from 1-4 RCTs (49-484) suggests large and statistically significant benefits of combined somatic and cognitive therapies (alone or in addition to TAU), relative to waitlist (alone or in addition to TAU), on improving self-rated PTSD symptomatology, the rate of remission, anxiety and depression symptoms, and sleeping difficulties in adults with PTSD over 3 months after trauma. However, heterogeneity is very high for self-rated PTSD symptomatology. Sub-analysis by specific intervention and trauma type suggests some differential effects of combined somatic and cognitive therapies, with the largest effect observed for emotional freedom technique (EFT) with military combat veterans. Very low quality evidence from 4 RCTs (N=544) suggests neither a clinically important nor statistically significant effect on discontinuation.

Somatic experiencing for delayed treatment (>3 months)

• Low to very low quality single-RCT (N=60) evidence suggests large and statistically significant benefits of somatic experiencing in addition to TAU relative to TAU-only on improving PTSD symptomatology (self-rated and clinician-rated) and depression symptoms, in adults with PTSD over 3 months after trauma. Very low quality evidence from this same RCT suggests higher drop-out may be associated with somatic experiencing, however, this effect is not statistically significant.

Resilience-oriented treatment for delayed treatment (>3 months)

• Low quality single-RCT (N=39) evidence suggests large and statistically significant benefits of resilience-oriented treatment relative to waitlist on improving self-rated PTSD symptomatology, anxiety and depression symptoms, in adults with PTSD over 3 months after trauma. Very low quality evidence from the same RCT suggests a non-significant effect on discontinuation.

Attention bias modification for delayed treatment (>3 months)

• Very low quality evidence from 2-3 RCTs (N=118-170) suggests clinically important but not statistically significant effects in favour of attention-placebo relative to attention bias modification on PTSD symptomatology (self-rated and clinician-rated) at endpoint, in adults with PTSD over 3 months after trauma. Moderate to very low quality evidence from 1-3 RCTs (N=72-170) suggests non-significant effects on anxiety and depression symptoms and discontinuation.

Couple interventions for delayed treatment (>3 months)

• Very low quality single-RCT (N=40) evidence suggests a large and statistically significant benefit of cognitive-behavioural conjoint therapy relative to waitlist on the rate of remission for PTSD symptoms, in adults with PTSD over 3 months after trauma. However, evidence from the same RCT suggests non-significant effects in the rate of response for PTSD symptoms, response for relationship difficulties, and remission for relationship difficulties. There is some evidence for higher drop-out associated with cognitive-behavioural conjoint therapy, however, this effect is not statistically significant.
• Very low quality single-RCT (N=41-57) evidence suggests large and statistically significant benefits of cognitive-behavioural conjoint therapy relative to psychoeducation sessions on PTSD symptomatology (self-rated and clinician-rated), anxiety symptoms and relationship difficulties at endpoint and 3-month follow-up, the rate of remission at endpoint (no follow-up available), and depression symptoms at 3-month follow-up (clinically important but not statistically significant at endpoint), in adults with PTSD over 3 months after trauma. Evidence from this same RCT suggests non-significant differences between cognitive-behavioural conjoint therapy and psychoeducation on discontinuation.

Parent training/family interventions for delayed treatment (>3 months)

• Low quality single-RCT (N=142) evidence suggests non-significant effects of family therapy relative to waitlist on self-rated PTSD symptomatology and anxiety symptoms at 4-month follow-up in adults with PTSD over 3 months after trauma. No endpoint data or other outcomes are available.
• Very low quality single-RCT (N=28-65) evidence suggests a moderate and statistically significant benefit of child-parent psychotherapy (using play) versus case management and individual treatment (for parent-only) on improving clinician-rated PTSD symptomatology, and a clinically important but not statistically significant benefit on the rate of remission, in adults with PTSD over 3 months after trauma. Evidence from this same RCT suggests a non-significant effect on discontinuation.

Self-help with support for delayed treatment (>3 months)

• Low to very low quality evidence from 1-6 RCTs (N=42-545) suggests large and statistically significant benefits of self-help with support (alone or in combination with TAU) relative to waitlist or TAU on improving self-rated PTSD symptomatology and anxiety and depression symptoms (at endpoint, 1-3 month and 1-year follow-up), clinician-rated PTSD symptomatology and functional impairment (at endpoint and 1-month follow-up [longest follow-up]), response, quality of life and sleeping difficulties (at endpoint), in adults with PTSD over 3 months after trauma. Very low quality evidence from 2 RCTs (N=211) suggests a clinically important but not statistically significant benefit of self-help with support on the rate of remission. Low to very low quality evidence from 1-7 RCTs (N=34-673) suggests non-significant effects of self-help with support on alcohol use disorder symptoms, substance use disorder symptoms (at endpoint and 3-month follow-up) and discontinuation. Sub-analysis of self-rated PTSD symptomatology by baseline severity showed non-significant subgroup differences.
• Very low quality single-RCT (N=85) evidence suggests neither clinically important nor statistically significant differences between self-help with support and trauma-focused CBT on self-rated PTSD symptomatology, dissociative symptoms, anxiety symptoms, or depression symptoms, at 2-month or 1-year follow-up (no endpoint data available) in adults with PTSD over 3 months after trauma.
• Very low quality single-RCT (N=43) evidence suggests a large and statistically significant benefit of a psychoeducational website without support relative to computerised trauma-focused CBT with support on anxiety symptoms at endpoint, in adults with PTSD over 3 months after trauma. However, this effect was not maintained at 3-month follow-up. Low to very low quality evidence from this same RCT (N=41-87) also suggests non-significant effects on clinician-rated PTSD symptomatology, response and depression symptoms at endpoint and 3-month follow-up, and discontinuation.

Self-help without support for delayed treatment (>3 months)

• Low to very low quality evidence from 2-5 RCTs (N=103-288) suggests moderate and statistically significant benefits of self-help without support relative to waitlist on improving self-rated PTSD symptomatology, the rate of remission (at endpoint and 3-6 month follow-up), response at endpoint (clinically important but not statistically significant at 3-6 month follow-up), and functional impairment and depression symptoms at endpoint (non-significant at 6-month follow-up), in adults with PTSD over 3 months after trauma. Very low quality evidence from 3 RCTs (N=121) suggests a clinically important but not statistically significant benefit of self-help without support on anxiety symptoms at endpoint (non-significant at 6-month follow-up). Low quality evidence from 7 RCTs (N=434) suggests higher drop-out may be associated with self-help without support, however, this effect is not statistically significant. Sub-analysis of self-rated PTSD symptomatology by baseline severity showed non-significant subgroup differences.
• Very low quality evidence from 5 RCTs (N=358-377) suggests moderate and statistically significant benefits of self-help without support relative to attention-placebo on improving self-rated PTSD symptomatology at endpoint, and a clinically important but not statistically significant benefit on improving depression symptoms at endpoint, in adults with PTSD over 3 months after trauma. These effects were not maintained at 1-month follow-up. Moderate to very low quality evidence from 1-4 RCTs (N=36-283) also suggests non-significant effects on clinician-rated PTSD symptomatology, remission, anxiety symptoms and discontinuation.

Trauma-focused CBT

• Evidence from 1 Australian economic evaluation conducted alongside a RCT (N = 354; missing data on approximately 27% of participants were imputed by multiple imputation) suggests that, compared with psychoeducation, trauma-focused CBT is likely to be cost-effective for the treatment of PTSD in adults with PTSD and at high distress but unlikely to be cost-effective for the treatment of PTSD in adults with PTSD and at low distress. This evidence is partially applicable to the UK context and is characterised by minor methodological limitations.
• Evidence from 1 Australian model-based economic study suggests that trauma-focused CBT is likely to be cost-effective for the treatment of PTSD in adults compared with treatment as usual. This evidence is partially applicable to the UK context and is characterised by potentially serious limitations.
• Evidence from 1 US before-after study suggests that trauma-focused CBT (exposure therapy /prolonged exposure) is likely to be more cost-effective compared with no treatment. This evidence is partially applicable to the UK and is characterised by potentially serious limitations.
• Evidence from 1 US economic evaluation conducted alongside a RCT with a preference arm (N=200; preference arm n=97, completers n=69; RCT n=103; completers n=58) suggests that trauma-focused CBT (exposure therapy / prolonged exposure) is likely to be more cost-effective compared with sertraline. This evidence is partially applicable to the UK and is characterised by potentially serious limitations.

Non-trauma-focused CBT

• Evidence from 1 US economic evaluation conducted alongside a RCT (N=101; at 1-year follow up: n=66) suggests that, compared with psychoeducation, non-trauma-focused CBT results in lower costs, similar effects on PTSD symptoms and modestly lower effects on depressive symptoms and functioning in adults with chronic combat-related PTSD and depressive disorder. This evidence is partially applicable to the UK and is characterised by potentially serious limitations.

Psychological, pharmacological and combined interventions

• Evidence from the guideline economic analysis suggests that brief TF-CBT individual (<8 sessions), psychoeducation, EMDR, combined somatic and cognitive therapies and self-help with support are the 5 most cost-effective interventions for the treatment of PTSD in adults. TF-CBT individual >12 sessions, counselling, combined TF-CBT + SSRI, group TF-CBT and present-centred therapy appear to be less cost-effective relative to other active interventions. Counselling and TF-CBT individual > 12 sessions were also found to be less cost-effective than no treatment in the base-case analysis. In-between, there is another group of interventions (SSRIs, TF-CBT individual 8-12 sessions, self-help without support, non-TF-CBT and IPT) that occupied middle cost effectiveness rankings (i.e. places 6-10) in the base-case analysis. The result for psychoeducation, which was found to be among the most cost-effective interventions, should be interpreted with great caution due to the limitations in the evidence. The economic analysis is directly applicable to the NICE decision-making context and is overall characterised by minor limitations, mainly relating to the NMAs that informed the analysis.

Interpreting the evidence

Included studies

Two studies of animal-assisted therapy for the treatment of PTSD in adults were identified for full-text review. Neither of these studies were included.

Excluded studies

Two studies were reviewed at full text and excluded from this review because the population was outside the scope (trial of people without PTSD), or a cross-over study where the first phase data were not available.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

Two studies of art therapy for the treatment of PTSD in adults were identified for full-text review. Neither of these studies were included.

Excluded studies

Two studies were reviewed at full text and excluded from this review because they were systematic reviews with no new useable data and any meta-analysis results were not appropriate to extract.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

Twenty-five studies of meditation or mindfulness based stress reduction (MBSR) for the treatment of PTSD in adults were identified for full-text review. Of these 25 studies, 9 RCTs (N=680) were included. There were 3 comparisons for meditation/MBSR (one study was in two comparisons).

There were no studies for early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms.

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, 8 RCTs (N=564) compared meditation/MBSR (alone or in addition to TAU) with TAU, attention-placebo or waitlist (Bormann et al. 2008; Bormann et al. 2012/2013 [one study reported across two papers]; Bränström et al. 2010/2012 [one study reported across two papers]; Kearney et al. 2013; Kearney et al. 2016; Levine et al. 2005; Possemato et al. 2016; Wahbeh et al. 2016/Colgan et al. 2016 [one study reported across two papers]). 1 RCT (N=114) compared meditation (in addition to TAU) with relaxation (in addition to TAU) (Wahbeh et al. 2016/Colgan et al. 2016 [one study reported across two papers]), and 1 RCT (N=116) compared MBSR (in addition to TAU) with present-centred therapy (in addition to TAU) (Polusny et al. 2015).

Sub-analyses were possible for the delayed treatment meditation/MBSR (alone or in addition to TAU) versus TAU/attention-placebo/waitlist comparison, comparing effects by multiplicity of trauma, specific comparison, diagnostic status at baseline, and trauma type.

Excluded studies

Sixteen studies were reviewed at full text and excluded from this review. The most common reasons for exclusion were systematic review with no new useable data and any meta-analysis results not appropriate to extract, or efficacy or safety data could not be extracted.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

One study of supported employment for the treatment of PTSD in adults was identified for full-text review. This RCT (N=85) was included in a single comparison for supported employment.

There were no studies for early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms.

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, 1 RCT (N=85) compared individual placement and support (IPS) supported employment with standard VA vocational rehabilitation programme (TAU).

Sub-analyses were not possible for supported employment.

Excluded studies

There were no studies that met criteria for full-text review that were excluded.

Included studies

Two studies of practical support for the treatment of PTSD in adults were identified for full-text review. Of these 2 studies, 1 RCT (N=41) was included in a single comparison for practical support.

There were no studies for early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms.

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, 1 RCT (N=41) compared practical support with treatment as usual (Weinstein et al. 2016).

Sub-analyses were not possible for practical support.

Excluded studies

One study was reviewed at full text and excluded from this review because the outcomes were not of interest.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

Ten studies of psychoeducation for the treatment of PTSD in adults were identified for full-text review. Of these 10 studies, 3 RCTs (N=689) were included. There were 2 comparisons for psychoeducation.

For early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms, 1 RCT (N=386) compared psychoeducation in addition to treatment as usual with treatment as usual-only (Jensen et al. 2016).

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, 2 RCTs (N=303) compared psychoeducation (alone or in addition to TAU) with waitlist or TAU (Ghafoori et al. 2016; Kaslow et al. 2010).

Sub-analyses were not possible for psychoeducational interventions.

Excluded studies

Seven studies were reviewed at full text and excluded from this review. The most common reasons for exclusion was that the intervention was not targeted at PTSD symptoms.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

Six studies of relaxation for the treatment of PTSD in adults were identified for full-text review. None of these studies were included.

Comparisons with trauma-focused CBT are presented in the Trauma-focused CBT section above.

Excluded studies

Six studies were reviewed at full text and excluded from this review because the comparison were outside the protocol (within-class comparison) or outcomes were not of interest, there was non-randomised group assignment, a small sample size (N<10 per arm), or the population was outside the scope (trial of soldiers on active service).

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

Two studies of peer support for the treatment of PTSD in adults were identified for full-text review. Neither of these studies were included.

Excluded studies

Two studies were reviewed at full text and excluded from this review because the intervention was not being targeted at PTSD symptoms, or it was a systematic review with no new useable data and any meta-analysis results were not appropriate to extract.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

One study of mentoring for the treatment of PTSD in adults was identified for full-text review. This study was not included.

Excluded studies

One study was reviewed at full text and excluded from this review because efficacy or safety data could not be extracted.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

Two studies of nature-assisted therapies for the treatment of PTSD in adults were identified for full-text review. Neither of these studies were included.

Excluded studies

Two studies were reviewed at full text and excluded from this review because of small sample size (N<10 per arm) or non-validated outcome measures.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

One study of spiritual intervention for the treatment of PTSD in adults was identified for full-text review. This study was not included.

Included studies

No studies assessing the cost effectiveness of psychosocial interventions for the treatment of PTSD in adults were identified. The search strategy for economic studies is provided in Appendix B.

Excluded studies

No economic studies on psychosocial interventions for the treatment of PTSD in adults were reviewed at full text and excluded.

Meditation/Mindfulness-based stress reduction (MBSR) for delayed treatment (>3 months)

• Low quality evidence from 4 RCTs (N=284) suggests a small but statistically significant benefit of meditation/MBSR (alone or in addition to TAU) relative to TAU, attention-placebo or waitlist on improving clinician-rated PTSD symptomatology at endpoint, and very low quality single-RCT (N=45) evidence suggests this benefit is maintained at 6-month follow-up, in adults with PTSD over 3 months after trauma. Moderate to very low quality evidence from 1-6 RCTs (N=47-450) suggests moderate and statistically significant benefits of meditation/MBSR on depression symptoms at endpoint and 1-6 month follow-up, and quality of life at endpoint and 4-month follow-up. However, low to very low quality evidence from 1-6 RCTs (N=29-387) suggests non-significant effects on self-rated PTSD symptomatology at endpoint or 1-4 month follow-up, the rate of remission, sleeping difficulties and emotional and behavioural problems at endpoint, the rate of response at endpoint or 4-month follow-up, and anxiety symptoms at endpoint or 3-month follow-up. Low quality evidence from 6 RCTs (N=424) suggests there may be higher drop-out associated with meditation/MBSR, however, this effect is not statistically significant.
• Low to very low quality single-RCT (N=77) evidence suggests moderate and statistically significant benefits of meditation (in addition to TAU) relative to relaxation (in addition to TAU) at study/treatment endpoint on improving self-rated PTSD symptomatology and depression symptoms, and a clinically important but not statistically significant benefit on the rate of response, in adults with PTSD over 3 months after trauma. However, low quality evidence from the same RCT suggests a non-significant effect on sleeping difficulties. No discontinuation evidence is available.
• Very low quality single-RCT (N=116) evidence suggests a moderate to large and statistically significant benefit of MBSR (in addition to TAU) relative to present-centred therapy (in addition to TAU) on improving self-rated PTSD symptomatology at endpoint and 2-month follow-up, and a delayed benefit on clinician-rated PTSD symptomatology at 2-month follow-up (non-significant at endpoint), in adults with PTSD over 3 months after trauma. However, evidence from the same RCT suggests non-significant differences between MBSR and present-centred therapy on the rate of remission and response (based on self-rated and clinician-rated measures), depression symptoms, and quality of life at endpoint and 2-month follow-up. Evidence from this RCT suggests higher drop-out may be associated with MBSR, however, this effect is not statistically significant.

Supported employment for delayed treatment (>3 months)

• Low quality single-RCT (N=85) evidence suggests a large and statistically significant benefit of individual placement and support (IPS) supported employment relative to standard VA vocational rehabilitation programme (TAU) on competitive employment (as measured by number of people who gained competitive employment and weeks competitively employed), in adults with PTSD over 3 months after trauma. However, low to very low quality evidence from this same RCT (N=57-85) suggests non-significant effects of IPS on PTSD symptomatology (self-rated or clinician-rated), the rate of response, depression symptoms and discontinuation.

Practical support for delayed treatment (>3 months)

• Low quality single-RCT (N=41) evidence suggests large and statistically significant benefits of practical support relative to TAU on improving self-rated PTSD symptomatology and depression symptoms, in adults with PTSD over 3 months after trauma. No evidence for other outcomes is available.

Psychoeducational interventions for early treatment (1-3 months)

• Low to very low quality single-RCT (N=225-386) evidence suggests non-significant effects of psychoeducation in addition to TAU relative to TAU-only for the early treatment of PTSD (initiated within 1-3 months of trauma) on self-rated PTSD symptomatology, anxiety symptoms, depression symptoms, and quality of life at 2-month follow-up (endpoint data not available), or discontinuation.

Psychoeducational interventions for delayed treatment (>3 months)

• Low to very low quality evidence from single-RCT analyses (N=59-89) suggests a moderate to large and statistically significant benefit of psychoeducation (alone or in addition to TAU) relative to waitlist or TAU on depression symptoms (at endpoint, and 1-, 6- and 12-month follow-up), in adults with PTSD over 3 months after trauma. However, evidence from single-RCT analyses suggests non-significant effects of psychoeducation on self-rated PTSD symptomatology (at endpoint, and 1-, 6- and 12-month follow-up), anxiety symptoms at 1-month follow-up, and suicide (at endpoint, 6-month and 1-year follow-up). Low quality evidence from 2 RCTs (N=303) suggests a moderate and statistically significant benefit on discontinuation, with lower drop-out associated with psychoeducation.

Interpreting the evidence

Included studies

Ten studies of acupuncture for the treatment of PTSD in adults were identified for full-text review. Of these 10 studies, 2 RCTs (N=222) were included. There were 2 comparisons for acupuncture.

There were no studies for early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms.

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, 1 RCT (N=84) compared acupuncture with waitlist (Hollifield et al. 2007), and 1 RCT (N=138) compared acupuncture with paroxetine (Wang et al. 2012).

Comparisons with trauma-focused CBT are presented in the Trauma-focused CBT section above.

Sub-analyses were not possible for acupuncture.

Excluded studies

Eight studies were reviewed at full text and excluded from this review. The most common reasons for exclusion were small sample size (N<10 per arm) or systematic review with no new useable data and any meta-analysis results not appropriate to extract.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

Eleven studies of exercise for the treatment of PTSD in adults were identified for full-text review. Of these 11 studies, 2 RCTs (N=128) were included in a single comparison for exercise.

There were no studies for early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms.

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, 2 RCTs (N=128) compared exercise in addition to treatment as usual with treatment as usual-only (Goldstein et al. 2018; Rosenbaum et al. 2011/Rosenbaum et al. 2015 [one study reported across two papers]).

Sub-analyses were not possible for exercise.

Excluded studies

Nine studies were reviewed at full text and excluded from this review. The most common reason for exclusion was that the paper was a systematic review with no new useable data and any meta-analysis results not appropriate to extract.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

Seven studies of repetitive transcranial magnetic stimulation (rTMS) for the treatment of PTSD in adults were identified for full-text review. Of these 7 studies, 1 RCT (N=20) was included in a single comparison for rTMS.

There were no studies for early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms.

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, 1 RCT (N=20) compared rTMS with sham stimulation (Watts et al. 2012).

Sub-analyses were not possible for rTMS.

