1Behavioural interventions (CBT/ GET/ pacing)

Study IDParticipantsInterventions/ comparatorsWithdrawals and adverse events
Cox (2002)154

Study design
Controlled trial

Level of evidence
2−
Number: 97
Adults or children?: Adults

Inclusion criteria: Confirmed diagnosis of CFS with no other secondary diagnosis; aged 15–60 years; current or pending inpatient for management approach

Exclusion criteria: Other specific diagnoses such as Parkinson's disease, MS, post-traumatic stress disorder, post-polio syndrome and/or personality disorder; aged 15–60 years; previous admission for management approach; previous management by CFS team as an outpatient; non-completion of inpatient treatment programme (staying less than 14 days)

Diagnosis/ case definition: CDC (1994)
Age: mean 33 yrs treatment group, 37 yrs control group
% Female: 79% treatment group, 83% control group
Duration of illness: median 56 months treatment group, 60.5 months control group
Baseline functioning: 92% not working or studying in treatment group, 97% in control group

Further details:
5% past history of anxiety, 13% in treatment group and 16% in control group had past history of depression
Recruited from NHS trust neurosciences centre
Diagnosis discussed with the medical team
Occupational Therapy Lifestyle management Programme uses principles of CBT and graded activity within a biopsychosocial framework. Series of 10 educational topics for daily management of CFS. waiting list control group

Number of participants in each group
61 in treatment group, 36 in control group
Withdrawals:
Questionnaire completion rate 6 months after discharge was 46/60 in the treatment group and 19/35 in the control group.

Adverse events:
Outcome 1Outcome 2Outcome 3Outcome 4
Outcome measured: SF-36 physical functioning

Results in intervention group
6 months(n=44): 20 improved, 14 stayed the same, 10 got worse
Results in control group
6 months(n=19): 7 improved, 3 stayed the same, 9 got worse

Comments
no significant difference between groups
Outcome measured
Health and Fatigue Qre - total fatigue

Results in intervention group
6 months (n=43): 25 improved, 6 stayed the same, 13 got worse
Results in control group
6 months (n=19): 10 improved, 3 stayed the same, 6 got worse

Comments
no significant difference between groups
Outcome measured
Fatigue
Perceived Fatigue Rating Scale

Results in intervention group
6 months (n=43): 26 improved, 17 got worse
Results in control group
6 months (n=19): 11 improved, 8 got worse

Comments
no significant difference between groups
Outcome measured
Emotional distress
Perceived Fatigue Rating Scale

Results in intervention group
6 months (n=43): 25 improved, 2 stayed the same, 16 got worse
Results in control group
6 months (n=19): 11 improved, 8 got worse

Comments
no significant difference between groups
Outcome 5Outcome 6Outcome 7Outcome 8
Outcome measured
Maintaining activity
Illness management questionnaire

Results in intervention group
6 months (n=43): 27 improved, 1 stayed the same, 15 got worse
Results in control group
6 months (n=19): 9 improved, 3 stayed the same, 7 got worse

Comments
significant difference in favour of treatment group (p=0.03)
Outcome measured
Accommodating to illness
Illness management questionnaire

Results in intervention group
6 months (n=43): 31 improved, 12 got worse
Results in control group
6 months (n=19): 7 improved, 1 stayed the same, 11 got worse

Comments
significant difference in favour of treatment group (p=0.02)
Outcome measured

Baseline values intervention group
Baseline values control group

Results in intervention group
Results in control group

Comments
Outcome measured

Baseline values intervention group
Baseline values control group

Results in intervention group
Results in control group
Comments
Additional comments: At discharge (end of treatment) there were significant differences between the gorups in: pain after exercise and total pain on the HFQ, fatigue and emotional distress on the PFRS, and maintaining activity, accommodating to illness and information seeking on the IMQ. At 3 months post-discharge there were significant differences between groups for length of current tiredness and pain after exercise on the HFQ, and maintaining activity, accommodating to illness and information seeking on the IMQ. At 6 months post-discharge there were significant differences between groups in health transition on the SF36, length of current tiredness on the HFQ and maintaining activity and accommodating to illness on the IMQ.
Study IDParticipantsInterventions/ comparatorsWithdrawals and adverse events
Cox (2002)155

Study design
Controlled trial

Level of evidence
2−
Number: 130
Adults or children?: Adults

Inclusion criteria: Confirmed diagnosis of CFS with no other secondary diagnoses, aged 15 – 60 years, current or pending inpatient for management approach, completion of treatment programme (for inpatient group)

Exclusion criteria: Other specific diagnoses such as Parkinson's disease, MS, PTSD, post polio syndrome and/ or personality disorder. Previous admission for specified CFS management approach, previous outpatient management by CFS team, non-completion of inpatient treatment programme.