Excluded studies

Six studies were reviewed at full text and excluded from this review. The most common reasons for exclusion were small sample size (N<10 per arm) or systematic review with no new useable data and any meta-analysis results not appropriate to extract.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

Fifteen studies of yoga for the treatment of PTSD in adults were identified for full-text review. Of these 15 studies, 3 RCTs (N=194) were included in a single comparison for yoga.

There were no studies for early treatment (intervention initiated 1-3 months posttrauma) of PTSD symptoms.

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, 3 RCTs (N=194) compared yoga (alone or in addition to TAU) with TAU or waitlist or attention-placebo (Jindani et al. 2015; Mitchell et al. 2014/Dick et al. 2014/Reddy et al. 2014 [one study reported across three papers]; van der Kolk et al. 2014).

Sub-analyses were not possible for yoga.

Excluded studies

Twelve studies were reviewed at full text and excluded from this review. The most common reasons for exclusion were that the paper was a systematic review with no new useable data and any meta-analysis results not appropriate to extract, or efficacy or safety data could not be extracted.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

Five studies of biofeedback or neurofeedback for the treatment of PTSD in adults were identified for full-text review. Of these 5 studies, 3 RCTs (N=102) were included in a single comparison for bio-/neuro-feedback.

There were no studies for early treatment (intervention initiated 1-3 months post-trauma) of PTSD symptoms.

For delayed treatment (intervention initiated more than 3 months post-trauma) of PTSD symptoms, 3 RCTs (N=102) compared biofeedback or neurofeedback (alone or in addition to TAU) with TAU or no treatment (Noohi et al. 2017; Tan et al. 2011; van der Kolk et al. 2016).

Sub-analyses were not possible for bio-/neuro-feedback.

Excluded studies

Two studies were reviewed at full text and excluded from this review due to outcomes not being of interest (no validated PTSD scale) or the paper was a systematic review with no new useable data and any meta-analysis results not appropriate to extract.

Studies not included in this review with reasons for their exclusions are provided in Appendix K.

Included studies

No studies assessing the cost effectiveness of other non-pharmacological interventions for the treatment of PTSD in adults were identified. The search strategy for economic studies is provided in Appendix B.

Excluded studies

No economic studies on other non-pharmacological interventions for the treatment of PTSD in adults were reviewed at full text and excluded.

Acupuncture for delayed treatment (>3 months)

• Very low quality single-RCT (N=48) evidence suggests large and statistically significant benefits of acupuncture relative to waitlist on improving self-rated PTSD symptomatology, the rate of remission, anxiety and depression symptoms, and functional impairment in adults with PTSD over 3 months after trauma. Evidence from this same RCT (N=56) suggests there may be higher drop-out associated with acupuncture, however, this effect is not statistically significant.
• Low to very low quality single-RCT (N=127) evidence suggests non-significant differences between acupuncture and paroxetine on clinician-rated PTSD symptomatology, anxiety and depression symptoms at endpoint, 3-month and 6-month follow-up, and discontinuation in adults with PTSD over 3 months after trauma.

Exercise for delayed treatment (>3 months)

• Low quality single-RCT (N=58) evidence suggests moderate to large and statistically significant benefits of exercise in addition to TAU relative to TAU-only on improving clinician-rated PTSD symptomatology, anxiety symptoms and sleeping difficulties, in adults with PTSD over 3 months after trauma. However, evidence from the same study suggests non-significant effects of exercise on self-rated PTSD symptomatology, depression symptoms or discontinuation.

Repetitive transcranial magnetic stimulation (rTMS) for delayed treatment (>3 months)

• Low quality single-RCT (N=20) evidence suggests large and statistically significant benefits of repetitive transcranial magnetic stimulation (rTMS) relative to sham stimulation on improving PTSD symptomatology (self-rated and clinician-rated) and depression symptoms, in adults with PTSD over 3 months after trauma. No evidence is available for discontinuation.

Yoga for delayed treatment (>3 months)

• Low to very low quality single-RCT analyses (N=50-60) suggests moderate and statistically significant benefits of yoga (in addition to TAU) relative to TAU or attention-placebo (in addition to TAU) on improving clinician-rated PTSD symptomatology, the rate of remission and sleeping difficulties in adults with PTSD over 3 months after trauma. Low to very low quality evidence from 1-3 RCTs (N=25-148) suggests clinically important, but not statistically significant benefits of yoga (alone or in addition to TAU) relative to TAU, attention-placebo or waitlist on self-rated PTSD symptomatology, dissociative symptoms at endpoint, and symptoms of alcohol use disorder at endpoint and 1-month follow-up. However, single-RCT (N=38) evidence suggests the effect on self-rated PTSD symptomatology is not maintained at 1-month follow-up, and there is no follow-up data available for dissociative symptoms. Low to very low quality evidence from 1-3 RCTs (N=38-148) suggests non-significant effects of yoga on anxiety and depression symptoms and symptoms of drug use disorder (at endpoint and 1-month follow-up). Very low quality evidence from 2 RCTs (N=118) suggests there may be higher drop-out associated with yoga, however, this effect is not statistically significant.

Bio-/neuro-feedback for delayed treatment (>3 months)

• Low to very low quality evidence from 1-3 RCTs (N=30-94) suggests large and statistically significant benefits of bio-/neuro-feedback (alone or in addition to TAU) relative to TAU or no treatment on improving self-rated PTSD symptomatology at endpoint and 4-6 week follow-up, clinician-rated PTSD symptomatology at 1-month follow-up (clinically important but not statistically significant at endpoint), remission at 1-month follow-up (clinically important but not statistically significant at endpoint) and depression symptoms at endpoint and 6-week follow-up, in adults with PTSD over 3 months after trauma. Very low quality evidence from 2 RCTs (N=72) suggests there may be higher drop-out associated with bio-/neuro-feedback, however, this effect is not statistically significant.

Interpreting the evidence

Cost effectiveness and resource use

No evidence on the cost effectiveness of other non-pharmacological interventions for the treatment of PTSD in adults was identified and no economic modelling was undertaken. The committee did not make any recommendations on other non-pharmacological interventions for the treatment of PTSD in adults due to uncertain or limited evidence of their benefits. As none of these interventions are in widespread use in routine clinical practice, the committee expressed the view that there would be no impact on resources.

Health economic evidence

Note: evidence resulting from the health economic search update was screened to reflect the final dates of the searches that were undertaken for the clinical reviews (see review protocols).

Database: Epub Ahead of Print, In-Process & Other Non-Indexed Citations, Ovid MEDLINE(R) Daily and Ovid MEDLINE(R), Embase, PsycINFO

Date of last search: 1 March 2018

Database: HTA, NHS EED

Date of last search: 1 March 2018

Trauma-focused CBT

Figure 2. Trauma-focused CBT versus waitlist or no treatment for early treatment (1-3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (IES change score); single-incident index trauma

Figure 3. Trauma-focused CBT versus waitlist or no treatment for early treatment (1-3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score); single-incident index trauma

Figure 4. Trauma-focused CBT versus waitlist or no treatment for early treatment (1-3 months) of clinically important symptoms/PTSD: Remission (number of people no longer meeting diagnostic criteria for PTSD); Single incident index trauma

Figure 5. Trauma-focused CBT versus waitlist or no treatment for early treatment (1-3 months) of clinically important symptoms/PTSD: Response self-rated (number of participants showing at least 50% improvement from baseline on IES); single-incident index trauma

Figure 6. Trauma-focused CBT versus waitlist or no treatment for early treatment (1-3 months) of clinically important symptoms/PTSD: Anxiety symptoms (HADS-A change score); single-incident index trauma

Figure 7. Trauma-focused CBT versus waitlist or no treatment for early treatment (1-3 months) of clinically important symptoms/PTSD: Depression symptoms (HADS-D change score); single-incident index trauma

Figure 8. Trauma-focused CBT versus waitlist or no treatment for early treatment (1-3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 9. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (PCL/SPTSS/HTQ/MPSS/PDS/PSS-SR/IES-R change score)

Figure 10. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 6-7 week follow-up (IES/HTQ change score)

Figure 11. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 3-month follow-up (HTQ change score)

Figure 12. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 8-month follow-up (PDS change score)

Figure 13. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 1-year follow-up (IES change score)

Figure 14. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at end-point (CAPS/HTQ/SI-PTSD/PSS-I change score)

Figure 15. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at 3-5 month follow-up (CAPS/PSS-I/HTQ change score)

Figure 16. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at endpoint (number of people no longer meeting diagnostic criteria for PTSD or no longer above clinical threshold on scale)

Figure 17. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at 3-6 month follow-up (number of people no longer meeting diagnostic criteria for PTSD or no longer above clinical threshold on scale)

Figure 18. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at 8-month follow-up (number of people no longer meeting diagnostic criteria for PTSD)

Figure 19. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response self-rated at endpoint (number of people showing clinically significant improvement (based on reliable change indices [RCI])≥50% improvement on PDS)

Figure 20. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response self-rated at 6-month follow-up (number of people showing ≥50% improvement on PDS)

Figure 21. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response clinician-rated (number of people showing improvement of at least 10 points on CAPS/clinically significant improvement on CAPS based on reliable change indices [RCI])

Figure 22. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at end-point (BAI/HADS-A/STAI State/HSCL-25 Anxiety/HAM-A change score)

Figure 23. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 2-month follow-up (STAI State change score)

Figure 24. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 5-6 month follow-up (BAI/HSCL-25 Anxiety change score)

Figure 25. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 1-year follow-up (STAI State change score)

Figure 26. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at endpoint (BDI/BDI-II/CES-D/HADS-D/HSCL-25 Depression/HAMD change score)

Figure 27. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 6-7 week follow-up (BDI/BDI-II change score)

Figure 28. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 3-6 month follow-up (BDI-II/CES-D/HSCL-25 Depression change score)

Figure 29. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 1-year follow-up (BDI change score)

Figure 30. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Dissociative symptoms at endpoint (DES change score)

Figure 31. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Dissociative symptoms at 2-month follow-up (DES change score)

Figure 32. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Dissociative symptoms at 1-year follow-up (DES change score)

Figure 33. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Emotional and behavioural problems: Anger (STAXI change score)

Figure 34. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Substance abuse (number of days of primary substance use in past 30 days; ASI-Lite change score)

Figure 35. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Global functioning (GAF change score)

Figure 36. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Functional impairment at endpoint (SDS/SAS-SR change score)

Figure 37. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Functional impairment at 6-month follow-up (SDS change score)

Figure 38. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Relationship difficulties (IIP change score)

Figure 39. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life at endpoint (WHO-5/SF-36 mental health/Q-LES-Q-SF/QOLI; change score)

Figure 40. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life at 6-week follow-up (WHO-5 change score)

Figure 41. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life at 3-month follow-up (WHO-5 change score)

Figure 42. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss of follow-up)

Sub-analysis by specific treatment: Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 43. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at end-point (PCL/SPTSS/HTQ/MPSS/PDS/PSS-SR/IES-R change score)

Figure 44. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS/HTQ/SI-PTSD/PSS-I change score)

Figure 45. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by diagnostic status at baseline: Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 46. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (PCL/SPTSS/HTQ/MPSS/PDS/PSS-SR/IES-R change score)

Figure 47. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS/HTQ/SI-PTSD/PSS-I change score)

Figure 48. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by diagnostic status at baseline: Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 49. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (PCL/SPTSS/HTQ/MPSS/PDS/PSS-SR/IES-R change score)

Figure 50. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS/HTQ/SI-PTSD/PSS-I change score)

Figure 51. Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 52. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (IES/IES-R/PDS/PSS-SR/HTQ/DTS/PCL/MPSS change score)

Figure 53. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 1-month follow-up (PCL/PDS change score)

Figure 54. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 3-4 month follow-up (PCL/PDS/IES-R change score)

Figure 55. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 5-6 month follow-up (IES-R/PDS change score)

Figure 56. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 9-12 month follow-up (PDS change score)

Figure 57. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS/HTQ/PSS-I/SI-PTSD change score)

Figure 58. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at 1-month follow-up (CAPS change score)

Figure 59. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at 3-4 month follow-up (CAPS change score)

Figure 60. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at 5-6 month follow-up (CAPS/HTQ/PSS-I/PDS change score)

Figure 61. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at 5-6 month follow-up (CAPS/PDS-I/CIDI-PT SD change score)

Figure 62. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at endpoint (number of people no longer meeting diagnostic criteria/above threshold on a scale for PTSD)

Figure 63. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at 1-3 month follow-up (number of people no longer meeting diagnostic criteria for PTSD)

Figure 64. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at 6-month follow-up (number of people no longer meeting diagnostic criteria for PTSD)

Figure 65. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at 1-year follow-up (number of people no longer meeting diagnostic criteria for PTSD)

Figure 66. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response self-rated at endpoint (number of people showing clinically significant improvement based on reliable change indices [RCI] on IES/IES-R/DTS)

Figure 67. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response self-rated at 6-month follow-up (number of people showing clinically significant improvement (based on reliable change indices [RCI] on PDS)

Figure 68. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response clinician-rated at endpoint (number of people showing clinically significant improvement based on reliable change indices [RCI]/improvement at least 12/30 points on CAPS)

Figure 69. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response clinician-rated at 1-month follow-up (number of people showing clinically significant improvement based on reliable change indices [RCI]/improvement of at least 12 points on CAPS)

Figure 70. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Dissociative symptoms at endpoint (DES change score)

Figure 71. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Dissociative symptoms at follow-up (DES change score); Multiple incident index trauma

Figure 72. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at endpoint (BAI/HAM-A/STAI State change score)

Figure 73. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 1-month follow-up (STAI State change score)

Figure 74. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 3-month follow-up (BAI/STAI State change score)

Figure 75. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 5-6 month follow-up (BAI/STAI State change score)

Figure 76. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 9-12 month follow-up (STAI State change score)

Figure 77. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at endpoint (BDI/BDI-II/CES-D/HAMD/MADRS change score)

Figure 78. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 1-month follow-up (BDI/BDI-II/HAMD change score)

Figure 79. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 3-4 month follow-up (BDI-II/HAMD change score)

Figure 80. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 5-6 month follow-up (BDI-II/HSCL-25 Depression/HAMD change score)

Figure 81. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 9-12 month follow-up (HAMD/BDI-II change score)

Figure 82. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Personality disorder symptoms (BSL change score); Multiple incident index trauma

Figure 83. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Alcohol use disorder symptoms at endpoint (AUDIT/SADQ change score)

Figure 84. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Alcohol use disorder symptoms at 3-5 month follow-up (AUDIT/SADQ change score)

Figure 85. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Alcohol use disorder symptoms at 9 month follow-up (SADQ change score)

Figure 86. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Alcohol use (TLFB: Percent days abstinent from alcohol, change score); Multiple incident index trauma

Figure 87. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Alcohol use (TLFB: Percent drinking days, change score); Unclear multiplicity of index trauma

Figure 88. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Alcohol use (TLFB: Drinks per drinking day, change score); Single incident index trauma

Figure 89. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Drug use (TLFB: Percent days abstinent from drugs, change score); Multiple incident index trauma

Figure 90. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Substance use (number of days primary substance use in past 30 days; ASI-Lite change score)

Figure 91. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Substance dependence remission at endpoint (number of people no longer meeting diagnostic criteria for substance dependence)

Figure 92. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Substance dependence remission at 5-6 month follow-up (number of people no longer meeting diagnostic criteria for substance dependence)

Figure 93. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Substance dependence remission at 9-month follow-up (number of people no longer meeting diagnostic criteria for substance dependence)

Figure 94. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Global functioning (GAF change score); Multiple incident index trauma

Figure 95. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Functional impairment (SDS/M2C change score/SAS endpoint)

Figure 96. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Emotional and behavioural problems: Aggression/Anger (AAS/DARS-7 change score); Multiple incident index trauma

Figure 97. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life (WHO-5/SF-12 change score); Multiple incident index trauma

Figure 98. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Relationship difficulties (ADAS change score); Multiple incident index trauma

Figure 99. Trauma-focused CBT + medication/TAU versus medication/TAU only (or+ attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by specific intervention: Trauma-focused CBT+ medication/TAU versus medication/TAU-only (or + attention –placebo) for delayed treatment (>3 months) of clinically important symptoms /PTSD

Figure 100. Trauma-focused CBT+ medication/TAU versus medication/TAU-only (or + attention –placebo) for delayed treatment (>3 months) of clinically important symptoms /PTSD: PTSD symptomatology for self-rated at endpoint (IES/IES-R/PDS/PSS-SR/HTQ/DTS/PCL/MPSS change score)

Figure 101. Trauma-focused CBT+ medication/TAU versus medication/TAU-only (or + attention –placebo) for delayed treatment (>3 months) of clinically important symptoms /PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS/HTQ/PSS-I/SI-PTSD change score)

Figure 102. Trauma-focused CBT+ medication/TAU versus medication/TAU-only (or + attention –placebo) for delayed treatment (>3 months) of clinically important symptoms /PTSD: Discontinuation (loss to follow-up)

Sub-analysis by diagnostic status at baseline: Trauma-focused CBT + medication/TAU versus medication/TAU-only (or + attention-placebo) for delayed treatment (> 3-months) of clinically important symptoms/PTSD

Figure 103. Trauma-focused CBT + medication/TAU versus medication/TAU-only (or + attention-placebo) for delayed treatment (> 3-months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (IES/IES-R/PDS/PSS-SR/HTQ/DTS/PCL/MPSS change score)

Figure 104. Trauma-focused CBT + medication/TAU versus medication/TAU-only (or + attention-placebo) for delayed treatment (> 3-months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at end-point (CAPS/HTQ/PSS-I/SI-PTSD change score)

Figure 105. Trauma-focused CBT + medication/TAU versus medication/TAU-only (or + attention-placebo) for delayed treatment (> 3-months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by trauma type: Trauma-focused CBT + medication/TAU versus medication/TAU-only (or + attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 106. Trauma-focused CBT + medication/TAU versus medication/TAU-only (or + attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (IES/IES-R/PDS/PSS-SR/HTQ/DTS/PCL/MPSS change score)

Figure 107. Trauma-focused CBT + medication/TAU versus medication/TAU-only (or + attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS/HTQ/PSS-I/SI-PTSD change score)

Figure 108. Trauma-focused CBT + medication/TAU versus medication/TAU-only (or + attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by personality disorder: Trauma-focused CBT+TAU versus TAU-only for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 109. Trauma-focused CBT+TAU versus TAU-only for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS change score)

Figure 110. Trauma-focused CBT+TAU versus TAU-only for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (PDS change score)

Figure 111. Trauma-focused CBT+TAU versus TAU-only for delayed treatment (>3 months) of clinically important symptoms/PTSD: Global functioning at endpoint (GAF change score)

Figure 112. Trauma-focused CBT+TAU versus TAU-only for delayed treatment (>3 months) of clinically important symptoms/PTSD: Dissociative symptoms at endpoint (DES change score)

Figure 113. Trauma-focused CBT+TAU versus TAU-only for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at endpoint (BDI-II change score)

Figure 114. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (IES/IES-R/PSS-SR change score)

Figure 115. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 3-month follow-up (PSS-SR change score)

Figure 116. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 6-month follow-up (PSS-SR change score)

Figure 117. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS/SI-PTSD change score)

Figure 118. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at 3-month follow-up (CAPS change score)

Figure 119. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at 6-month follow-up (CAPS change score)

Figure 120. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at endpoint (number of people no longer meeting diagnostic criteria or no longer above clinical threshold on scale for PTSD)

Figure 121. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at 3-month follow-up (number of people no longer above clinical threshold on scale for PTSD)

Figure 122. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at 6-month follow-up (number of people no longer meeting diagnostic criteria for PTSD)

Figure 123. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response self-rated at endpoint (number of people showing clinically significant improvement based on reliable change indices [RCI] on IES)

Figure 124. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response self-rated at 15-month follow-up (number of people showing clinically significant improvement based on reliable change indices [RCI] on IES)

Figure 125. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Dissociative symptoms at endpoint (DES/CAPS dissociation cluster change score)

Figure 126. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Dissociative symptoms at 3-month follow-up (CAPS dissociation cluster change score)

Figure 127. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Dissociative symptoms at 6-month follow-up (DES change score)

Figure 128. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at endpoint (STAI State/HADS-A/HAM-A change score)

Figure 129. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 6-month follow-up (STAI State change score)

Figure 130. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at endpoint (BDI/BDI-II/HADS-D/MADRS change score)

Figure 131. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 3-month follow-up (BDI change score)

Figure 132. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 6-month follow-up (BDI change score)

Figure 133. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Functional impairment (SDS change score)

Figure 134. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by specific intervention: Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 135. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (IES/IES-R/PSS-SR change score)

Figure 136. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS/SI-PTSD change score)

Figure 137. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by diagnostic status at baseline: Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 138. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (IES/IES-R/PSS-SR change score)

Figure 139. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS/SI-PTSD change score)

Figure 140. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by trauma type: Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 141. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (IES/IES-R/PSS-SR change score)

Figure 142. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS/SI-PTSD change score)

Figure 143. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by diagnostic status at baseline: Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 144. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (PCL/PDS/PSS-SR change score)

Figure 145. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS/PSS-I change score)

Figure 146. Trauma-focused CBT (±TAU) versus eye movement desensitisation and reprocessing (EMDR; ±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 147. Trauma-focused CBT (±TAU) versus non-trauma-focused CBT (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at follow-up (PCL change score); Multiple incident index trauma

Figure 148. Trauma-focused CBT (±TAU) versus non-trauma-focused CBT (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (PSS-I/CAPS change score)

Figure 149. Trauma-focused CBT (±TAU) versus non-trauma-focused CBT (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at follow-up (CAPS change score); Multiple incident index trauma

Figure 150. Trauma-focused CBT (±TAU) versus non-trauma-focused CBT (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at endpoint (number of people no longer meeting diagnostic criteria for PTSD)

Figure 151. Trauma-focused CBT (±TAU) versus non-trauma-focused CBT (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at follow-up (number of people no longer meeting diagnostic criteria for PTSD); Multiple incident index trauma

Figure 152. Trauma-focused CBT (±TAU) versus non-trauma-focused CBT (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response clinician-rated at endpoint (number of people showing clinically significant improvement based on reliable change indices [RCI] on PSS-I)

Figure 153. Trauma-focused CBT (±TAU) versus non-trauma-focused CBT (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms (STAI Stage change score)

Figure 154. Trauma-focused CBT (±TAU) versus non-trauma-focused CBT (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at endpoint (BDI change score)

Figure 155. Trauma-focused CBT (±TAU) versus non-trauma-focused CBT (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at follow-up (BDI/HAMD change score); Multiple incident index trauma

Figure 156. Trauma-focused CBT (±TAU) versus non-trauma-focused CBT (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Sleeping difficulties (PSQI change score); Multiple incident index trauma

Figure 157. Trauma-focused CBT (±TAU) versus non-trauma-focused CBT (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life (QOLI/SF-36 MH change score); Multiple incident index trauma

Figure 158. Trauma-focused CBT (±TAU) versus non-trauma-focused CBT (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 159. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (PCL/PDS/PSS-SR change score)

Figure 160. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 2-4 month follow-up (PCL/PDS/PSS-SR change score)

Figure 161. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 6-8 month follow-up (PCL/PDS/PSS-SR change score)

Figure 162. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 1-year follow-up (PCL/PDS/PSS-SR change score)

Figure 163. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 2-year follow-up (PCL change score)

Figure 164. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (PCL change score)

Figure 165. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at 3-month follow-up (CAPS change score)

Figure 166. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at 6-month follow-up (CAPS change score)

Figure 167. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at 1-year follow-up (CAPS/PSS-I/CIDI-PTSD change score)

Figure 168. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at 2-year follow-up (CAPS change score)

Figure 169. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at endpoint (number of people no longer meeting diagnostic criteria or no longer above threshold on a scale for PTSD)

Figure 170. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at 3-month follow-up (number of people no longer meeting diagnostic criteria or no longer above threshold on a scale for PTSD)

Figure 171. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at 6-8 month follow-up (number of people no longer meeting diagnostic criteria or no longer above threshold on a scale for PTSD)

Figure 172. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at 1-year follow-up (number of people no longer meeting diagnostic criteria or no longer above threshold on a scale for PTSD)

Figure 173. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response clinician-rated (number of people showing clinically significant improvement on PSS-I based on reliable change indices [RCI])

Figure 174. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at endpoint (BAI/STAI State/BSI anxiety/HAM-A change score)

Figure 175. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 3-month follow-up (BAI/STAI State change score)

Figure 176. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 6-8 month follow-up (BAI/STAI State/HAM-A change score)

Figure 177. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 1-year follow-up (STAI State change score)

Figure 178. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 2-year follow-up (STAI State change score)

Figure 179. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at endpoint (BDI/BDI-II/BDI-13/BSI Depression change score)

Figure 180. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 3-month follow-up (BDI/BDI-II change score)

Figure 181. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 6-8 month follow-up (BDI-II/BDI-13 change score)

Figure 182. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 1-year follow-up (BDI change score)

Figure 183. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 2-year follow-up (BDI change score)

Figure 184. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Functional impairment (SDS change score); Unclear multiplicity of index event

Figure 185. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Global functioning (GAF change score); Single incident index trauma

Figure 186. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Relationship difficulties (IIP change score); Multiple incident index trauma

Figure 187. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life at endpoint (QOLI/Q-LES-Q-SF/SF-12 change score)

Figure 188. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life at 3-4 month follow-up (Q-LES-Q-SF/SF-12 change score)

Figure 189. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life at 6-month follow-up (Q-LES-Q-SF/SF-12 change score)

Figure 190. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life at 1-year follow-up (SF-12 change score)

Figure 191. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by trauma type: Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 192. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (PCL/PDS/PSS-SR change score)

Figure 193. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS/PSS-I change score)

Figure 194. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by specific intervention: Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 195. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (PCL/PDS/PSS-SR change score)

Figure 196. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS/PSS-I change score)

Figure 197. Trauma-focused CBT (±TAU) versus counselling (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 198. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (PCL change score)

Figure 199. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 2-3 month follow-up (PCL change score)

Figure 200. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 4-month follow-up (PCL change score)

Figure 201. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 6-month follow-up (PCL change score)

Figure 202. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS change score)

Figure 203. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at 1-3 month follow-up (CAPS change score)

Figure 204. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at 4-month follow-up (CAPS change score)

Figure 205. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at 6-month follow-up (CAPS change score)

Figure 206. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at endpoint (number of people no longer meeting diagnostic criteria for PTSD)

Figure 207. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at 1-3 month follow-up (number of people no longer meeting diagnostic criteria for PTSD)

Figure 208. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at 6-month follow-up (number of people no longer meeting diagnostic criteria for PTSD)

Figure 209. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response clinician-rated at endpoint (number of people showing clinically significant improvement based on reliable change indices [RCI] on PSS-I/at least 10-point improvement on CAPS)

Figure 210. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response clinician-rated at 3-month follow-up (number of people showing at least 10-point improvement on CAPS)

Figure 211. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response clinician-rated at 6-month follow-up (number of people showing at least 10-point improvement on CAPS)

Figure 212. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Dissociative symptoms (DES change score); Multiple incident index trauma

Figure 213. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at endpoint (BAI/STAI State/BSI Anxiety change score)

Figure 214. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 3-month follow-up (BAI/STAI State change score)

Figure 215. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 6-month follow-up (BAI/STAI State change score)

Figure 216. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at endpoint (BDI/BDI-II/QIDS/BSI Depression change score)

Figure 217. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 2-3 month follow-up (BDI/BDI-II/QIDS change score)

Figure 218. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 4-month follow-up (QIDS change score)

Figure 219. Depression symptoms at 6-month follow-up (BDI/BDI-II/QIDS score)

Figure 220. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Emotional and behavioural problems: Anger (STAXI change score); Multiple incident index trauma

Figure 221. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life (QOLI change score); Multiple incident index trauma

Figure 222. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loos to follow-up)

Sub-analysis by specific intervention: Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 223. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (PCL change score)

Figure 224. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS change score)

Figure 225. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by diagnostic status at baseline: Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 226. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (PCL change score)

Figure 227. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS change score)

Figure 228. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by trauma type: Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 229. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (PCL change score)

Figure 230. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS change score)

Figure 231. Trauma-focused CBT (±TAU) versus present-centred therapy (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 232. Trauma-focused CBT (+TAU) versus metacognitive (+TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (PDS change score); Single incident index trauma

Figure 233. Trauma-focused CBT (+TAU) versus metacognitive (+TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission (number of people no longer meeting diagnostic criteria for PTSD)

Figure 234. Trauma-focused CBT (+TAU) versus metacognitive (+TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response self-rated (number of people showing clinically significant improvement based on at least 10-point improvement on IES)

Figure 235. Trauma-focused CBT (+TAU) versus metacognitive (+TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms (BAI change score); Single incident index trauma

Figure 236. Trauma-focused CBT (+TAU) versus metacognitive (+TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (BDI-II change score); Single incident index trauma

Figure 237. Trauma-focused CBT (+TAU) versus metacognitive (+TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 238. Trauma-focused CBT versus interpersonal psychotherapy (IPT) for delayed treatment (>3-months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score)

Figure 239. Trauma-focused CBT versus interpersonal psychotherapy (IPT) for delayed treatment (>3-months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (PSS-SR change score)

Figure 240. Trauma-focused CBT versus interpersonal psychotherapy (IPT) for delayed treatment (>3-months) of clinically important symptoms/PTSD: Remission (number of people scoring <20 on CAPS)

Figure 241. Trauma-focused CBT versus interpersonal psychotherapy (IPT) for delayed treatment (>3-months) of clinically important symptoms/PTSD: Response (number of people showing ≥30% improvement on CAPS)

Figure 242. Trauma-focused CBT versus interpersonal psychotherapy (IPT) for delayed treatment (>3-months) of clinically important symptoms/PTSD: Depression symptoms (HAMD change score)

Figure 243. Trauma-focused CBT versus interpersonal psychotherapy (IPT) for delayed treatment (>3-months) of clinically important symptoms/PTSD: Functional impairment (SAS change score)

Figure 244. Trauma-focused CBT versus interpersonal psychotherapy (IPT) for delayed treatment (>3-months) of clinically important symptoms/PTSD: Quality of life (Q-LES-Q-SF change score)

Figure 245. Trauma-focused CBT versus interpersonal psychotherapy (IPT) for delayed treatment (>3-months) of clinically important symptoms/PTSD: Relationship difficulties (IIP change score)

Figure 246. Trauma-focused CBT versus interpersonal psychotherapy (IPT) for delayed treatment (>3-months) of clinically important symptoms/PTSD: Discontinuation (loss of follow-up)

Figure 247. Trauma-focused CBT (+TAU) versus psychodynamic therapy (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (IES change score); Single incident index trauma

Figure 248. Trauma-focused CBT (±TAU) versus self-help (without support; ± TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score)

Figure 249. Trauma-focused CBT (±TAU) versus self-help (without support; ± TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at endpoint (number of people no longer meeting diagnostic criteria or scoring below clinical threshold on a scale)

Figure 250. Trauma-focused CBT (±TAU) versus self-help (without support; ± TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at 6-month follow-up (number of people scoring <14 on PDS) S)

Figure 251. Trauma-focused CBT (±TAU) versus self-help (without support; ± TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response at endpoint (number of people showing ≥50% improvement on PDS)

Figure 252. Trauma-focused CBT (±TAU) versus self-help (without support; ± TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response at 6-month follow-up (number of people showing ≥50% improvement on PDS)

Figure 253. Trauma-focused CBT (±TAU) versus self-help (without support; ± TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at endpoint (BDI-II change score)

Figure 254. Trauma-focused CBT (±TAU) versus self-help (without support; ± TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 6-month follow-up (BDI-II change score)

Figure 255. Trauma-focused CBT (±TAU) versus self-help (without support; ± TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at endpoint (BAI change score)

Figure 256. Trauma-focused CBT (±TAU) versus self-help (without support; ± TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 6-month follow-up (BAI change score)

Figure 257. Trauma-focused CBT (±TAU) versus self-help (without support; ± TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Functional impairment at endpoint (SDS change score)

Figure 258. Trauma-focused CBT (±TAU) versus self-help (without support; ± TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Functional impairment at 6-month follow-up (SDS change score)

Figure 259. Trauma-focused CBT (±TAU) versus self-help (without support; ± TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 260. Trauma-focused CBT versus self-help support for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (IES change score); Single incident index trauma

Figure 261. Trauma-focused CBT versus self-help support for delayed treatment (>3 months) of clinically important symptoms/PTSD: Dissociative symptoms (DES change score); Single incident index trauma

Figure 262. Trauma-focused CBT versus self-help support for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms (STAI State change score); Single incident index trauma

Figure 263. Trauma-focused CBT versus self-help support for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (BDI change score); Single incident index trauma

Figure 264. Trauma-focused CBT (+TAU) versus hypnotherapy (+TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (IES change score); Single incident index trauma

Figure 265. Trauma-focused CBT versus psychoeducational session for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (IES change score)

Figure 266. Trauma-focused CBT versus psychoeducational session for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 3-month follow-up (IES change score)

Figure 267. Trauma-focused CBT versus psychoeducational session for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 6-month follow-up (IES change score)

Figure 268. Trauma-focused CBT versus psychoeducational session for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 269. Trauma-focused CBT (±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (PCL/PSS-SR change score)

Figure 270. Trauma-focused CBT (±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 3-month follow-up (PCL/PSS-SR change score)

Figure 271. Trauma-focused CBT (±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS change score)

Figure 272. Trauma-focused CBT (±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at 3-month follow-up (CAPS change score)

Figure 273. Trauma-focused CBT (±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at endpoint (number of people scoring <20 on CAPS)

Figure 274. Trauma-focused CBT (±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at 3-month follow-up (number of people scoring <20 on CAPS)

Figure 275. Trauma-focused CBT (±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response (number of people showing ≥30% improvement on CAPS)

Figure 276. Trauma-focused CBT (±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Dissociative symptoms (CAPS dissociation cluster change score); Unclear multiplicity of index trauma

Figure 277. Trauma-focused CBT (±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms (SCL-90: Anxiety; change score); unclear multiplicity of index trauma

Figure 278. Trauma-focused CBT (±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at endpoint (HAMD/BDI change score)

Figure 279. Trauma-focused CBT (±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 3-month follow-up (BDI change score)

Figure 280. Trauma-focused CBT (±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Functional impairment (SAS change score)

Figure 281. Trauma-focused CBT (±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life (Q-LES-Q-SF change score)

Figure 282. Trauma-focused CBT (±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Relationship difficulties (IIP change score)

Figure 283. Trauma-focused CBT (±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 284. Trauma-focused CBT versus acupuncture for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (PSS-SR change score); unclear multiplicity of index trauma

Figure 285. Trauma-focused CBT versus acupuncture for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission (number of people scoring <16 on PSS-SR); unclear multiplicity of index trauma

Figure 286. Trauma-focused CBT versus acupuncture for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (HSCL-25: Depression, change score); unclear multiplicity of index trauma

Figure 287. Trauma-focused CBT versus acupuncture for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms (HSCL-25; Anxiety, change score); unclear multiplicity of index trauma

Figure 288. Trauma-focused CBT versus acupuncture for delayed treatment (>3 months) of clinically important symptoms/PTSD: Functional impairment (SDS change score); unclear multiplicity of index trauma

Figure 289. Trauma-focused CBT versus acupuncture for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 290. Trauma-focused CBT versus SSRIs for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (HTQ/PDS change score)

Figure 291. Trauma-focused CBT versus SSRIs for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 1-year follow-up (PDS change score)

Figure 292. Trauma-focused CBT versus SSRIs for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (PSS-I/SI-PTSD change score)

Figure 293. Trauma-focused CBT versus SSRIs for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission (number of people no longer meeting diagnostic criteria for PTSD)

Figure 294. Trauma-focused CBT versus SSRIs for delayed treatment (>3 months) of clinically important symptoms/PTSD: Dissociative symptoms (DES change score)

Figure 295. Trauma-focused CBT versus SSRIs for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at endpoint (HAM-A/STAI State change score)

Figure 296. Trauma-focused CBT versus SSRIs for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 1-year follow-up (STAI State change score)

Figure 297. Trauma-focused CBT versus SSRIs for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at endpoint (HAMD/BDI/BDI-II change score)

Figure 298. Trauma-focused CBT versus SSRIs for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 1-year follow-up (BDI-II change score)

Figure 299. Trauma-focused CBT versus SSRIs for delayed treatment (>3 months) of clinically important symptoms/PTSD: Functional impairment (SDS change score)

Figure 300. Trauma-focused CBT versus SSRIs for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life (WHO-5 change score)

Figure 301. Trauma-focused CBT versus SSRIs for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 302. Trauma-focused CBT + SSRIs versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (HTQ change score)

Figure 303. Trauma-focused CBT + SSRIs versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms (HAM-A change score)

Figure 304. Trauma-focused CBT + SSRIs versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (HAMD change score)

Figure 305. Trauma-focused CBT + SSRIs versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Functional impairment (SDS change score)

Figure 306. Trauma-focused CBT + SSRIs versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life (WHO-5 change score)

Figure 307. Trauma-focused CBT + SSRIs versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Non-trauma-focused CBP

Figure 308. Non-trauma-focused CBT (±TAU) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report (PCL/DTS/PDS/PSS-SR/MPSS-SR change score)

Figure 309. Non-trauma-focused CBT (±TAU) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS change score)

Figure 310. Non-trauma-focused CBT (±TAU) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at 3-month follow-up (CAPS change score)

Figure 311. Non-trauma-focused CBT (±TAU) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at endpoint (number of people no longer meeting diagnostic criteria/above threshold on a scale for PTSD)

Figure 312. Non-trauma-focused CBT (±TAU) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at 3-month follow-up (number of people no longer meeting diagnostic criteria)

Figure 313. Non-trauma-focused CBT (±TAU) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Dissociative symptoms (DES; change score)

Figure 314. Non-trauma-focused CBT (±TAU) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Sleeping difficulties (ISI/SQI change score)

Figure 315. Non-trauma-focused CBT (±TAU) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptom at endpoint (BDI/BDI-II change score)

Figure 316. Non-trauma-focused CBT (±TAU) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 3-month follow-up (BDI change score)

Figure 317. Non-trauma-focused CBT (±TAU) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Alcohol use (TLFB Number of drinking days; change score); Multiple incident index trauma

Figure 318. Non-trauma-focused CBT (±TAU) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Drug use (TLFB Number of drug use days; change score); Multiple incident index trauma

Figure 319. Non-trauma-focused CBT (±TAU) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by specific intervention: Non-trauma-focused CBT (±TAU) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 320. Non-trauma-focused CBT (±TAU) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report (PCL/DTS/PDS/PSS-SR/MPSS-SR change score)

Figure 321. Non-trauma-focused CBT (±TAU) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS change score)

Figure 322. Non-trauma-focused CBT (±TAU) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by diagnostic status at baseline

Figure 323. Non-trauma-focused CBT (±TAU) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report (PCL/DTS/PDS/PSS-SR/MPSS-SR change score)

Figure 324. Non-trauma-focused CBT (±TAU) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS change score)

Figure 325. Non-trauma-focused CBT (±TAU) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by trauma type

Figure 326. Non-trauma-focused CBT (±TAU) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report (PCL/DTS/PDS/PSS-SR/MPSS-SR change score)

Figure 327. Non-trauma-focused CBT (±TAU) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS change score)

Figure 328. Non-trauma-focused CBT (±TAU) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 329. Non-trauma-focused CBT (±TAU) versus attention-placebo (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report at endpoint (PCL/PSS-SR change score)

Figure 330. Non-trauma-focused CBT (±TAU) versus attention-placebo (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report at 3-month follow-up (PCL change score)

Figure 331. Non-trauma-focused CBT (±TAU) versus attention-placebo (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score)

Figure 332. Non-trauma-focused CBT (±TAU) versus attention-placebo (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response (number of people showing clinically significant improvement, based on reliable change indices [RCI])

Figure 333. Non-trauma-focused CBT (±TAU) versus attention-placebo (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (CES-D change score); Multiple incident index trauma

Figure 334. Non-trauma-focused CBT (±TAU) versus attention-placebo (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Drug use (Substance use Inventory: Number of days participants used drugs during the past 7 days; change score)

Figure 335. Non-trauma-focused CBT (±TAU) versus attention-placebo (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life at endpoint (SF-36 change score)

Figure 336. Non-trauma-focused CBT (±TAU) versus attention-placebo (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life at 3-month follow-up (SF-36 change score)

Figure 337. Non-trauma-focused CBT (±TAU) versus attention-placebo (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 338. Non-trauma-focused CBT (+TAU) versus psychoeducational group (+TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report (DTS change score); Multiple incident index trauma

Figure 339. Non-trauma-focused CBT (+TAU) versus psychoeducational group (+TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score); Multiple incident index trauma

Figure 340. Non-trauma-focused CBT (+TAU) versus psychoeducational group (+TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (HAMD change score); Multiple incident index trauma

Figure 341. Non-trauma-focused CBT (+TAU) versus psychoeducational group (+TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 342. Non-trauma-focused CBT versus counselling for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (PSS-I change score)

Figure 343. Non-trauma-focused CBT versus counselling for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission (number of people no longer meeting diagnostic criteria for PTSD)

Figure 344. Non-trauma-focused CBT versus counselling for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response (number of people showing clinically significant improvement based on reliable change indices [RCI] on PSS-I)

Figure 345. Non-trauma-focused CBT versus counselling for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms (STAI State change score)

Figure 346. Non-trauma-focused CBT versus counselling for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (BDI change score)

Figure 347. Non-trauma-focused CBT versus counselling for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 348. Non-trauma-focused CBT versus present-centred therapy for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score); Multiple incident index trauma

Figure 349. Non-trauma-focused CBT versus present-centred therapy for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission (number of people no longer meeting diagnostic criteria for PTSD); Multiple incident index trauma

Figure 350. Non-trauma-focused CBT versus present-centred therapy for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (BDI change score); Multiple incident index trauma