Diagnosis/ case definition: CDC (1994)

Age: treatment group mean 33 yrs, control group mean 37 yrs

% Female: 79% treatment group, 83% control group

Duration of illness: median 56 months treatment gorup, 60.5 months comparison group

Baseline functioning: 92% not working or studying in the treatment group, 97% in the control group

Further details:
None
Recruited through a neuroscience Centre and a national ME charity. Experimental group were inpatients, control group were from the inpatient waiting list. 80% in the treatment group and 78% in the control group reported 'infection/ virus' at onset. 84% in each group currently on medication including tricyclic antidepressant, 5HT uptake inhibitor, hypnotic, analgesics.
Combined CBT and graded activity intervention not described

Number of participants in each group
65 inpatient group, 37 control group
Withdrawals: 5 withdrew from experimental group, 18 from the control group

Adverse events:
Results
Outcome 1Outcome 2Outcome 3Outcome 4
Outcome measured: Functional status
SF36: physical functioning; role functioning - physical; bodily pain; general health; vitality; social functioning; role functioning - emotional; mental health; reported health transition

Baseline values intervention group
27 (22.03); 11.4 (24.6); 35.07 (22.72); 31.23 (19.21); 19.04 (18.19); 25.04 (19.34); 53.21 (46.23); 58.18 (18.25); 3.2 (1.05)
Baseline values control group
27.86 (19.6); 5.71 (19.26); 33.86 (21.78); 24.97 (15.29); 15.57 (15.52); 29.27 (24.43); 66.67 (43.54); 60.91 (23.57); 3.4 (1.1)

Results in intervention group
30.46 (21.48); 13.07 (22.55); 40.4 (26.34); 31.3 (16.8); 25.23 (19.88); 33.52 (26.11); 70.44 (38.86); 61.73 (20.78); 2.7 (1)
Results in control group
30 (22.36); 5.26 (13.38); 44.95 (25.73); 34.63 (21.25); 25.79 (23.35); 38.82 (27.92); 70.18 (39.9); 68.63 (21.72); 2.9 (1.15)

Comments
no significant differences between groups on on any subscale at 6 months
Outcome measured
Scale of Perceived Ability and Recovery current level of ability; future level to achieve; influenced by you; influenced by others

Baseline values intervention group
32.2 (14.9); 88.6 (15.4); 77.4 (20.8); 70.9 (21.6)
Baseline values control group
37.4 (13.6); 84.9 (14.6); 84 (21.4); 75.1 (21.2)

Results in intervention group
38.3 (20.2); 82.9 (16.9); 71.7 (19.1); 61.2 (21)
Results in control group
41.1 (18.2); 82.6 (16.6); 72.6 (15.2); 63.2 (21.1)

Comments
no significant differences between groups
Outcome measured
Disability
visual analogue scale (max 40)

Baseline values intervention group
29.9 (6.8)
Baseline values control group
29.9 (7)

Results in intervention group
28 (7.9)
Results in control group
27.3 (23.3)

Comments
no significant difference between groups
Outcome measured
Mood
Hospital anxiety and depression scale

Baseline values intervention group
Anxiety: 8.5 (4.7), depression: 7.9 (3.8)
Baseline values control group
Anxiety: 8.9 (5.6), depression: 7.9 (4.2)

Results in intervention group
Anxiety: on discharge 7.8 (4.2), 3 months post discharge 7.3 (4.4), 6 months post discharge 7 (4.7); Depression: on discharge 6.5 (3.8), 3 months post discharge 6.7 (4.6), 6 months post discharge 7.3 (4.5)
Results in control group
Anxiety: on discharge 8.5 (5.4), 3 months post discharge 8 (5.5), 6 months post discharge 6.8 (5.4); depression: on discharge 7.4 (4.4), 3 months post discharge 7 (3.6), 6 months post discharge 5.7 (3.6)

Comments
no significant differences between groups
Outcome 5Outcome 6Outcome 7Outcome 8
Outcome measured
Fatigue
Chalder fatigue questionnaire (total fatigue)

Baseline values intervention group
23.7 (7.4)
Baseline values control group
23.8 (6.1)

Results in intervention group
on discharge 20.8 (7.7), 3 months post discharge 18.3 (7.9), 6 months post discharge 19.6 (7.8)
Results in control group
on discharge: 21.8 (7), 3 months post discharge 20 (7.8), 6 months post discharge 21 96)

Comments
no significant differences between groups
Outcome measured
Pain
max score 6

Baseline values intervention group
4.2 (1.7)
Baseline values control group
4 (1.5)

Results in intervention group
on discharge 3.3 (1.7), 3 months post discharge 3.4 (1.7), 6 months post discharge 3.7 (1.7)
Results in control group
on discharge 4 (1.7), 3 months post discharge 3.7 (1.9), 6 months post discharge 3.7 (1.3)