Figure 351. Non-trauma-focused CBT versus present-centred therapy for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Present-centred therapy (+TAU)

Figure 352. Present-centred therapy (+TAU) versus TAU for early treatment (1-3 months) clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score); Multiple incident index trauma

Figure 353. Present-centred therapy (+TAU) versus TAU for early treatment (1-3 months) clinically important symptoms/PTSD: Response (number of people showing improvement of at least 26 points on CAPS); Multiple incident index trauma

Figure 354. Present-centred therapy (+TAU) versus TAU for early treatment (1-3 months) clinically important symptoms/PTSD: Depression symptoms (BDI change score); Multiple incident index trauma

Figure 355. Present-centred therapy (+TAU) versus TAU for early treatment (1-3 months) clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 356. Present-centred therapy versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score)

Figure 357. Present-centred therapy versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission (number of people no longer meeting diagnostic criteria for PTSD)

Figure 358. Present-centred therapy versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Dissociative symptoms (DES; change score)

Figure 359. Present-centred therapy versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms (STAI state; change score)

Figure 360. Present-centred therapy versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (BDI change score)

Figure 361. Present-centred therapy versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Emotional and behavioural problems: Anger (STAXI change score)

Figure 362. Present-centred therapy versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life (QOLI change score)

Figure 363. Present-centred therapy versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Metacognitive therapy

Figure 364. Metacognitive therapy (±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (IES/PDS change score)

Figure 365. Metacognitive therapy (±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response self-rated at endpoint (number of people showing clinically significant improvement based on at least 10-point improvement on IES)

Figure 366. Metacognitive therapy (±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms (BAI change score)

Figure 367. Metacognitive therapy (±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (BDI-II change score)

Figure 368. Metacognitive therapy (±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Reconsolidation of traumatic memories (RTM) intervention

Figure 369. Reconsolidation of traumatic memories (RTM) intervention + TAU versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (PSS-I change score)

Figure 370. Reconsolidation of traumatic memories (RTM) intervention + TAU versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (PCL endpoint score)

Figure 371. Reconsolidation of traumatic memories (RTM) intervention + TAU versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Single-session behavioural therapy

Figure 372. Single-session behavioural therapy versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 6-week follow-up (TSSC change score)

Figure 373. Single-session behavioural therapy versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at 6-8 week follow-up (CAPS change score)

Figure 374. Single-session behavioural therapy versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response at 6-week follow-up (number of people rated as ‘much’ or ‘very much’ improved on CGI-I)

Figure 375. Single-session behavioural therapy versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Functional impairment at 6-8 week follow-up (WSA change score)

Figure 376. Single-session behavioural therapy versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 6-8 week follow-up (BDI change score)

Problem solving

Figure 377. Problem solving versus supportive counselling for early treatment (1-3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report (IES-R endpoint score); Single incident trauma

Figure 378. Problem solving versus supportive counselling for early treatment (1-3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Eye movement desensitisation and reprocessing (EMDR)

Figure 379. Eye movement desensitisation and reprocessing (EMDR) versus supportive counselling for early treatment (1-3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (SPRINT change score); Multiple incident index trauma

Figure 380. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report at endpoint (IES/IES-R/Trauma Symptoms Inventory/PDS/PSS-SR change scores/M-PTSD endpoint)

Figure 381. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report at 1-month follow-up (IES-R change score)

Figure 382. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (SI-PTSD/CAPS change score)

Figure 383. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at endpoint (number of people no longer meeting diagnostic criteria for PTSD)

Figure 384. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at 1-month follow-up (number of people no longer meeting diagnostic criteria for PTSD)

Figure 385. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response self-rated (number of people showing clinically significant improvement, based on reliable change indices [RCI] on IES)

Figure 386. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Dissociative symptoms (DES change score)

Figure 387. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms (STAI State/HAM-A change score)

Figure 388. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at endpoint (BDI/BDI-II /MADRS change score)

Figure 389. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 1-month follow-up (BDI-II change score)

Figure 390. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Functional impairment (SDS change score)

Figure 391. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by diagnostic status at baseline: Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 392. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report at endpoint (IES/IES-R/Trauma Symptoms Inventory/PDS/PSS-SR change scores/M-PTSD endpoint)

Figure 393. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (SI-PTSD/CAPS change score)

Figure 394. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by trauma type

Figure 395. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report at endpoint (IES/IES-R/Trauma Symptoms Inventory/PDS/PSS-SR change scores/M-PTSD endpoint)

Figure 396. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (SI-PTSD/CAPS change score)

Figure 397. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 398. Eye movement desensitisation and reprocessing (EMDR) versus pill placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score)

Figure 399. Eye movement desensitisation and reprocessing (EMDR) versus pill placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission (number of people scoring <20 on CAPS)

Figure 400. Eye movement desensitisation and reprocessing (EMDR) versus pill placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (BDI-II; change score)

Figure 401. Eye movement desensitisation and reprocessing (EMDR) versus pill placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 402. Eye movement desensitisation and reprocessing (EMDR) versus supportive counselling for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (IES change score)

Figure 403. Eye movement desensitisation and reprocessing (EMDR) versus supportive counselling for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms (STAI State; change score)

Figure 404. Eye movement desensitisation and reprocessing (EMDR) versus supportive counselling for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (BDI change score)

Figure 405. Eye movement desensitisation and reprocessing (EMDR) versus supportive counselling for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 406. Eye movement desensitisation and reprocessing (EMDR) versus non-trauma-focused CBT for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score); Multiple incident index trauma

Figure 407. Eye movement desensitisation and reprocessing (EMDR) versus non-trauma-focused CBT for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (HTQ change score); Multiple incident index trauma

Figure 408. Eye movement desensitisation and reprocessing (EMDR) versus non-trauma-focused CBT for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response (number of people showing improvement of at least 10 points on CAPS at 3-month follow-up)

Figure 409. Eye movement desensitisation and reprocessing (EMDR) versus non-trauma-focused CBT for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms (HSCL-25: Anxiety, change score); Multiple incident index trauma

Figure 410. Eye movement desensitisation and reprocessing (EMDR) versus non-trauma-focused CBT for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (HSCL-25: Depression; Change score); Multiple incident index trauma

Figure 411. Eye movement desensitisation and reprocessing (EMDR) versus non-trauma-focused CBT for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 412. Eye movement desensitisation and reprocessing (EMDR) versus ‘other active psych intervention’ for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (IES change score); Multiple incident index trauma

Figure 413. Eye movement desensitisation and reprocessing (EMDR) versus ‘other active psych intervention’ for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (BDI change score); Multiple incident index trauma

Figure 414. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (IES/PSS-SR change score at endpoint/follow-up)

Figure 415. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 3-month follow-up (PSS-SR change score at endpoint/follow-up)

Figure 416. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 6-month follow-up (IES-R change score)

Figure 417. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS change score)

Figure 418. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at 3-month follow-up (CAPS change score)

Figure 419. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at 6-month follow-up (CAPS change score)

Figure 420. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at endpoint (number of people no longer meeting diagnostic criteria or no longer above clinical threshold on a scale for PTSD)

Figure 421. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at 3-month follow-up (number of people no longer above clinical threshold on a scale for PTSD)

Figure 422. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at 6-month follow-up (number of people no longer meeting diagnostic criteria for PTSD)

Figure 423. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Dissociative symptoms (CAPS dissociation cluster change score); Unclear multiplicity of index trauma

Figure 424. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms (HADS-A/STAI state change score at endpoint/follow-up)

Figure 425. Depression symptoms at endpoint (BDI change score)

Figure 426. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 3-6 month follow-up (BDI/HADS-D change score)

Figure 427. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life (Functional Assessment of Quality of Life in MS; change score)

Figure 428. Eye movement desensitisation and reprocessing (EMDR; ±TAU) versus relaxation (±TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 429. Eye movement desensitisation and reprocessing (EMDR) versus combined somatic and cognitive therapies for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report (PCL-C change score); Single incident trauma

Figure 430. Eye movement desensitisation and reprocessing (EMDR) versus combined somatic and cognitive therapies for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score); Single incident trauma

Figure 431. Eye movement desensitisation and reprocessing (EMDR) versus combined somatic and cognitive therapies for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response self-rated (number of people showing clinically significant improvement (based on reliable change indices [RCI]) on PCL-C); Single incident trauma

Figure 432. Eye movement desensitisation and reprocessing (EMDR) versus combined somatic and cognitive therapies for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response clinician-rated (number of people showing clinically significant improvement [based on RCI] on CAPS); Single incident trauma

Figure 433. Eye movement desensitisation and reprocessing (EMDR) versus combined somatic and cognitive therapies for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms (HADS-A change score); Single incident trauma

Figure 434. Eye movement desensitisation and reprocessing (EMDR) versus combined somatic and cognitive therapies for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (HADS-D change score); Single incident trauma

Figure 435. Eye movement desensitisation and reprocessing (EMDR) versus combined somatic and cognitive therapies for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life (Satisfaction with Life Scale; change score); Single incident trauma

Figure 436. Eye movement desensitisation and reprocessing (EMDR) versus combined somatic and cognitive therapies for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 437. Eye movement desensitisation and reprocessing (EMDR) versus fluoxetine for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score); Multiple incident index trauma

Figure 438. Eye movement desensitisation and reprocessing (EMDR) versus fluoxetine for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission (number of people scoring <20 on CAPS); Multiple incident index trauma

Figure 439. Eye movement desensitisation and reprocessing (EMDR) versus fluoxetine for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (BDI-II change score); Multiple incident index trauma

Figure 440. Eye movement desensitisation and reprocessing (EMDR) versus fluoxetine for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Hypnotherapy

Figure 441. Hypnotherapy + TAU versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (IES change score)

Figure 442. Hypnotherapy followed by trauma-focused CBT versus symptom monitoring followed by trauma-focused CBT for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score); Multiple incident index trauma

Figure 443. Hypnotherapy followed by trauma-focused CBT versus symptom monitoring followed by trauma-focused CBT for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (BDI-II change score); Multiple incident index trauma

Figure 444. Hypnotherapy followed by trauma-focused CBT versus symptom monitoring followed by trauma-focused CBT for delayed treatment (>3 months) of clinically important symptoms/PTSD: Sleeping difficulties (PSQI change score); Multiple incident index trauma

Figure 445. Hypnotherapy followed by trauma-focused CBT versus symptom monitoring followed by trauma-focused CBT for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 446. Hypnotherapy (+ TAU) versus zolpidem (+ TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report (IES change score); Multiple incident index trauma

Figure 447. Hypnotherapy (+ TAU) versus zolpidem (+ TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (BDI change score); Multiple incident index trauma

Figure 448. Hypnotherapy (+ TAU) versus zolpidem (+ TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up); Multiple incident index trauma

Psychodynamic therapy

Figure 449. Psychodynamic therapy (± TAU) versus waitlist (+ TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (IES change score)

Figure 450. Psychodynamic therapy (± TAU) versus waitlist (+ TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission (number of people no longer met criteria for PTSD based on HTQ DSM-IV PTSD algorithm)

Figure 451. Psychodynamic therapy (± TAU) versus waitlist (+ TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms (HSCL-25: Anxiety; change score)

Figure 452. Psychodynamic therapy (± TAU) versus waitlist (+ TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (HSCL-25: Depression; change score)

Figure 453. Psychodynamic therapy (± TAU) versus waitlist (+ TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Interpersonal psychotherapy

Figure 454. Interpersonal psychotherapy (IPT) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score); Multiple incident index trauma

Figure 455. Interpersonal psychotherapy (IPT) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission (number of people no longer meeting diagnostic criteria for PTSD)

Figure 456. Interpersonal psychotherapy (IPT) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (HAMD change score); Multiple incident index trauma

Figure 457. Interpersonal psychotherapy (IPT) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 458. Interpersonal psychotherapy (IPT) versus relaxation for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score)

Figure 459. Interpersonal psychotherapy (IPT) versus relaxation for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (PSS-SR change score)

Figure 460. Interpersonal psychotherapy (IPT) versus relaxation for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission (number of people scoring <20 on CAPS)

Figure 461. Interpersonal psychotherapy (IPT) versus relaxation for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response (number of people showing ≥30% improvement on CAPS)

Figure 462. Interpersonal psychotherapy (IPT) versus relaxation for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (HAMD change score)

Figure 463. Interpersonal psychotherapy (IPT) versus relaxation for delayed treatment (>3 months) of clinically important symptoms/PTSD: Functional impairment (SAS change score)

Figure 464. Interpersonal psychotherapy (IPT) versus relaxation for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life (Q-LES-Q-SF change score)

Figure 465. Interpersonal psychotherapy (IPT) versus relaxation for delayed treatment (>3 months) of clinically important symptoms/PTSD: Relationship difficulties (IIP change score)

Figure 466. Interpersonal psychotherapy (IPT) versus relaxation for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Counselling

Figure 467. Counselling (± TAU) versus TAU or waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (PCL/PDS/HTQ change score)

Figure 468. Counselling (± TAU) versus TAU or waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 1-4 month follow-up (HTQ/PDS change score)

Figure 469. Counselling (± TAU) versus TAU or waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 8-12 month follow-up (PDS change score)

Figure 470. Counselling (± TAU) versus TAU or waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS change score)

Figure 471. Counselling (± TAU) versus TAU or waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at 1-year follow-up (CIDI-PTSD change score)

Figure 472. Counselling (± TAU) versus TAU or waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at endpoint (number of people no longer meeting diagnostic criteria or no longer above clinical threshold on a scale for PTSD)

Figure 473. Counselling (± TAU) versus TAU or waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at 8-12 month follow-up (number of people no longer meeting diagnostic criteria for PTSD)

Figure 474. Counselling (± TAU) versus TAU or waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at endpoint (BAI/STAI State change score)

Figure 475. Counselling (± TAU) versus TAU or waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 1-month follow-up (HSCL Anxiety change score)

Figure 476. Counselling (± TAU) versus TAU or waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at endpoint (BDI change score)

Figure 477. Counselling (± TAU) versus TAU or waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 1-month follow-up (HSCL Depression change score)

Figure 478. Counselling (± TAU) versus TAU or waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Functional impairment (SDS change score)

Figure 479. Counselling (± TAU) versus TAU or waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Global functioning (GAF change score)

Figure 480. Counselling (± TAU) versus TAU or waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life at endpoint (Q-LES-Q-SF/SF-12 change score)

Figure 481. Counselling (± TAU) versus TAU or waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life at follow-up (SF-12 change score); Multiple incident index trauma

Figure 482. Counselling (± TAU) versus TAU or waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Combined somatic and cognitive therapies

Figure 483. Combined somatic and cognitive therapies (± TAU) versus waitlist (± TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (PCL/MPSS change score)

Figure 484. Combined somatic and cognitive therapies (± TAU) versus waitlist (± TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission (number of people scoring <50 on PCL)

Figure 485. Combined somatic and cognitive therapies (± TAU) versus waitlist (± TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms (SA-45 Anxiety T-score; change score)

Figure 486. Combined somatic and cognitive therapies (± TAU) versus waitlist (± TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (SA-45 Depression T-score change score)

Figure 487. Combined somatic and cognitive therapies (± TAU) versus waitlist (± TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Sleeping difficulties (ISI change score)

Figure 488. Combined somatic and cognitive therapies (± TAU) versus waitlist (± TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by specific intervention: Combined somatic and cognitive therapies (± TAU) versus waitlist (± TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 489. Combined somatic and cognitive therapies (± TAU) versus waitlist (± TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (PCL/MPSS change score)

Figure 490. Combined somatic and cognitive therapies (± TAU) versus waitlist (± TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by trauma type: Combined somatic and cognitive therapies (± TAU) versus waitlist (± TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 491. Combined somatic and cognitive therapies (± TAU) versus waitlist (± TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (PCL/MPSS change score)

Figure 492. Combined somatic and cognitive therapies (± TAU) versus waitlist (± TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Somatic experiencing

Figure 493. Somatic experiencing + TAU versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report (PDS change score)

Figure 494. Somatic experiencing + TAU versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score)

Figure 495. Somatic experiencing + TAU versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (CES-D change score)

Figure 496. Somatic experiencing + TAU versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Resilience-oriented treatment

Figure 497. Resilience-oriented treatment versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report (PDS change score)

Figure 498. Resilience-oriented treatment versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms (STAI state change score)

Figure 499. Resilience-oriented treatment versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (BDI-II change score)

Figure 500. Resilience-oriented treatment versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Attention bias modification

Figure 501. Attrition bias modification versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report (PCL/SRIP; change score)

Figure 502. Attrition bias modification versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score); Multiple incident index trauma

Figure 503. Attrition bias modification versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms (HADS-A change score); Multiple incident index trauma

Figure 504. Attrition bias modification versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (PHQ-9/HADS-D change score); Multiple incident index trauma

Figure 505. Attrition bias modification versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Couple intervention

Figure 506. Couple intervention versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response (number of people showing improvement of at least 10 points on CAPS)

Figure 507. Couple intervention versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission (number of people who no longer met DSM-IV-TR diagnostic criteria and CAPS score<45)

Figure 508. Couple intervention versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response for relationship difficulties (number of participants showing improvement of at least 10 points on DAS)

Figure 509. Couple intervention versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission for relationship difficulties (number of participants scoring ≥98 on DAS)

Figure 510. Couple intervention versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 511. Couple intervention versus psycho-education sessions for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score); Multiple incident index trauma

Figure 512. Couple intervention versus psycho-education sessions for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (PCL-M change score); Multiple incident index trauma

Figure 513. Couple intervention versus psycho-education sessions for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission (number of people scoring <45 on CAPS at endpoint)

Figure 514. Couple intervention versus psycho-education sessions for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms (STAI: State change score); Multiple incident index trauma

Figure 515. Couple intervention versus psycho-education sessions for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (CES-D change score); Multiple incident index trauma

Figure 516. Couple intervention versus psycho-education sessions for delayed treatment (>3 months) of clinically important symptoms/PTSD: Relationship difficulties (DAS change score); Multiple incident index trauma

Figure 517. Couple intervention versus psycho-education sessions for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Family therapy

Figure 518. Family therapy versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report at 4-month follow-up (UCLA PTSD-RI; change score)

Figure 519. Family therapy versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 4-month follow-up (STAI: State; change score)

Child-parent psychotherapy

Figure 520. Child-parent psychotherapy (using play) versus case management and individual treatment (for parent only) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score)

Figure 521. Child-parent psychotherapy (using play) versus case management and individual treatment (for parent only) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission (number of people no longer meeting diagnostic criteria for PTSD)

Figure 522. Child-parent psychotherapy (using play) versus case management and individual treatment (for parent only) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Self-help with support

Figure 523. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (IES endpoint/IES-R/PDS/PCL-5 change score)

Figure 524. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 1-3 month follow-up (IES/PCL-5/PDS change score)

Figure 525. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 1-year follow-up (IES change score)

Figure 526. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score); Single incident trauma

Figure 527. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response (number of people showing clinically significant improvement, based on reliable change indices [RCI], on IES-R/PDS)

Figure 528. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission (number of people no longer above threshold on CAPS/<20 on PDS)

Figure 529. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Functional impairment (SDS change score); Single incident trauma

Figure 530. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life (QOLI/EUROHIS-QOL change score)

Figure 531. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Sleeping difficulties (SCL-90: Sleeping problems; change score)

Figure 532. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at endpoint (BAI/BSI: Anxiety/HSCL-25: Anxiety/SCL-90: Anxiety change score)

Figure 533. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 1-2 month follow-up (BAI/STAI State change score)

Figure 534. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 1-year follow-up (STAI State change score)

Figure 535. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at endpoint (BDI/BDI-II/BSI: Depression/HSCL-25: Depression/SCL-90: Depression change score)

Figure 536. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 1-2 month follow-up (BDI change score)

Figure 537. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 1-year follow-up (BDI change score)

Figure 538. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Alcohol use disorder symptoms (AUDIT change score); Single incident trauma

Figure 539. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Substance use disorder symptoms (TLFB: Number of days abstinent from alcohol in the last 90 days; change score); Unclear multiplicity of index trauma

Figure 540. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 541. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Sub-analysis by baseline severity: PTSD symptomatology self-rated at endpoint (IES endpoint/IES-R/PDS/PCL-5 change score)

Sub-analysis by specific intervention: Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 542. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (IES endpoint/IES-R/PDS/PCL-5 change score)

Figure 543. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score)

Figure 544. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by diagnostic status at baseline: Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 545. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (IES endpoint/IES-R/PDS/PCL-5 change score)

Figure 546. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score)

Figure 547. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by trauma type: Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 548. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (IES endpoint/IES-R/PDS/PCL-5 change score)

Figure 549. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score)

Figure 550. Self-help with support (± TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 551. Self-help with support versus self-help without support for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (PSS-I change score); single incident trauma

Figure 552. Self-help with support versus self-help without support for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response (number of people showing clinically significant improvement, based on reliable change indices [RCI], on PSS-I); single incident trauma

Figure 553. Self-help with support versus self-help without support for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms (FDAS change score); single incident trauma

Figure 554. Self-help with support versus self-help without support for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (CES-D change score); single incident trauma

Figure 555. Self-help with support versus self-help without support for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Self-help (without support)

Figure 556. Self-help (without support) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (IES-R/PCL-C/PDS change scores)

Figure 557. Self-help (without support) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission (number of people no longer meeting diagnostic criteria for PTSD or no longer above clinical threshold on scale); single incident trauma

Figure 558. Self-help (without support) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response at endpoint (number of people showing improvement of at least 10 points on PCL-C/clinically significant improvement, based on reliable change indices [RCI] on CAPS/≥50% improvement on PDS)

Figure 559. Self-help (without support) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response at 3-6 month follow-up (number of people showing clinically significant improvement, based on reliable change indices [RCI], on CAPS/≥50% improvement on PDS)

Figure 560. Self-help (without support) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Functional impairment at endpoint (SDS/B-IPF change score)

Figure 561. Self-help (without support) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Functional impairment at 6-month follow-up (SDS change score)

Figure 562. Self-help (without support) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at endpoint (BAI/STAI State/GAD-7 change score)

Figure 563. Self-help (without support) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 6-month follow-up (BAI change score)

Figure 564. Self-help (without support) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at endpoint (BDI-II/PHQ-8/PHQ-9 change score)

Figure 565. Self-help (without support) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 6-month follow-up (BDI-II change score)

Figure 566. Self-help (without support) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 567. Self-help (without support) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Sub-analysis by baseline severity: PTSD symptomatology self-rated (IES-R/PCL-C/PDS change scores)

Sub-analysis by specific intervention: Self-help (without support) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 568. Self-help (without support) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (IES-R/PCL-C/PDS change scores)

Figure 569. Self-help (without support) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by diagnostic status at baseline: Self-help (without support) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 570. Self-help (without support) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (IES-R/PCL-C/PDS change scores)

Figure 571. Self-help (without support) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by trauma type: Self-help (without support) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 572. Self-help (without support) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (IES-R/PCL-C/PDS change scores)

Figure 573. Self-help (without support) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 574. Self-help (without support) versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report at endpoint (PDS/IES change score)

Figure 575. Self-help (without support) versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report at 1-month follow-up (PDS change score)

Figure 576. Self-help (without support) versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (PSS-I change score)

Figure 577. Self-help (without support) versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission (number of people no longer meeting diagnostic criteria for PTSD)

Figure 578. Self-help (without support) versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at endpoint (CES-D/BDI-II change score)

Figure 579. Self-help (without support) versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 1-month follow-up (CES-D/BDI-II change score)

Figure 580. Self-help (without support) versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at endpoint (STAI State change score)

Figure 581. Self-help (without support) versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 1-month follow-up (STAI State change score)

Figure 582. Self-help (without support) versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by specific intervention: Self-help (without support) versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 583. Self-help (without support) versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report at endpoint (PDS/IES change score)

Figure 584. Self-help (without support) versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (PSS-I change score)

Figure 585. Self-help (without support) versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by diagnostic status at baseline: Self-help (without support) versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 586. Self-help (without support) versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report at endpoint (PDS/IES change score)

Figure 587. Self-help (without support) versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (PSS-I change score)

Figure 588. Self-help (without support) versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by trauma type: Self-help (without support) versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 589. Self-help (without support) versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report at endpoint (PDS/IES change score)

Figure 590. Self-help (without support) versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (PSS-I change score)

Figure 591. Self-help (without support) versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Meditation/Mindfulness-based stress reduction

Figure 592. Meditation/Mindfulness-based stress reduction (MBSR; ±TAU) versus TAU/attention-placebo/waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report at endpoint (PCL change score)

Figure 593. Meditation/Mindfulness-based stress reduction (MBSR; ±TAU) versus TAU/attention-placebo/waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report at 1-4 month follow-up (PCL change score)

Figure 594. Meditation/Mindfulness-based stress reduction (MBSR; ±TAU) versus TAU/attention-placebo/waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS/PSS-I change score)

Figure 595. Meditation/Mindfulness-based stress reduction (MBSR; ±TAU) versus TAU/attention-placebo/waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at 6-month follow-up (PSS-I change score)

Figure 596. Meditation/Mindfulness-based stress reduction (MBSR; ±TAU) versus TAU/attention-placebo/waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission (number of people scoring below clinical threshold on a scale)

Figure 597. Meditation/Mindfulness-based stress reduction (MBSR; ±TAU) versus TAU/attention-placebo/waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response at endpoint (number of people showing clinically significant improvement based on RCI [≥10/11 points on PCL-C])

Figure 598. Meditation/Mindfulness-based stress reduction (MBSR; ±TAU) versus TAU/attention-placebo/waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response at 4-month follow-up (number of people showing clinically significant improvement based on RCI [≥10 points on PCL-C])

Figure 599. Meditation/Mindfulness-based stress reduction (MBSR; ±TAU) versus TAU/attention-placebo/waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at endpoint (BSI Anxiety/HADS-A change score)

Figure 600. Meditation/Mindfulness-based stress reduction (MBSR; ±TAU) versus TAU/attention-placebo/waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 3-month follow-up (HADS-A change score)

Figure 601. Meditation/Mindfulness-based stress reduction (MBSR; ±TAU) versus TAU/attention-placebo/waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at endpoint (BDI/BSI Depression/HADS-D/PHQ-9 change score)

Figure 602. Meditation/Mindfulness-based stress reduction (MBSR; ±TAU) versus TAU/attention-placebo/waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 1-6 month follow-up (HADS-D/PHQ-9 change score)

Figure 603. Meditation/Mindfulness-based stress reduction (MBSR; ±TAU) versus TAU/attention-placebo/waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Sleeping difficulties (PSQI change score)

Figure 604. Meditation/Mindfulness-based stress reduction (MBSR; ±TAU) versus TAU/attention-placebo/waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Emotional and behavioural problems (STAXI-2 change score)

Figure 605. Meditation/Mindfulness-based stress reduction (MBSR; ±TAU) versus TAU/attention-placebo/waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life at endpoint (Q-LES-Q-SF/SF-8/12 Mental Component summary [MCS] change score)

Figure 606. Meditation/Mindfulness-based stress reduction (MBSR; ±TAU) versus TAU/attention-placebo/waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life at 4-month follow-up (SF-8 Mental Component summary [MCS] change score)

Figure 607. Meditation/Mindfulness-based stress reduction (MBSR; ±TAU) versus TAU/attention-placebo/waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by specific comparison: Meditation/Mindfulness-based stress reduction (MBSR) versus control for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 608. Meditation/Mindfulness-based stress reduction (MBSR) versus control for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report at endpoint (PCL change score)

Figure 609. Meditation/Mindfulness-based stress reduction (MBSR) versus control for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS change score)

Figure 610. Meditation/Mindfulness-based stress reduction (MBSR) versus control for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by diagnostic status at baseline: Meditation/Mindfulness-based stress reduction (MBSR) versus control for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 611. Meditation/Mindfulness-based stress reduction (MBSR) versus control for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report at endpoint (PCL change score)

Figure 612. Meditation/Mindfulness-based stress reduction (MBSR) versus control for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS change score)

Figure 613. Meditation/Mindfulness-based stress reduction (MBSR) versus control for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Sub-analysis by trauma type: Meditation/Mindfulness-based stress reduction (MBSR) versus control for delayed treatment (>3 months) of clinically important symptoms/PTSD

Figure 614. Meditation/Mindfulness-based stress reduction (MBSR) versus control for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report at endpoint (PCL change score)

Figure 615. Meditation/Mindfulness-based stress reduction (MBSR) versus control for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS/PSS-I change score)

Figure 616. Meditation/Mindfulness-based stress reduction (MBSR) versus control for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Meditation

Figure 617. Meditation (+TAU) versus relaxation (+TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report (PCL change score)

Figure 618. Meditation (+TAU) versus relaxation (+TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response (number of people showing clinically significant improvement based on RCI [≥11 points on PCL-C])

Figure 619. Meditation (+TAU) versus relaxation (+TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (BDI change score)

Figure 620. Meditation (+TAU) versus relaxation (+TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Sleeping difficulties (PSQI change score)

Mindfulness-based stress reduction

Figure 621. Mindfulness-based stress reduction (MBSR; +TAU) versus present-centred therapy (+TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (PCL change score); Multiple incident index trauma

Figure 622. Mindfulness-based stress reduction (MBSR; +TAU) versus present-centred therapy (+TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score); Multiple incident index trauma

Figure 623. Mindfulness-based stress reduction (MBSR; +TAU) versus present-centred therapy (+TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission (number of people no longer meeting diagnostic criteria for PTSD); Multiple incident index trauma

Figure 624. Mindfulness-based stress reduction (MBSR; +TAU) versus present-centred therapy (+TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response self-rated (number of people showing improvement of at least 10 points on PCL); Multiple incident index trauma

Figure 625. Mindfulness-based stress reduction (MBSR; +TAU) versus present-centred therapy (+TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response clinician-rated (number of people showing improvement of at least 10 points on CAPS); Multiple incident index trauma

Figure 626. Mindfulness-based stress reduction (MBSR; +TAU) versus present-centred therapy (+TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (PHQ-9 change score); Multiple incident index trauma

Figure 627. Mindfulness-based stress reduction (MBSR; +TAU) versus present-centred therapy (+TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life (WHO-QoL-BREF change score); Multiple incident index trauma

Figure 628. Mindfulness-based stress reduction (MBSR; +TAU) versus present-centred therapy (+TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Individual placement and support (IPS) supported employment

Figure 629. Individual placement and support (IPS) supported employment versus standard VA vocational rehabilitation programme (TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score)

Figure 630. Individual placement and support (IPS) supported employment versus standard VA vocational rehabilitation programme (TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (DTS change score)

Figure 631. Individual placement and support (IPS) supported employment versus standard VA vocational rehabilitation programme (TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Response (number of people rated as ‘much’ or ‘very much’ improved on CGI-I)

Figure 632. Individual placement and support (IPS) supported employment versus standard VA vocational rehabilitation programme (TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (QIDS change score)

Figure 633. Individual placement and support (IPS) supported employment versus standard VA vocational rehabilitation programme (TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Competitive employment (number of people who gained competitive employment)

Figure 634. Individual placement and support (IPS) supported employment versus standard VA vocational rehabilitation programme (TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Competitive employment (weeks competitively employed)

Figure 635. Individual placement and support (IPS) supported employment versus standard VA vocational rehabilitation programme (TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation

Practical support

Figure 636. Practical support versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (PDS change score)

Figure 637. Practical support versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (CES-D change score)

Psychoeducation

Figure 638. Psychoeducation (+TAU) versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 2-month follow-up (HTQ-IV change score)

Figure 639. Psychoeducation (+TAU) versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 2-month follow-up (HADS-A endpoint score)

Figure 640. Psychoeducation (+TAU) versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 2-month follow-up (HADS-D endpoint score)

Figure 641. Psychoeducation (+TAU) versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Quality of life at 2-month follow-up (SF-12 MCS)

Figure 642. Psychoeducation (+TAU) versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 643. Psychoeducation (±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (DTS change score)

Figure 644. Psychoeducation (±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 1-month follow-up (PCL change score)

Figure 645. Psychoeducation (±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 6-month follow-up (DTS change score)

Figure 646. Psychoeducation (±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 12-month follow-up (DTS change score)

Figure 647. Psychoeducation (±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 1-month follow-up (BSI Anxiety change score)

Figure 648. Psychoeducation (±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at endpoint (BDI-II change score)

Figure 649. Psychoeducation (±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 1-month follow-up (BSI Depression change score)

Figure 650. Psychoeducation (±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 6-month follow-up (BDI-II change score)

Figure 651. Psychoeducation (±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 12-month follow-up (BDI-II change score)

Figure 652. Psychoeducation (±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Suicide (BSS change score); Multiple incident index trauma

Figure 653. Psychoeducation (±TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Acupuncture

Figure 654. Acupuncture versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated (PSS-SR change score)

Figure 655. Acupuncture versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission (number of people scoring <16 on PSS-SR)

Figure 656. Acupuncture versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (HSCL-25: Depression, change score)

Figure 657. Acupuncture versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms (HSCL-25: Anxiety, change score)

Figure 658. Acupuncture versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Functional impairment (SDS change score)

Figure 659. Acupuncture versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Figure 660. Acupuncture versus paroxetine for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score); Single incident trauma

Figure 661. Acupuncture versus paroxetine for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms (HAM-A change score); Single incident trauma

Figure 662. Acupuncture versus paroxetine for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (HAMD change score); Single incident trauma

Figure 663. Acupuncture versus paroxetine for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Exercise

Figure 664. Exercise (+TAU) versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report (PCL change score)

Figure 665. Exercise (+TAU) versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score)

Figure 666. Exercise (+TAU) versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms (DASS: Anxiety; change score)

Figure 667. Exercise (+TAU) versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (DASS: Depression; change score)

Figure 668. Exercise (+TAU) versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Sleeping difficulties (PSQI change score)

Figure 669. Exercise (+TAU) versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Repetitive transcranial magnetic stimulation (rTMS)

Figure 670. Repetitive transcranial magnetic stimulation (rTMS) versus sham stimulation for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report (PCL change score)

Figure 671. Repetitive transcranial magnetic stimulation (rTMS) versus sham stimulation for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score)

Figure 672. Repetitive transcranial magnetic stimulation (rTMS) versus sham stimulation for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms (BDI change score)

Yoga

Figure 673. Yoga (± TAU) versus TAU/waitlist/attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report at endpoint (PCL/DTS change score)

Figure 674. Yoga (± TAU) versus TAU/waitlist/attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-report at 1-month follow-up (PCL change score)

Figure 675. Yoga (± TAU) versus TAU/waitlist/attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated (CAPS change score)

Figure 676. Yoga (± TAU) versus TAU/waitlist/attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission (number of people no longer meeting diagnostic criteria for PTSD)

Figure 677. Yoga (± TAU) versus TAU/waitlist/attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Dissociative symptoms (DES change score)

Figure 678. Yoga (± TAU) versus TAU/waitlist/attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at endpoint (DASS: Anxiety/STAI: State; change score)

Figure 679. Yoga (± TAU) versus TAU/waitlist/attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Anxiety symptoms at 1-month follow-up (STAI: State; change score)

Figure 680. Yoga (± TAU) versus TAU/waitlist/attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at endpoint (BDI-II/DASS Depression/CES-D change score)

Figure 681. Yoga (± TAU) versus TAU/waitlist/attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 1-month follow-up

Figure 682. Yoga (± TAU) versus TAU/waitlist/attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Symptoms of alcohol use disorder at endpoint (AUDIT change score)

Figure 683. Yoga (± TAU) versus TAU/waitlist/attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Symptoms of alcohol use disorder at 1-month follow-up (AUDIT change score)

Figure 684. Yoga (± TAU) versus TAU/waitlist/attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Symptoms of drug use disorder at endpoint (DUDIT change score)

Figure 685. Yoga (± TAU) versus TAU/waitlist/attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Symptoms of drug use disorder at 1-month follow-up (DUDIT change score)

Figure 686. Yoga (± TAU) versus TAU/waitlist/attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Sleeping difficulties (ISI change score)

Figure 687. Yoga (± TAU) versus TAU/waitlist/attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Bio-/neuro-feedback (±TAU) versus TAU or no treatment

Figure 688. Bio-/neuro-feedback (± TAU) versus TAU or no treatment for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at endpoint (PCL/DTS/IES-R change score)

Figure 689. Bio-/neuro-feedback (± TAU) versus TAU or no treatment for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology self-rated at 4-6 week follow-up (DTS/IES-R change score)

Figure 690. Bio-/neuro-feedback (± TAU) versus TAU or no treatment for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at endpoint (CAPS change score)

Figure 691. Bio-/neuro-feedback (± TAU) versus TAU or no treatment for delayed treatment (>3 months) of clinically important symptoms/PTSD: PTSD symptomatology clinician-rated at 1-month follow-up (CAPS change score)

Figure 692. Bio-/neuro-feedback (± TAU) versus TAU or no treatment for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at endpoint (number of people no longer meeting diagnostic criteria)

Figure 693. Bio-/neuro-feedback (± TAU) versus TAU or no treatment for delayed treatment (>3 months) of clinically important symptoms/PTSD: Remission at 1-month follow-up (number of people no longer meeting diagnostic criteria)

Figure 694. Bio-/neuro-feedback (± TAU) versus TAU or no treatment for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at endpoint (BDI change score)

Figure 695. Bio-/neuro-feedback (± TAU) versus TAU or no treatment for delayed treatment (>3 months) of clinically important symptoms/PTSD: Depression symptoms at 6-week follow-up (BDI change score)

Figure 696. Bio-/neuro-feedback (± TAU) versus TAU or no treatment for delayed treatment (>3 months) of clinically important symptoms/PTSD: Discontinuation (loss to follow-up)

Trauma-focused CBT

Table 115. Clinical evidence profile: Trauma-focused CBT versus waitlist or no treatment for early treatment (1-3 months) of clinically important symptoms/PTSD

Table 116. Clinical evidence profile: Trauma-focused CBT versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 117. Clinical evidence profile: Trauma-focused CBT + medication/TAU versus medication/TAU-only (or + attention-placebo) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 118. Clinical evidence profile: Trauma-focused CBT (+/- TAU) versus eye movement desensitisation and reprocessing (EMDR; +/- TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 119. Clinical evidence profile: Trauma-focused CBT (+/- TAU) versus non-trauma-focused CBT (+/- TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 120. Clinical evidence profile: Trauma-focused CBT (+/- TAU) versus counselling (+/- TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 121. Clinical evidence profile: Trauma-focused CBT (+/- TAU) versus present-centred therapy (+/- TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 122. Clinical evidence profile: Trauma-focused CBT (+ TAU) versus metacognitive (+ TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 123. Clinical evidence profile: Trauma-focused CBT versus interpersonal psychotherapy (IPT) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 124. Clinical evidence profile: Trauma-focused CBT (+ TAU) versus psychodynamic therapy (+ TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 125. Clinical evidence profile: Trauma-focused CBT (+/- TAU) versus self-help (without support; +/- TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 126. Clinical evidence profile: Trauma-focused CBT versus self-help with support for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 127. Clinical evidence profile: Trauma-focused CBT (+ TAU) versus hypnotherapy (+ TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 128. Clinical evidence profile: Trauma-focused CBT versus psychoeducational session for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 129. Clinical evidence profile: Trauma-focused CBT (+/- TAU) versus relaxation (+/- TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 130. Clinical evidence profile: Trauma-focused CBT versus acupuncture for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 131. Clinical evidence profile: Trauma-focused CBT versus SSRIs for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 132. Clinical evidence profile: Trauma-focused CBT + SSRIs versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD

Non-trauma-focused CBT

Table 133. Clinical evidence profile: Non-trauma-focused CBT (+/- TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 134. Clinical evidence profile: Non-trauma-focused CBT (+/- TAU) versus attention-placebo (+/- TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 135. Clinical evidence profile: Non-trauma-focused CBT (+ TAU) versus psychoeducational group (+ TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 136. Clinical evidence profile: Non-trauma-focused CBT versus counselling for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 137. Clinical evidence profile: Non-trauma-focused CBT versus present-centred therapy for delayed treatment (>3 months) of clinically important symptoms/PTSD

Present-centred therapy

Table 138. Clinical evidence profile: Present-centred therapy (+ TAU) versus TAU for early treatment (1-3 months) of clinically important symptoms/PTSD

Table 139. Clinical evidence profile: Present-centred therapy versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD

Cognitive therapies

Table 140. Clinical evidence profile: Metacognitive therapy (+/- TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 141. Clinical evidence profile: Reconsolidation of traumatic memories (RTM) intervention + TAU versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD

Behavioural therapies

Table 142. Clinical evidence profile: Single-session behavioural therapy versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD

Problem solving

Table 143. Clinical evidence profile: Problem solving versus supportive counselling for early treatment (1-3 months) of clinically important symptoms/PTSD

Eye movement desensitisation and reprocessing (EMDR)

Table 144. Clinical evidence profile: Eye movement desensitisation and reprocessing (EMDR) versus supportive counselling for early treatment (1-3 months) of clinically important symptoms/PTSD

Table 145. Clinical evidence profile: Eye movement desensitisation and reprocessing (EMDR; +/- TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 146. Clinical evidence profile: Eye movement desensitisation and reprocessing (EMDR) versus pill placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 147. Clinical evidence profile: Eye movement desensitisation and reprocessing (EMDR) versus supportive counselling for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 148. Clinical evidence profile: Eye movement desensitisation and reprocessing (EMDR) versus non-trauma-focused CBT for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 149. Clinical evidence profile: Eye movement desensitisation and reprocessing (EMDR) versus ‘other active psych intervention’ for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 150. Clinical evidence profile: Eye movement desensitisation and reprocessing (EMDR; +/- TAU) versus relaxation (+/- TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 151. Clinical evidence profile: Eye movement desensitisation and reprocessing (EMDR) versus combined somatic and cognitive therapies for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 152. Clinical evidence profile: Eye movement desensitisation and reprocessing (EMDR) versus fluoxetine for delayed treatment (>3 months) of clinically important symptoms/PTSD

Hypnotherapy

Table 153. Clinical evidence profile: Hypnotherapy + TAU versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 154. Clinical evidence profile: Hypnotherapy followed by trauma-focused CBT versus symptom monitoring followed by trauma-focused CBT for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 155. Clinical evidence profile: Hypnotherapy (+ TAU) versus zolpidem (+ TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Interpersonal psychotherapy (IPT)

Table 156. Clinical evidence profile: Interpersonal psychotherapy (IPT) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD

Psychodynamic therapies

Table 157. Clinical evidence profile: Psychodynamic therapy (+/- TAU) versus waitlist (+/- TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 158. Clinical evidence profile: Interpersonal psychotherapy (IPT) versus relaxation for delayed treatment (>3 months) of clinically important symptoms/PTSD

Counselling

Table 159. Clinical evidence profile: Counselling (+/- TAU) versus TAU or waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD

Combined somatic and cognitive therapies

Table 160. Clinical evidence profile: Combined somatic and cognitive therapies (+/- TAU) versus waitlist (+/- TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Somatic experiencing

Table 161. Clinical evidence profile: Somatic experiencing + TAU versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD

Resilience-oriented treatment

Table 162. Clinical evidence profile: Resilience-oriented treatment versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD

Attention bias modification

Table 163. Clinical evidence profile: Attention bias modification versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD

Couple interventions

Table 164. Clinical evidence profile: Couple intervention versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 165. Clinical evidence profile: Couple intervention versus psychoeducation sessions for delayed treatment (>3 months) of clinically important symptoms/PTSD

Parent training/family interventions

Table 166. Clinical evidence profile: Family therapy versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 167. Clinical evidence profile: Child-parent psychotherapy (using play) versus case management and individual treatment (for parent-only) for delayed treatment (>3 months) of clinically important symptoms/PTSD (in parent)

Self-help with support

Table 168. Clinical evidence profile: Self-help with support (+/- TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 169. Clinical evidence profile: Self-help with support versus self-help without support for delayed treatment (>3 months) of clinically important symptoms/PTSD

Self-help (without support)

Table 170. Clinical evidence profile: Self-help (without support) versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 171. Clinical evidence profile: Self-help (without support) versus attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD

Meditation/Mindfulness-based stress reduction (MBSR)

Table 172. Clinical evidence profile: Meditation/Mindfulness-based stress reduction (MBSR; +/- TAU) versus TAU/attention-placebo/waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 173. Clinical evidence profile: Meditation (+ TAU) versus relaxation (+ TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 174. Clinical evidence profile: Mindfulness-based stress reduction (MBSR; + TAU) versus present-centred therapy (+ TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Supported employment

Table 175. Clinical evidence profile: Individual placement and support (IPS) supported employment versus standard VA vocational rehabilitation programme (TAU) for delayed treatment (>3 months) of clinically important symptoms/PTSD

Practical support

Table 176. Clinical evidence profile: Practical support versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD

Psychoeducation

Table 177. Clinical evidence profile: Psychoeducation (+ TAU) versus TAU for early treatment (1-3 months) of clinically important symptoms/PTSD

Table 178. Clinical evidence profile: Psychoeducation (+/- TAU) versus waitlist or TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD

Acupuncture

Table 179. Clinical evidence profile: Acupuncture versus waitlist for delayed treatment (>3 months) of clinically important symptoms/PTSD

Table 180. Clinical evidence profile: Acupuncture versus paroxetine for delayed treatment (>3 months) of clinically important symptoms/PTSD

Exercise

Table 181. Clinical evidence profile: Exercise (+ TAU) versus TAU for delayed treatment (>3 months) of clinically important symptoms/PTSD

Repetitive transcranial magnetic stimulation (rTMS)

Table 182. Clinical evidence profile: Repetitive transcranial magnetic stimulation (rTMS) versus sham stimulation for delayed treatment (>3 months) of clinically important symptoms/PTSD

Yoga

Table 183. Clinical evidence profile: Yoga (+/- TAU) versus TAU/waitlist/attention-placebo for delayed treatment (>3 months) of clinically important symptoms/PTSD

Bio-/neuro-feedback

Table 184. Clinical evidence profile: Bio-/neuro-feedback (+/- TAU) versus TAU or no treatment for delayed treatment (>3 months) of clinically important symptoms/PTSD

Population

The study population of the economic model comprised adults with PTSD, who initiate treatment for PTSD in a community setting, although they may receive care in other settings over the time horizon of the analysis. This was decided because the majority of adults with PTSD initiate treatment for PTSD in a community setting in UK routine practice.

No distinction was made between adults with single trauma and those with multiple traumas as there was no adequate evidence to demonstrate that the effectiveness of interventions was affected by this factor.

The starting age of the cohorts considered in the economic model was set at 39 years, to reflect the mean age of adults with PTSD presenting to healthcare services. The estimate of 39 years was based on a study of all consecutive patients who were referred for assessment for possible PTSD between April 2001 and August 2008 in a UK NHS outpatient clinic and were subsequently offered cognitive therapy for PTSD (Ehlers et al., 2013).

The percentage of women in each cohort at the start of the model was estimated to be 51.6%, calculated using the proportion of women in the general population aged 39 years (i.e. the average age of population initiating treatment) obtained from general statistics for the UK population (Office for National Statistics, 2017b), and data on the percentage of people screened positive for PTSD by age and sex reported in the most recent adult psychiatric morbidity household survey conducted in England (McManus et al., 2016).

Determining the starting age and gender mix of the cohorts was necessary in order to estimate mortality risks in the model; moreover, the gender mix was used at the estimation of QALYs, as the base-case economic analysis utilised gender-specific utility data, as described later.

Interventions assessed

The range of interventions assessed in the economic analysis was determined by the availability of relevant clinical data included in the guideline systematic review of psychological interventions for the treatment of adults with clinically important PTSD symptoms. Network meta-analysis (NMA) was employed for synthesis of the available efficacy data. Details of the NMA undertaken to inform the economic analysis are provided in the ‘Efficacy data and methods of evidence synthesis’ section. The guideline economic analysis assessed psychological, pharmacological and combined psychological and pharmacological interventions that were connected to the network of evidence and were thus possible to include in the NMA. Hypnotherapy and psychosocial interventions such as meditation, mindfulness-based stress reduction, supported employment, peer support and practical support, as well as physical interventions such as exercise, yoga, acupuncture, bio-neuro-feedback and repetitive transcranial magnetic stimulation (r-TMS) were not included in the analysis as they were not part of the decision problem. Relaxation was included as a control intervention that provided additional indirect comparisons across interventions of interest.

Based on the advice of the committee, only effective interventions that had been tested on at least 50 people across the RCTs included in the NMAs assessing efficacy at treatment endpoint were considered in the economic analysis, as this was deemed as the minimum evidence that would be adequate to support a practice recommendation.

Interventions that belonged to the trauma-focused cognitive behavioural therapy (TF-CBT) class were not considered separately according to their type, as the description of the type of TF-CBT was not always clear in the publications. However, based on reported resource use in each RCT included in the NMA, TF-CBT interventions were categorised according to their mode of delivery in individual, group and mixed (where the intervention was delivered by a combination of individual and group sessions). Each of these categories was further subdivided, as relevant, to those comprising fewer than 8 sessions, 8-12 sessions, and more than 12 sessions, and were considered separately in the NMA and the economic analysis, to reflect the different intervention costs and, potentially, different efficacy associated with each sub-category.

Based on the available evidence, the following interventions were considered in the economic analysis of interventions for the treatment of adults with PTSD:

• Psychoeducation
• Counselling
• TF-CBT individual <8 sessions
• TF-CBT individual 8-12 sessions
• TF-CBT individual >12 sessions
• TF-CBT group 8-12 sessions
• non-TF-CBT
• Eye Movement Desensitisation Reprocessing (EMDR)
• Present-centred therapy
• Interpersonal psychotherapy
• Combined somatic and cognitive therapies
• Self-help with support
• Self-help without support
• Selective serotonin reuptake inhibitors (SSRIs)
• TF-CBT individual 8-12 sessions + SSRIs
• No treatment, reflected in the waitlist arms of RCTs included in the guideline systematic review and NMA.

Model structure

A hybrid decision-analytic model consisting of a decision-tree followed by a three-state Markov model was constructed using Microsoft Office Excel 2013. The model estimated the total costs and benefits associated with provision of effective treatment options in adults with PTSD. The structure of the model, which aimed to simulate the course of PTSD and relevant clinical practice in the UK, was also driven by the availability of clinical data.

According to the model structure, hypothetical cohorts of adults with PTSD were initiated on each of the treatment options assessed, including no treatment. The duration of a full course of initial treatment was 12 weeks for drugs and varied between 6 and 16 weeks for non-pharmacological interventions. The duration of combined interventions was determined by the component with the longest duration. For modelling purposes relating to estimation of QALYs, the duration of a full course of treatment was assumed to be 3 months (12 weeks), without this assumption affecting resource use associated with each intervention. Following a course of treatment, people in each cohort either remitted (that is, they did not meet criteria for a PTSD diagnosis) or did not remit. Those initiated on pharmacological or combined treatment were given a further 3 months of maintenance pharmacological therapy if they had remitted. In the 3 months of follow-up after treatment completion, people who remitted (‘no PTSD’) could remain in remission, relapse to a PTSD state or die. Those who did not remit, could remain in the PTSD state, remit (and move to a ‘no PTSD’ state) or die. The two distinct periods in the decision-tree (full course of treatment and 3-month follow-up) were informed by the results of respective NMAs (although the 3-month follow-up period was informed by the results of the NMA only in a sensitivity analysis, as discussed later). The length of the follow-up period immediately post-treatment was set at 3 months as this was the period for which most RCT follow-up data were available across interventions.

After that point, people in each cohort, both those who remitted and those who did not remit, were entered into the Markov component of the economic model, in either the ‘PTSD’ or the ‘no PTSD’ health states, depending on their state at the end of the decision-tree. In each cycle of the Markov model, they could remain in the same health state or move between the two states of ‘PTSD’ and ‘no PTSD’ or move to the death state (absorbing state). The Markov model was run in 3-month cycles, for consistency with the duration of the two periods of the decision-tree, that is, a full course of treatment (which lasted, on average, 3 months) and another 3-month follow-up period (the length of which was determined by data availability). A half-cycle correction was applied. Due to lack of long-term comparative clinical data, transitions between the ‘PTSD’ and ‘no PTSD’ health states in the Markov component of the model were assumed to be independent of the intervention received at the decision-tree part of the model. The transition probability to the death state depended on the PTSD status of each person in the population.

The time horizon of the analysis was 3 years, consisting of the 6 months of the decision tree and another 2.5 years (10 x 3-month cycles) in the Markov component of the economic model. This time frame was considered to be long enough to capture longer-term costs and effects of treatment, without significant extrapolation over the course of PTSD.

Death was not considered during provision of interventions, as no relevant differential mortality data are available. However, the presence of PTSD is associated with an increase in mortality (Ahmadi et al., 2011). For this reason, death was considered at follow-up, both in the first 3 months of follow-up (decision-tree) and in the Markov component of the model.

A proportion of adults who received pharmacological or combined treatment were assumed to experience side effects from medication which resulted in a reduction in their HRQoL over the period they received pharmacological treatment (i.e. 3 or 6 months) and incurred extra costs for their management, which comprised GP visits and pharmacological treatment.

The structure of the economic model for interventions for treatment of PTSD in adults is shown in Figure 697.

Figure 697. Schematic diagram of the economic model structure: interventions for the treatment of PTSD in adults

Costs and outcomes considered in the analysis

The economic analysis adopted the perspective of the NHS and personal social services (PSS), as recommended by NICE (NICE, 2014). Costs consisted of intervention costs (healthcare professional time, drug acquisition and equipment/infrastructure required for self-help interventions), as well as other costs incurred by adults PTSD who did not remit following treatment or who experienced a relapse following remission and by those who remitted, including primary, community and secondary health care and personal social services. Costs of management of common side effects from pharmacological treatment in people receiving pharmacological or combined treatment were also considered in the analysis. The cost year was 2017.

The measure of outcome was the Quality Adjusted Life Year (QALY), which incorporated utilities associated with the health states of PTSD and no PTSD, as well as utility decrements due to common side effects associated with pharmacological treatment.

Efficacy data and methods of evidence synthesis

Other clinical input parameters

Other clinical input parameters included

• the baseline (no treatment) probability of remission, which was applied as the baseline in the decision-tree and also across all treatment options in the Markov part of the model
• the risk of relapse following remission, which was independent of the intervention received at the start of the model.
• the risk of development of side effects from SSRIs
• mortality associated with PTSD and no PTSD health states

Utility data and estimation of quality adjusted life years (QALYs)

In order to express outcomes in the form of QALYs, the health states of the economic model (remission, response not reaching remission, no response or relapse) need to be linked to appropriate utility scores. Utility scores represent the HRQoL associated with specific health states on a scale from 0 (death) to 1 (perfect health); they are estimated using preference-based measures that capture people’s preferences on the HRQoL experienced in the health states under consideration.

The systematic review of utility data on PTSD-related heath states identified 2 studies that reported utility data corresponding to PTSD-related health states in adults that met inclusion criteria (Freed et al., 2009; Haagsma et al., 2012; Mihalopoulos et al., 2015). There were 4 studies that were excluded after obtaining full text, and these are reported in Appendix K, together with reasons for exclusion.

Freed and colleagues (2009) reported utility scores derived from a random sample of 808 veterans (79% male; 12% met criteria for PTSD) who attended four primary care clinics in the US and who completed the PTSD Checklist (PCL), the Clinician-Administered PTSD Scale, the Mini-International Neuropsychiatric Interview and the Medical Outcomes Survey Short Form–36 (SF-36). SF-36 ratings were used to estimate utility scores after conversion to SF-6D and use of the UK adult general population algorithm, which was derived using the standard gamble (SG) technique (Brazier et al., 2002). The authors reported utility data for veterans with PTSD (n=711) and veterans without PTSD (n=97), before and after adjustment for confounders such as gender, employment status, presence of disability as well as mental and physical health comorbidities (chronic obstructive pulmonary disorder, mood disorder, anxiety disorder, substance use disorder).

Haagsma and colleagues (2012) reported mean EQ-5D utility scores derived from 1,781 injury patients aged 15 years and older who attended the Emergency Department of the Dutch Injury Surveillance System. The sample consisted of victims of traffic, home and leisure, occupational and sport accidents. Injuries varied from minor to severe injury, single or multiple injury, requiring hospitalisation or not. The Impact of Event Scale (IES) was used to assess symptoms of post-traumatic stress indicative of PTSD. The UK EQ-5D tariff, formed using the time trade-off (TTO) technique, was used (Dolan, 1997). The authors reported utility scores from 73 injury patients with PTSD symptoms (IES-score ≥ 35) and 1,708 patients without PTSD symptoms (IES < 35).

Mihalopoulos and colleagues (2015) reported utility data from adults participating in the National Survey of Mental Health and Wellbeing conducted in Australia in 1997. People were categorised into those with or without a current diagnosis of PTSD (according to DSM-IV criteria) and whether or not they had been receiving evidence-based treatments over the last 12 months. HRQoL was measured using the generic Assessment of Quality of Life (AQoL) measure, which was subsequently converted to the AQoL-4D preference-based measure. The scale includes 12 items (personal care, household tasks, ability to move around the house and community, personal relationships, relationships with other people, relationships with family, vision, hearing, communication with others, sleeping habits, feelings in general, and level of pain or discomfort) rated using 4 levels. Preferences for AQoL-4D health states have been elicited from a sample of the Australian general population using time trade-off (TTO). The study provided data gender-specific data for people who were PTSD-free following evidence-based treatment [i.e. people with a diagnosis of PTSD within the last 12 months but without a current (30-day) diagnosis, who had received evidence-based treatment over the last 12 months] and people with PTSD [i.e. people with a diagnosis of PTSD within the last 12 months including the last 30-days who had not been receiving evidence-based treatment over the last 12 months].

An overview of the study characteristics, the methods used to define health states, and the health-state utility values reported by each of the two studies is provided in Table 197.

Table 197. Summary of available health-state utility data for PTSD in adults

According to NICE guidance on the selection of utility values for use in cost-utility analysis (NICE, 2013), the measurement of changes in HRQoL should be reported directly from people with the condition examined, or, if this is not possible, by their carers, and the valuation of health states should be based on public preferences elicited using a choice-based method, such as the TTO or SG, in a representative sample of the UK population. NICE recommends the EQ-5D (Brooks, 1996; Dolan, 1997) as the preferred measure of HRQoL in adults for use in cost-utility analysis.

Of the reported data, those from Haagsma and colleagues (2012) are based on EQ-5D ratings and UK population preferences and thus directly meet NICE criteria. However, the committee noted that they reflect HRQoL of people with injuries, so utility values may have been greatly affected by physical symptoms, which are likely to be more severe in people with PTSD. Moreover, utility values in people who have never had PTSD are expected to be higher than those in people who have remitted from PTSD, who are the focus of the economic analysis.

The data from Freed and colleagues (2009) were derived from US veterans and were based on values elicited from the UK population using SG, thus partially meeting NICE criteria. The committee noted the narrow difference between PTSD and no-PTSD health states after adjustment for confounders and the high prevalence of comorbidities characterising the study population (veterans). They also noted that the utility values in people who have never had PTSD are expected to be higher than those in people who have remitted from PTSD, who are the focus of the economic analysis.

The data from Mihalopoulos and colleagues (2015) were derived from Australian adults who had experienced PTSD. The utility values express Australian population’s preferences but meet NICE criteria regarding the method of preference elicitation. The committee noted that the utility data correspond directly to the model health states of interest, i.e. people with PTSD and people who remitted from PTSD. The committee noted that the difference between the PTSD and no-PTSD health state values were the narrowest among the 3 datasets (compared with unadjusted data from Freed and colleagues) but expressed the view that they probably reflected a conservative but realistic estimate of the difference in the utility between people experiencing PTSD and those who have remitted.

Based on the above considerations, the committee selected the data from Mihalopoulos and colleagues to inform the guideline economic analysis of interventions for adults with PTSD. The analysis utilised separately the utility data for men and women. Gender-specific data, as reported in the study, were used. The adjusted data from Freed and colleagues, which indicated a narrower utility benefit following remission from PTSD, were used in sensitivity analysis; the same utility values for each health state were used for both men and women, as the paper did not provide gender-specific utility data.

Changes in utility between the states of ‘PTSD’ and ‘no PTSD’ were assumed to occur linearly over the time period of the change. When running the probabilistic analysis, the utility value of the ‘no PTSD’ health state was not allowed to become lower than that of the ‘PTSD’ health state. In iterations where the utility of the ‘no PTSD’ health state was lower than the utility of the ‘PTSD’ health state, the former was forced to equal the latter.

Side effects from SSRIs are expected to have a negative impact on people’s HRQoL. Sullivan and colleagues (2004) applied regression analysis on EQ-5D data (UK tariffs) obtained from participants in the 2000 national US Medical Expenditure Panel Survey to derive age-adjusted utility values for health states associated with depression and with side effects of antidepressants. Health states were defined based on descriptions in the International Classification of Diseases (9th Edition) [ICD-9] and the Clinical Classification Categories (CCC) [clinically homogenous groupings of ICD-9 codes derived by the Agency for Healthcare Research and Quality]. The authors reported a mean utility decrement due to side effects from antidepressants ranging from -0.044 (diarrhoea) to -0.129 (excitation, insomnia and anxiety), with a mean decrement of -0.087; the mean utility of treated depression was 0.848. These data translate into a 10.3% reduction in utility due to side effects of antidepressants, which was applied to people who experienced side effects from SSRIs in the economic model, over the period they received SSRI treatment.

Intervention resource use and costs

Intervention costs were estimated by combining resource use associated with each intervention with appropriate unit costs.

Costs associated with the PTSD and ‘no PTSD’ health states

The costs of the PTSD and PTSD-free states in the Markov component of the economic model were estimated using health and personal social service usage data from the Adult Psychiatry Morbidity Survey conducted in England in 2014 (McManus et al., 2016), supplemented with resource use data from other national sources and the committee’s expert opinion. The survey reported the percentage of adults with PTSD and adults without PTSD that were currently receiving pharmacological or psychological treatment and/or had been using a range of health and personal social services over the last quarter or year for a mental or emotional problem. These services included inpatient hospital stays, outpatient visits, and contacts with GPs, psychiatrists, psychologists, community psychiatric nurses, community learning disability nurses, other nursing services, social workers, self-help and support groups, home help or home care, outreach or family support workers and community day-care centres. However, the exact resource use of each service (e.g. number of psychological treatment sessions, number of outpatient visits) was not reported as relevant information was not collected in the survey. The reported percentages of survey respondents using the services over a period of time were extrapolated, where needed, in order to estimate the percentage of adults with and without PTSD using each service on an annual basis. The mean number of sessions for adults receiving psychological treatment was taken from an annual report on the use of IAPT services (NHS Digital, Community and Mental Health team 2016). The average length of stay for adults receiving inpatient care was taken from national hospital episode statistics (NHS Digital, 2017). Furthermore, the committee made estimates on the number of visits and the time spent on each visit where relevant, in order to provide a total resource use estimate for each type of service. Information on the number of GP visits for adults with mental health problems was sought from published UK evidence (Kontopantelis et al., 2015). The resource use estimates were then combined with appropriate unit costs taken from national sources (Curtis and Burns, 2017, NHS Improvement, 2017) in order to estimate an overall annual health and personal social service cost incurred by adults with PTSD and by those without PTSD. Unit costs included wages/salary, salary on costs, capital and other overheads, as well as qualification costs.

Details on the data and the committee’s estimates used to estimate the annual costs associated with the PTSD and no PTSD health states are provided in Table 202.

Table 202. Annual health and personal social service costs incurred by adults with PTSD and adults without PTSD (2017 prices)

Using the annual cost figures, 3-monthly health and personal social care costs were then estimated for the two states of ‘PTSD’ (£293) and ‘no PTSD’ (£27) of the economic model. People moving between the two health states of PTSD and no PTSD in every cycle of the model were assumed to incur 50% of the PTSD cost and 50% of the no PTSD cost within the cycle they transitioned between the two health states.

Health and personal social service costs were assumed to be the same across all arms of the economic model during the period of initial (3-month) treatment and therefore were excluded from further consideration.

Because the estimated health state-related costs were based to a large degree on the committee’s expert opinion, a sensitivity analysis was conducted, in which costs associated with the PTSD state were varied by ±50%, to explore the impact of the health state cost estimates on the results of the economic analysis.

All costs were expressed in 2017 prices, uplifted, where necessary, using the Hospital and Community Health Services Pay and Prices Index (Curtis & Burns, 2017). Costs and QALYs were discounted at an annual rate of 3.5%, according to NICE guidance (NICE, 2014).

Cost of management of side effects from the pharmacological component of combined treatment

People who experienced common side effects were assumed to have one extra GP contact every 3 months costing £37 (Curtis & Burns, 2017) and to incur a cost of £3 over the same period for medication relating to the management of common side effects.

Discounting

Costs and benefits were discounted at an annual rate of 3.5% as recommended by NICE (2014).

Handling uncertainty

Model input parameters were synthesised in a probabilistic analysis. This means that the input parameters were assigned probabilistic distributions (rather than being expressed as point estimates); this approach allowed more comprehensive consideration of the uncertainty characterising the input parameters and captured the non-linearity characterising the economic model structure. Subsequently, 10,000 iterations were performed, each drawing random values out of the distributions fitted onto the model input parameters. Results (mean costs and QALYs for each intervention) were averaged across the 10,000 iterations. This exercise provides more accurate estimates than those derived from a deterministic analysis (which utilises the mean value of each input parameter ignoring any uncertainty around the mean), by capturing the non-linearity characterising the economic model structure (Briggs et al., 2006).

The distributions of the log-odds ratios of relative effects of all treatments versus no treatment were obtained from the respective NMAs, defined directly from values recorded in each of the 10,000 iterations used after thinning the 300,000 iterations performed in WinBUGS or OpenBUGS, as relevant.

Beta distribution was assigned to the following parameters: the baseline probability of remission (probability of remission of no treatment between 0-6 months and probability of remission across all interventions from 6 months onwards); the probability of relapse; the proportion of people experiencing side effects from SSRIs; and the utility values (including the disutility due to side effects from SSRIs), after applying the method of moments on data reported in the relevant literature.

The hazard ratio of death of people with PTSD versus people without PTSD was assigned a log-normal distribution.

Uncertainty in psychological intervention costs was taken into account by assigning probability distributions to the number of individually delivered psychological therapy sessions, based on intervention completion data and data on mean number of sessions reported in the RCTs that informed the economic analysis. The number of therapist sessions per person attending group psychological interventions was not assigned a probability distribution because the number of group sessions remains the same, whether a participant attends the full course of treatment or a lower number of sessions. The therapist time spent on self-help programmes was assigned a normal distribution. The unit cost of therapists delivering psychological interventions, as well as the unit cost of GPs, were also assigned a normal distribution.

NHS/PSS costs associated with the ‘PTSD’ and ‘no PTSD’ health states were assigned a gamma distribution.

Table 203 reports the mean values of all input parameters utilised in the economic model and provides details on the types of distributions assigned to each input parameter and the methods employed to define their range.

Table 203. Input parameters (deterministic values and probability distributions) that informed the economic model of interventions for the treatment of PTSD in adults

A number of different analyses were undertaken, using the 2 sets of available efficacy data (changes in PTSD symptom scores and dichotomous remission) and 2 alternative assumptions on the efficacy of interventions at the 3-month follow-up (based on the respective continuous change score data). Consequently, 3 separate probabilistic analyses were undertaken:

• Analysis A: efficacy data at treatment endpoint were derived from the NMA of continuous data (changes in PTSD symptom scores), transformed to log-odds ratios of remission; the probability of remission of all active interventions at 3-6 months was conservatively assumed to equal that of no treatment. This analysis formed the base-case economic analysis.
• Analysis B: efficacy data at treatment endpoint were derived from the NMA of continuous data (changes in PTSD symptom scores), transformed to log-odds ratios of remission; the relative effect of active interventions versus no treatment at 3-6 months was derived from the NMA of changes in PTSD symptom scores between baseline and 1-4 month follow-up, also transformed to log-odds ratios of remission,.
• Analysis C: efficacy data at treatment endpoint were derived from the NMA of dichotomous remission data; the probability of remission of all active interventions at 3-6 months was assumed to equal that of no treatment, as dichotomous remission follow-up data were very limited.

A number of deterministic one-way sensitivity analyses were also employed to explore the impact of alternative hypotheses on the results. The following scenarios were explored:

• The annual risk of relapse was varied between 0.05 and 0.20 (base-case value was 0.10)
• Use of alternative utility values of 0.61 and 0.64 for the PTSD and no PTSD health states, respectively, reported in Freed and colleagues (2005)
• The PTSD health state cost was changed by ± 50%.

Presentation of the results

Results of the economic analysis are presented as follows:

Results are reported separately for each cohort examined in the economic model. In each analysis, mean total costs and QALYs are presented for each intervention, averaged across 10,000 iterations of the model. An incremental analysis is provided for each cohort, in table format, where all options have been listed from the most to the least effective (in terms of QALYs gained). Options that are dominated by absolute dominance (that is, they are less effective and more costly than one or more other options) or by extended dominance (that is, they are less effective and more costly than a linear combination of two alternative options) are excluded from further analysis. Subsequently, incremental cost-effectiveness ratios (ICERs) are calculated for all pairs of consecutive options remaining in analysis.

ICERs are calculated by the following formula:

In addition to ICERs, the mean net monetary benefit (NMB) of each intervention is presented. This is defined by the following formula:

Incremental mean costs and effects (QALYs) of each intervention versus no treatment are also presented in the form of cost effectiveness planes.

The probability of each intervention being the most cost-effective option at the NICE lower cost effectiveness threshold of £20,000/QALY is provided, calculated as the proportion of iterations (out of the 10,000 iterations run) in which the intervention has had the highest NMB among all interventions considered in the analysis.

The mean ranking in terms of cost effectiveness is also reported for each intervention (out of the 10,000 iterations run), where a rank of 1 is best.

The probabilities of each intervention being cost-effective at various cost effectiveness thresholds are illustrated in cost-effectiveness acceptability curves (CEACs). Finally, the cost-effectiveness acceptability frontiers (CEAFs) are also plotted; these show the treatment option with the highest mean NMB over different cost effectiveness thresholds, and the probability that the option with the highest NMB is the most cost-effective among those assessed (Fenwick et al., 2001).

Validation of the economic model

The economic model (including the conceptual model and the identification and selection of input parameters) was developed by the health economist in collaboration with a health economics sub-group formed by members of the committee. As part of the model validation, all inputs and model formulae were systematically checked; the model was tested for logical consistency by setting input parameters to null and extreme values and examining whether results changed in the expected direction. The base-case results and results of sensitivity analyses were discussed with the committee to confirm their plausibility.

Analysis A (base-case): efficacy at treatment endpoint based on NMA of continuous data (changes in PTSD symptom scores); no beneficial effect beyond treatment endpoint

The results of the base-case economic analysis are provided in Table 204. This table provides mean QALYs and mean total costs for each intervention assessed in the economic analysis, as well as the results of incremental analysis, the mean NMB of each intervention, and its mean ranking by cost effectiveness (where a rank of 1 is best). Interventions have been ordered from the most to the least effective in terms of number of QALYs gained. According to the results, TF-CBT individual < 8 sessions was the most clinically and cost-effective intervention, however, its probability of being the most cost-effective option was only 0.28. Psychoeducation was the second most cost-effective intervention, followed by EMDR, combined somatic and cognitive therapies, self-help with support, SSRI, self-help without support, TF-CBT individual 8-12 sessions, IPT, non-TF-CBT, present-centred therapy, TF-CBT group 8-12 sessions, combined TF-CBT individual 8-12 sessions + SSRI, no treatment, TF-CBT individual >12 sessions, and counselling.

Table 204. Analysis A, base-case results of economic modelling: interventions for the treatment of PTSD in adults [efficacy at treatment endpoint based on NMA of continuous data (changes in PTSD symptom scores); no beneficial effect beyond treatment endpoint]

Figure 701 provides the cost effectiveness plane of the analysis. Each intervention is placed on the plane according to its incremental costs and QALYs compared with no treatment, which is placed at the origin.

Figure 701. Analysis A (base-case): Cost-effectiveness plane of interventions for the treatment of PTSD in adults [efficacy at treatment endpoint based on NMA of continuous data (changes in PTSD symptom scores); no beneficial effect beyond treatment endpoint]

The CEAC and CEAF of the analysis are shown in Figure 702 and Figure 703, respectively. It can be seen that psychoeducation is the most cost-effective intervention for up to a cost effectiveness threshold of £9,000/QALY, with a probability that exceeds 0.49. TF-CBT individual < 8 sessions is the most cost-effective option for higher cost effectiveness thresholds and up to £40,000/QALY, but its probability of being cost-effective does not exceed 0.30 at any cost effectiveness threshold. It should be noted that, although TF-CBT individual <8 sessions is the most cost-effective option at a cost effectiveness threshold of £9,000/QALY and above, it does not have the highest probability of being cost-effective at any point beyond this threshold. In contrast, psychoeducation shows the highest probability of being cost-effective, despite of the fact that it has a lower mean NMB compared with TF-CBT individual < 8 sessions for cost effectiveness thresholds of £9,000/QALY and above. This means that, for cost effectiveness thresholds of £9,000/QALY and above, TF-CBT individual < 8 sessions has the highest mean NMB across the 10,000 iterations, but psychoeducation has a higher NMB than TF-CBT individual < 8 sessions in a larger number of iterations (which translates into a higher probability of psychoeducation being costeffective). This finding is explained by the close NMB values between the TF-CBT individual < 8 sessions and psychoeducation across iterations (which, on average, are higher for TF-CBT individual < 8 sessions) and the more positive skew in the distribution of the NMB of psychoeducation, in comparison to the distribution of the NMB of TF-CBT individual < 8 sessions (this phenomenon is explained in detail in Fenwick et al., 2001).

Figure 702. Analysis A (base-case): Cost-effectiveness acceptability curves of interventions for the treatment of PTSD in adults [efficacy at treatment endpoint based on NMA of continuous data (changes in PTSD symptom scores); no beneficial effect beyond treatment endpoint]

Figure 703. Analysis A (base-case): Cost-effectiveness acceptability frontier of interventions for the treatment of PTSD in adults [efficacy at treatment endpoint based on NMA of continuous data (changes in PTSD symptom scores); no beneficial effect beyond treatment endpoint]

Results were robust to the scenarios explored through deterministic analysis. The top 7 most cost-effective interventions remained the same, although in some of the analyses their relative ranking changed, in particular when an alternative set of utility values was attached to the model health states.

Analysis B: efficacy at treatment endpoint based on NMA of continuous data (changes in PTSD symptoms scores); beneficial effect up to 3-month follow-up (obtained from NMA of continuous data at 1-4 month follow-up)

The results of this analysis are provided in Table 205. TF-CBT individual < 8 sessions was the most cost-effective intervention, followed by psychoeducation, combined somatic and cognitive therapies and EMDR. These were followed by self-help with support, self-help without support, SSRI, IPT, TF-CBT individual 8-12 sessions, non-TF-CBT, TF-CBT individual >12 sessions, present-centred therapy, TF-CBT group 8-12 sessions, TF-CBT individual 8-12 sessions + SSRI, counselling and, finally, no treatment. The probability of TF-CBT individual < 8 sessions being the most cost-effective intervention was only 0.18.

Table 205. Analysis B, results of economic modelling: interventions for the treatment of PTSD in adults [efficacy at treatment endpoint based on NMA of continuous data (changes in PTSD symptoms scores); beneficial effect up to 3-month follow-up (obtained from NMA of continuous data at 1-4 month follow-up)]

Figure 704 provides the cost effectiveness plane of the analysis. Each intervention is placed on the plane according to its incremental costs and QALYs compared with no treatment.

Figure 704. Analysis B: Cost-effectiveness plane of interventions for the treatment of PTSD in adults [efficacy at treatment endpoint based on NMA of continuous data (changes in PTSD symptoms scores); beneficial effect up to 3-month follow-up (obtained from NMA of continuous data at 1-4 month follow-up)]

The CEAC and CEAF of the analysis are shown in Figure 705 and Figure 706, respectively. Psychoeducation is the most cost-effective intervention for up to a cost effectiveness threshold of £12,000/QALY, with a probability that exceeds 0.40. TF-CBT individual < 8 sessions is the most cost-effective option for higher cost effectiveness thresholds and up to £40,000/QALY, but its probability of being cost-effective does not go beyond 0.21 at any cost effectiveness threshold. Similar to analysis A, it can be seen that although TF-CBT individual <8 sessions is the most cost-effective option at a cost effectiveness threshold of £12,000/QALY and above, it does not have the highest probability of being cost-effective at any point beyond this threshold. In contrast, psychoeducation shows the highest probability of being cost-effective, despite of the fact that it has a lower mean NMB compared with TF-CBT individual < 8 sessions for cost effectiveness thresholds of £12,000/QALY and above. As with analysis A, this finding is attributable to the close NMB values between the TF-CBT individual < 8 sessions and psychoeducation across iterations and the more positive skew in the distribution of the NMB of psychoeducation, in comparison to the distribution of the NMB of TF-CBT individual < 8 sessions.

Figure 705. Analysis B: Cost-effectiveness acceptability curves of interventions for the treatment of PTSD in adults [efficacy at treatment endpoint based on NMA of continuous data (changes in PTSD symptoms scores); beneficial effect up to 3-month follow-up (obtained from NMA of continuous data at 1-4 month follow-up)]

Figure 706. Analysis B: Cost-effectiveness acceptability frontier of interventions for the treatment of PTSD in adults [efficacy at treatment endpoint based on NMA of continuous data (changes in PTSD symptoms scores); beneficial effect up to 3-month follow-up (obtained from NMA of continuous data at 1-4 month follow-up)]

Results were overall robust to the scenarios explored through deterministic analysis. The top 7 most cost-effective interventions remained the same, although in some of the analyses their relative ranking changed, in particular when an alternative set of utility values was attached to the model health states.

Analysis C: efficacy at treatment endpoint based on NMA of dichotomous remission data; no beneficial effect beyond treatment endpoint

The results of this analysis are provided in Table 206. In contrast to the other two analyses, non-TF-CBT was found to be the most effective and cost-effective intervention, followed, regarding cost effectiveness, by EMDR and TF-CBT individual 8-12 sessions. These were followed by IPT, SSRI, self-help without support, self-help with support, present-centred therapy, TF-CBT individual 8-12 sessions + SSRI, TF-CBT individual >12 sessions, counselling, TF-CBT group 8-12 sessions and no treatment. The probability of non-TF-CBT being the most cost-effective intervention was 0.42.

Table 206. Analysis C, results of economic modelling: interventions for the treatment of PTSD in adults [efficacy at treatment endpoint based on NMA of dichotomous remission data; no beneficial effect beyond treatment endpoint]

Figure 707 provides the cost effectiveness plane of the analysis. Each intervention is placed on the plane according to its incremental costs and QALYs compared with no treatment.

Figure 707. Analysis C: Cost-effectiveness plane of interventions for the treatment of PTSD in adults [efficacy at treatment endpoint based on NMA of dichotomous remission data; no beneficial effect beyond treatment endpoint]

The CEAC and CEAF of the analysis are shown in Figure 708 and Figure 709, respectively.

Non-TF-CBT is the most cost-effective option at any cost effectiveness threshold between zero and £40,000/QALY, with a probability of being cost-effective of 0.42 at the NICE lower cost effectiveness threshold of £20,000/QALY.

Figure 708. Analysis C: Cost-effectiveness acceptability curves of interventions for the treatment of PTSD in adults [efficacy at treatment endpoint based on NMA of dichotomous remission data; no beneficial effect beyond treatment endpoint]

Figure 709. Analysis C: Cost-effectiveness acceptability frontier of interventions for the treatment of PTSD in adults [efficacy at treatment endpoint based on NMA of dichotomous remission data; no beneficial effect beyond treatment endpoint]

Results were overall robust to the scenarios tested through deterministic sensitivity analysis, and the top 7 most cost-effective interventions remained the same, with some changes in relative ranking, in particular when an alternative set of utility values was attached to the model health states.

Discussion – conclusions, strengths and limitations of economic analysis

The guideline economic analysis assessed the cost effectiveness of a range of psychological interventions, as well as SSRIs and combined TF-CBT with SSRIs, for the treatment of PTSD in adults. The interventions assessed were determined by the availability of efficacy data obtained from the NMAs that were conducted to inform this guideline. TF-CBT interventions were categorised according to their mode of delivery in individual, group and mixed (where the intervention was delivered in a combination of individual and group sessions). Each of these categories was further subdivided, as relevant, to those comprising fewer than 8 sessions, 8-12 sessions, and more than 12 sessions, and were considered separately in the NMA and the economic analysis, to reflect the different intervention costs and, potentially, different efficacy associated with each sub-category.

The base-case analysis utilised continuous efficacy data at treatment endpoint, comprising changes in PTSD symptom scores, which were transformed to log-odds ratios of remission using a published formula; this analysis conservatively assumed that the beneficial effect of interventions lasts only until treatment endpoint and that after this period, the probability of remission is equal to that of baseline treatment (no treatment). An alternative scenario, which assumed a beneficial treatment effect of up to 3 months post-treatment (based on continuous follow-up data) was also explored in a second analysis. Finally, a third analysis which utilised more limited dichotomous efficacy data at treatment endpoint, and which also assumed no further treatment effect beyond treatment endpoint, was tested in an attempt to validate the conclusions of the base-case analysis. However, it needs to be noted that the definition of remission is different between this analysis and the base-case analysis: in the analysis that derived remission from continuous data (changes in PTSD symptom scale scores), remission was defined as a final score below a hypothetical cut-off point on a PTSD symptom scale with an underlying normal distribution. In contrast, in the analysis that utilised dichotomous remission data, remission was defined, in most studies, as a loss of PTSD diagnosis using DSM, ICD or similar criteria, and, in a small number of studies, as a final score below a cut-off point on a PTSD symptom scale.

In the base-case analysis (which utilised continuous data at treatment endpoint and assumed no treatment effect beyond treatment endpoint), the order of interventions from the most to the least cost-effective for the treatment of PTSD in adults was: TF-CBT individual < 8 sessions, psychoeducation, EMDR, combined somatic and cognitive therapies, self-help with support, SSRI, self-help without support, TF-CBT individual 8-12 sessions, IPT, non-TF-CBT, present-centred therapy, TF-CBT group 8-12 sessions, combined TF-CBT individual 8-12 sessions + SSRI, no treatment, TF-CBT individual >12 sessions, and counselling. The probability of TF-CBT individual < 8 sessions being the most cost-effective treatment option was 0.28.

When a beneficial effect of up to 3 months post-treatment was assumed, there were no dramatic changes in the results; the ranking of combined somatic and cognitive therapies, self-help without support and IPT improved by one place, whereas EMDR and TF-CBT individual 8-12 sessions dropped one place in ranking. The order of interventions became TF-CBT individual < 8 sessions, psychoeducation, combined somatic and cognitive therapies, EMDR, self-help with support, self-help without support, SSRI, IPT, TF-CBT individual 8-12 sessions, non-TF-CBT, TF-CBT individual >12 sessions, present-centred therapy, TF-CBT group 8-12 sessions, TF-CBT individual 8-12 sessions + SSRI, counselling, and no treatment. The probability of TF-CBT individual < 8 sessions being the most cost-effective treatment option was 0.18.

When dichotomous remission data were used, there were more important changes in the results with non-TF-CBT becoming the most cost-effective intervention followed by EMDR, TF-CBT individual 8-12 sessions, IPT, SSRI, self-help without support, self-help with support, present-centred therapy, TF-CBT individual 8-12 sessions + SSRI, TF-CBT individual >12 sessions, counselling, TF-CBT group 8-12 sessions, and no treatment. The probability of non-TF-CBT being the most cost-effective treatment was 0.42.

Results of the economic analysis were robust to changes in input parameters tested in deterministic sensitivity analysis.

Overall, across the 3 analyses, TF-CBT individual < 8 sessions, psychoeducation, EMDR, combined somatic and cognitive therapies and self-help with support appear to be the most cost-effective interventions for the treatment of PTSD in adults, as they all ranked in the top 5 places in the base-case economic analysis and on at least one of the secondary analyses (it is noted that, with the exception of EMDR and self-help with support, dichotomous remission data were not available for the other 3 interventions and therefore these were not considered in the respective secondary economic analysis). TF-CBT individual > 12 sessions, counselling, combined TF-CBT + SSRI, group TF-CBT and present-centred therapy do not appear to be cost-effective relative to other active interventions assessed, as they all ranked in the bottom 5 places among active interventions in all 3 economic analyses. Counselling and TF-CBT individual > 12 sessions, in particular, were found to be less cost-effective than no treatment in the base-case analysis. In-between, there is another group of interventions (SSRIs, TF-CBT individual 8-12 sessions, self-help without support, non-TF-CBT, IPT) that occupied middle cost effectiveness rankings (i.e. places 6-10) in the 2 analyses that utilised continuous data at treatment endpoint; these interventions showed an improved cost effectiveness in the analysis that utilised dichotomous remission data at treatment endpoint, with non-TF-CBT becoming the most cost-effective option in this analysis; however, this secondary analysis utilised efficacy data from a more limited number of interventions and did not include 3 of the interventions that were shown to be among the most cost-effective options in the analyses that utilised continuous data at treatment endpoint (i.e. TF-CBT individual < 8 sessions, psychoeducation, and combined somatic and cognitive therapies).

One thing worth noting is that increasing the number of sessions of individual TF-CBT does not appear to translate into higher efficacy or cost effectiveness, as shown in the results of the NMA and the economic analysis, respectively. However, this may be attributable to the populations in the studies assessing individual TF-CBT of different intensity: it is likely that participants who were recruited in trials that assessed a higher number of individual TF-CBT sessions had also more severe symptoms of PTSD at baseline, and therefore might have a more limited response to treatment compared with participants in trials that tested a smaller number of individual TF-CBT sessions. It is also worth noting that group TF-CBT does not appear to be effective or cost-effective relative to individual forms of TF-CBT in adults with PTSD.

The analysis utilised clinical effectiveness parameters derived from NMAs. This methodology enabled evidence synthesis from both direct and indirect comparisons between interventions, and allowed simultaneous inference on all treatments examined in pair-wise trial comparisons while respecting randomisation (Caldwell et al., 2005; Lu & Ades, 2004). The quality and limitations of RCTs considered in the NMAs have unavoidably impacted on the quality of the economic model clinical input parameters. For example, economic results may be have been affected by reporting and publication bias.

Effects for some interventions were informed by limited evidence: TF-CBT group 8-12 sessions, present centred therapy and IPT were tested on 57, 99 and 55 individuals, respectively, regarding the change in PTSD symptoms scores at treatment endpoint. In the outcome of remission, non-TF-CBT, TF-CBT group 8-12 sessions, IPT, present-centred therapy, self-help without support, SSRI and TF-CBT individual 8-12 sessions + SSRI were tested on fewer than 100 participants each. Even more limited evidence was available in the NMA of continuous follow-up data: effects for combined somatic and cognitive therapies, IPT and self-help without support were based on data from fewer than 50 participants for each intervention, whereas effects for TF-CBT individual >12 sessions, present-centred therapy and self-help with support were based on data from 50-100 participants each. It should be noted that TF-CBT individual 8-12 sessions had the most robust evidence base across all outcomes assessed in NMA.

It is also noted that, regarding changes in continuous PTSD symptoms scores at treatment endpoint, psychoeducation has been tested on 152 participants across 2 trials. However, the relative effect of psychoeducation versus no treatment in the respective NMA of continuous was in fact determined by data reported in one trial (Chambers 2014), in which psychoeducation (tested on 131 participants) was compared with TF-CBT individual < 8 sessions. In that trial, psychoeducation had a moderately lower effect than its comparator, which was marginally statistically significant. However, the effect of TF-CBT individual < 8 sessions versus waitlist was very large in the NMA and this resulted in a rather large relative effect of psychoeducation versus waitlist as well (median odds ratio 39.63), which, combined with its low intervention cost, determined its high cost effectiveness in the economic analysis. It is worth noting that the effect of psychoeducation versus wait list was characterised by particularly high uncertainty, as indicated by its very wide 95% credible intervals (1.03 to 1,446.64).

Global inconsistency checks and further inconsistency checks through node-splitting indicated that there was no inconsistency between direct and indirect evidence considered in the NMA that utilised continuous data at treatment endpoint (changes in PTSD symptom scale scores). In contrast, some evidence of inconsistency was identified in the NMA of continuous data at 1-4 month follow-up (which was utilised in analysis B) and the NMA of dichotomous remission data at treatment endpoint (which was utilised in analysis C). Therefore, economic analysis A appears to be the only one that utilised NMA data with no inconsistency between direct and indirect evidence. Moreover, heterogeneity across all NMAs was found to be high. It is also noted that the relative effects of most interventions versus waitlist were very large and characterised, in many cases, by considerably wide 95% credible intervals. These findings need to be taken into account when interpreting the results of the NMAs but also the cost effectiveness results.

The economic model did not consider discontinuation in the model structure due to the relatively limited discontinuation data available. However, for the NMA that informed the economic analysis, ITT continuous data were extracted, where available. This means that discontinuation has been implicitly taken into account in the economic model outcomes. Moreover, the probabilistic analysis took into account the completion rates of the interventions assessed in the RCTs that informed the economic analysis, so that the number of sessions reflected, up to a degree, the attrition rates characterising each intervention.

The baseline risk of remission was estimated based on 664 people aged 16-85 years, who participated in the 2007 Australian National Survey of Mental Health and Wellbeing and had experienced PTSD at some point in their life. The risk of relapse was not possible to estimate using published evidence, and therefore was based on an assumption following the committee’s advice. However, a range of values was tested in deterministic sensitivity analysis. Other data, such as the increased risk of death associated with PTSD, and the risk of developing common side effects from SSRIs were based on published evidence.

The time horizon of the analysis was 3 years, which were considered adequate to capture longer terms and costs associated with a course of treatment for PTSD without significant extrapolation over the course of PTSD.

Utility data used in the economic model were derived from a systematic review of studies reporting utility data for PTSD-related health states. The review included three studies. One study met the NICE preferences for the type of utility data to be used in economic evaluation. However, these data were deemed unsuitable by the committee, due to concerns on the eligibility of the study participants. The economic analysis considered utility data from one study on adults with and without current PTSD diagnosis, who participated in a national mental health survey in Australia and provided HRQoL ratings that were transformed into utility data using the AQoL-4D preference-based measure; deterministic sensitivity analysis used SF-6D utility data derived from veterans with and without PTSD in the US.

Intervention costs were estimated based on relevant information provided in the studies included in the NMA supplemented by the committee’s expert opinion, in order to reflect routine NHS practice.

NHS and PSS costs incurred by adults with PTSD and those remitting from PTSD were based on resource use data reported in the most recent (2014) Adult Psychiatric Morbidity Survey conducted in England for people with PTSD and people without PTSD, combined with the committee’s expert opinion, other published sources of relevant resource use data and national unit costs. The committee determined the exact resource use associated with each resource use component (e.g. number of visits to health professionals), due to lack of any relevant information. This exercise determined the costs of the PTSD and the no PTSD health states, which were estimated to approximate £1,173 and £110, respectively, per annum.

According to a cost of illness study conducted in Northern Ireland (Ferry et al., 2015), the total direct NHS/PSS cost incurred by people with PTSD in Northern Ireland was 32,975,590 in 2008 prices, and 74,935 people were estimated to have PTSD within 12 months. This translates to a cost per person with PTSD of £518 in 2017 prices, which is a figure considerably lower than that estimated for the guideline economic analysis for adults with PTSD (£1,173). However, the study used a different methodology for the estimation of costs, which may justify, at least partially, the difference between the two figures. On the other hand, annual cost figures for children with PTSD and children recovering from PTSD reported for children (Shearer et al., 2018) [£2,596 and £1,114, respectively] are considerably higher that the respective figures estimated for adults with PTSD in the guideline economic analysis. However, these costs for children were estimated for participants in a RCT, where all utilised healthcare resources. In contrast, the figures estimated for the guideline economic analysis for adults with/without PTSD were based on survey data, in which a significant proportion of people did not receive any treatment for their mental or emotional problem. In any case, deterministic analysis explored the impact of a ± 50% change in the NHS/PSS cost of the PTSD health state on the results of the economic analysis.

Overall conclusions from the guideline economic analysis

The guideline base-case economic analysis suggests that TF-CBT individual < 8 sessions, psychoeducation, EMDR, combined somatic and cognitive therapies and self-help with support are the 5 most cost-effective interventions for the treatment of PTSD in adults. TF-CBT individual > 12 sessions, counselling, combined TF-CBT + SSRI, group TF-CBT and present-centred therapy appear to be less cost-effective relative to other active interventions.

Counselling and TF-CBT individual > 12 sessions were also found to be less cost-effective than no treatment in the base-case analysis. In-between, there is another group of interventions (SSRIs, TF-CBT individual 8-12 sessions, self-help without support, non-TF-CBT, IPT) that occupied middle cost effectiveness rankings (i.e. places 6-10) in the basecase analysis.

The result for psychoeducation, which was found to be among the most cost-effective interventions, should be interpreted with great caution due to limitations in the evidence base and the considerably high uncertainty characterising its efficacy estimate. Moreover, the NMA that informed the base-case analysis was characterised by high between-study heterogeneity, as well as large effects and considerable uncertainty for some interventions, and this should be taken into account when interpreting the results of the analysis.

Results from the alternative scenarios explored in the other two analyses (i.e. consideration of efficacy data derived from the NMAs of continuous 1-4 month follow-up data and of dichotomous remission data) are somewhat different from the base-case analysis, in particular those derived from use of dichotomous remission data, which included a smaller number of interventions due to unavailability of relevant data; the results from these analyses should be interpreted with caution due to the limitations characterising the respective evidence base and the NMAs that informed them (limited evidence base, evidence of inconsistency between direct and indirect evidence, high heterogeneity, large effects and considerable uncertainty for some interventions).

References

• Ahmadi N, Hajsadeghi F, Mirshkarlo HB et al (2011) Post-traumatic stress disorder, coronary atherosclerosis, and mortality. American Journal of Cardiology 108(1), 29–33 [PubMed: 21530936]
• Anderson HD, Pace WD, Libby AM et al. (2012) Rates of 5 common antidepressant side effects among new adult and adolescent cases of depression: a retrospective US claims study. Clinical Therapeutics 34(1), 113–23 [PubMed: 22177545]
• Brazier J, Roberts J, Deverill M (2002) The estimation of a preference-based measure of health from the SF-36. Journal of Health Economics 21(2), 271–92 [PubMed: 11939242]
• Breslau N, Kessler RC, Chilcoat HD et al. (1998) Trauma and posttraumatic stress disorder in the community: the 1996 Detroit Area Survey of Trauma. Archives of General Psychiatry 55(7), 626–32 [PubMed: 9672053]
• Briggs A, Sculpher M, Claxton K (2006) Decision Modelling for Health Economic Evaluation. New York, NY: Oxford University Press
• British Association for Behavioural & Cognitive Psychotherapies (2016) Criteria and guidelines for re-accreditation as a behavioural and/or cognitive psychotherapist. British Association for Behavioural & Cognitive Psychotherapies. Available from: http://www​.babcp.com​/files/Accreditation​/CBP/Full/CBP-Full-Guidelines-V5-0614​.pdf
• Brooks R (1996) EuroQol: the current state of play. Health Policy 37(1), 53–72 [PubMed: 10158943]
• Caldwell DM, Ades AE, Higgins JP (2005) Simultaneous comparison of multiple treatments: combining direct and indirect evidence. BMJ 331(7521), 897–900 [PMC free article: PMC1255806] [PubMed: 16223826]
• Chapman C, Mills K, Slade T et al. (2012) Remission from post-traumatic stress disorder in the general population. Psychological Medicine 42(8), 1695–703 [PubMed: 22166813]
• Chinn S (2000) A simple method for converting an odds ratio to effect size for use in meta-analysis. Statistics in Medicine 19(22), 3127–31 [PubMed: 11113947]
• Cohen J (1969) Statistical power analysis for the behavioral sciences. New York: Academic Press
• Cooper H, Hedges LV, Valentine JC (2009) The Handbook of Research Synthesis and Meta-analysis. New York: Russel Sage Foundation.
• Coupland C, Dhiman P, Morriss R et al. (2011) Antidepressant use and risk of adverse outcomes in older people: population based cohort study. BMJ 343: d4551. [PMC free article: PMC3149102] [PubMed: 21810886]
• Curtis L and Burns A (2017) Unit Costs of Health & Social Care 2017. Canterbury: PSSRU, University of Kent.
• Dias S, Ades AE, Welton NJ et al. (2018) Network Meta-analysis for Decision-Making, Hoboken NJ: Wiley. Chapter 4: Generalised Linear Models, p93–154
• Dias S, Welton NJ, Sutton AJ et al. (2011a, last updated September 2016) NICE DSU Technical support document 2: a generalised linear modelling framework for pairwise and network meta-analysis of randomised controlled trials. Available from http://nicedsu​.org.uk​/technical-support-documents​/technical-support-documents/ [PubMed: 27466657]
• Dias S, Welton NJ, Sutton AJ et al. (2011b, updated April 2014) NICE DSU Technical support document 4: inconsistency in networks of evidence based on randomised controlled trials. Available from http://nicedsu​.org.uk​/technical-support-documents​/technical-support-documents/ [PubMed: 27466656]
• Dolan P (1997) Modeling valuations for EuroQol health states. Medical Care 35(11), 1095–108. [PubMed: 9366889]
• Ehlers A, Grey N, Wild J et al. (2013) Implementation of cognitive therapy for PTSD in routine clinical care: effectiveness and moderators of outcome in a consecutive sample. Behaviour Research and Therapy, 51(11), 742–52 [PMC free article: PMC3897916] [PubMed: 24076408]
• Fenwick E, Claxton K, Sculpher M (2001) Representing uncertainty: the role of cost-effectiveness acceptability curves. Health Economics 10(8), 779–87 [PubMed: 11747057]
• Ferry FR, Brady SE, Bunting BP et al. (2015) The Economic Burden of PTSD in Northern Ireland. Journal of Traumatic Stress 28(3), 191–7 [PubMed: 25990825]
• Freed MC, Yeager DE, Liu X et al. (2009) Preference-weighted health status of PTSD among veterans: an outcome for cost-effectiveness analysis using clinical data. Psychiatric Services 60(9), 1230–8 [PubMed: 19723738]
• Furukawa TA, Noma H, Caldwell DM et al. (2014) Waiting list may be a nocebo condition in psychotherapy trials: a contribution from network meta-analysis. Acta Psychiatrica Scandinavica 130(3), 181–92 [PubMed: 24697518]
• Haagsma JA, Polinder S, Olff M et al. (2012) Posttraumatic stress symptoms and health-related quality of life: a two year follow up study of injury treated at the emergency department. BMC Psychiatry 12: 1 [PMC free article: PMC3276433] [PubMed: 22230388]
• Jakobsen JC, Katakam KK, Schou A et al. (2017) Selective serotonin reuptake inhibitors versus placebo in patients with major depressive disorder. A systematic review with meta-analysis and Trial Sequential Analysis. BMC Psychiatry 17(1): 58 [PMC free article: PMC5299662] [PubMed: 28178949]
• Kontopantelis E, Olier I, Planner C et al. (2015) Primary care consultation rates among people with and without severe mental illness: a UK cohort study using the Clinical Practice Research Datalink. BMJ Open 5(12): e008650 [PMC free article: PMC4691766] [PubMed: 26674496]
• Lu G and Ades AE (2004) Combination of direct and indirect evidence in mixed treatment comparisons. Statistics in Medicine 23(20), 3105–24 [PubMed: 15449338]
• Lunn DJ, Thomas A, Best N et al. (2000) WinBUGS-A Bayesian modelling framework: Concepts, structure, and extensibility. Statistics and Computing 10, 325–337
• Mavridis D, Giannatsi M, Cipriani A et al. (2015) A primer on network meta-analysis with emphasis on mental health. Evidence-Based Mental Health 18(2), 40–6 [PubMed: 25908686]
• McManus S, Bebbington P, Jenkins R et al. (2016) Mental health and wellbeing in England: Adult Psychiatric Morbidity Survey 2014. Leeds: NHS Digital
• Mihalopoulos C, Magnus A, Lal A et al. (2015) Is implementation of the 2013 Australian treatment guidelines for posttraumatic stress disorder cost-effective compared to current practice? A cost-utility analysis using QALYs and DALYs. The Australian and New Zealand Journal of Psychiatry 49(4), 360–76 [PubMed: 25348698]
• Morina N, Wicherts JM, Lobbrecht J et al. (2014) Remission from post-traumatic stress disorder in adults: a systematic review and meta-analysis of long term outcome studies. Clinical Psychology Review 34(3), 249–55 [PubMed: 24681171]
• National College for Teaching and Leadership (2016) Review of clinical and educational psychology training arrangements. NHS Health Education England
• National Institute for Health and Clinical Excellence (2008) Social Value Judgements: Principles for the Development of NICE Guidance (2nd ed.) [PubMed: 27905706]
• National Institute for Health and Care Excellence (2013) Guide to the Methods of Technology Appraisal 2013 (PMG 9) [PubMed: 27905712]
• National Institute for Health and Care Excellence (2014, last updated April 2017) Developing NICE guidelines: the manual (PMG 20) [PubMed: 26677490]
• Netten A, Knight J, Dennett J et al. (1998) Development of a ready reckoner for staff costs in the NHS, Vols 1 & 2. Canterbury: PSSRU, University of Kent
• NHS Digital (2017) Hospital Episode Statistics for England. Admitted Patient Care statistics, 2016-17. NHS Digital
• NHS Digital, Community and Mental Health team (2016) Psychological Therapies: Annual report on the use of IAPT services. England, 2015-16
• NHS Improvement (2017) National Schedule of Reference Costs, 2016-17 - NHS trusts and NHS foundation trusts. Available from: https://improvement​.nhs​.uk/resources/reference-costs/
• NHS Business Services Authority, NHS Prescription Services (2018) NHS England and Wales. Electronic Drug Tariff. Issue: March 2018. Compiled on the behalf of the Department of Health. Available from: https://www​.nhsbsa.nhs​.uk/pharmacies-gp-practices-and-appliance-contractors​/drug-tariff
• Office for National Statistics (2017a) National Life Tables, England, 1980-82 to 2014-2016. Available from: http://www​.ons.gov.uk​/peoplepopulationandcommunity​/birthsdeathsandmarriages​/lifeexpectancies​/bulletins​/nationallifetablesunitedkingdom​/2015-09-23 [Accessed on 27/05/2016]
• Office for National Statistics (2017b) Population Estimates for UK, England and Wales, Scotland and Northern Ireland: Mid-2016. Available from: http://www​.ons.gov.uk​/peoplepopulationandcommunity​/birthsdeathsandmarriages​/lifeexpectancies​/bulletins​/nationallifetablesunitedkingdom​/2015-09-23 [Accessed on 27/05/2016]
• Pietrzak RH, Feder A, Singh R et al. (2014) Trajectories of PTSD risk and resilience in World Trade Center responders: an 8-year prospective cohort study. Psychological Medicine 44(1), 205–19 [PubMed: 23551932]
• Resic, PA, Williams LF, Suvak MK et al. (2012) Long-term outcomes of cognitive-behavioral treatments for posttraumatic stress disorder among female rape survivors. Journal of Consulting and Clinical Psychology 80(2), 201–10 [PMC free article: PMC3336190] [PubMed: 22182261]
• Rosellini AJ, Liu H, Petukhova MV, et al. (2018) Recovery from DSM-IV post-traumatic stress disorder in the WHO World Mental Health surveys. Psychological Medicine 48(3), 437–450 [PMC free article: PMC5758426] [PubMed: 28720167]
• Shearer J, Papanikolaou N, Meiser-Stedman R et al. (2018) Cost-effectiveness of cognitive therapy as an early intervention for post-traumatic stress disorder in children and adolescents: a trial based evaluation and model. Journal of Child Psychology and Psychiatry 59(7), 773–780 [PMC free article: PMC6597355] [PubMed: 29197091]
• Solomon Z and Mikulincer M (2006) Trajectories of PTSD: a 20-year longitudinal study. American Journal of Psychiatry 163(4), 659–66 [PubMed: 16585441]