Comments
significant difference between groups on discharge (p<0.05)
Outcome measured
Profile of Fatigue Related Symptoms Fatigue; emotional distress; cognitive difficulties; somatic symptoms

Baseline values intervention group
4.2 (1.3); 2.3 (1.3); 3.4 (1.4); 2.8 (1.4)
Baseline values control group
4.3 (1.3); 2.4 (1.8); 3.2 (1.6); 2.9 (1.4)

Results in intervention group
on discharge: 3.7 (1.5); 2.1 (1.4); 3.1 (1.4); 2.4 (1.4), 3 months post discharge: 3.7 (1.6); 2.1 (1.5); 3 (1.4); 2.3 (1.5), 6 months post discharge: 4 (1.5); 2 (1.3); 3 (1.6); 2.3 (1.4)
Results in control group
on discharge: 4.4 (1.3); 2.5 (1.7); 3.2 (1.4); 2.7 (1.5), 3 months post discharge: 4.1 (1.3); 2.3 (1.7); 3 (1.4); 2.6 (1.4), 6 months post discharge: 3.7 (1.5); 2 (1.5); 2.9 (1.1); 2.3 91.3)

Comments
significant difference between groups in fatigue scores at discharge (p<0.02), improvement in fatigue on discharge (p<0.003) and improvement in emotional distress on discharge (p<0.03)
Outcome measured
Illness Management Questionnaire maintaining activity; accommodating the illness; focusing on symptoms; information seeking

Baseline values intervention group
3.2 (1); 3.9 (1); 3.4 (1); 3.9 (1)
Baseline values control group
3.5 (1); 3.8 (1.); 3.4 (1); 3.8 (1)

Results in intervention group
on discharge: 2.9 (1); 4.3 (1); 3.2 (1); 4 (1), 3 months post discharge: 3 (1); 4.3 (1); 3.1 (1); 3.7 (1), 6 months post discharge: 2.9 (1); 4.3 (1); 3.1 (1); 3.5 (1)
Results in control group
on discharge: 3.5 91); 3.7 (1); 3.2 (1); 3.3 (1), 3 months post discharge: 3.5 91); 3.8 (1); 3.4 (1); 3.3 (1), 6 months post discharge: 3.4 (1); 3.8 (1); 3.2 (1); 3.1 (1)
Comments
significant difference between groups on discharge for maintaining activity, accommodating to illness and information seeking; and at 3 and 6 months post discharge for maintaining activity and accommodating to illness
Study IDParticipantsInterventions/ comparatorsWithdrawals and adverse events
Deale (1997)147

Study design
RCT

Level of evidence
1++
Number: 60
Adults or children?: Not stated

Inclusion criteria: Consecutive referrals. Patients taking antidepressant medication or anxiolytics were eligible if dose was stable for 3 months before entry and during the trial. Excluded if had somatisation disorder, severe depression, ongoing physical investigations, concurrent new treatment and inability to attnd all treatment sessions

Exclusion criteria:

Diagnosis/ case definition: Oxford

Age: Mean 31 (sd=9) in CBT group, mean 38 (sd=11) in relaxation group

% Female: 70% female in CBT group, 67% in relaxation group

Duration of illness: Mean 3.4 (sd=2.1) years in CBT group, mean 4.6 (sd=3.3) years in realxation group

Baseline functioning: Both groups had near maximum scores on measures of functional impairment and fatigue, scores on general health questionnaire were moderate, but depression was not marked.

Further details:
5 patients had additional diagnoses of dysrhthmia, 9 had major depression, 3 had anxiety disorders, and 6 had both depression and anxiety disorders
Patients recruited from specialist CFS clinic, No significant differences between group for marital status, social class, proportion unemployed, proportion with psychiatric diagnosis, use of antidepressants or patient attribution of symptoms to physical illness. 12 patients used antidepressants and 2 used anxiolytics
Also met CDC 94 criteria
CBT
Patients recevied either 13 sessions over 4–6 months of CBT (graded activity and cognitive restructuring) or relaxation
Patients were seen individually

Number of participants in each group
30 in each group
Withdrawals: 7 patients dropped out of treatment and completed no more clinical measures: 3 from CBT, 1 found it ineffective, 1 felt too ill to attend as an outpatient (received inpatient CBT and improved), 1 improved and wanted no further treatment. 4 patients withdrew from relaxation, 1 felt to ill to continue, 1 gave no reason & 2 found relaxation exercises overly tiring.

Adverse events:
Results
Outcome 1Outcome 2Outcome 3Outcome 4
Outcome measured: Improvement in physical functioning
Proportion improved at 6 month follow-up. Increase of 50 or more from pre-treatment to 6 months follow-up or end score of 83+ on physical functioning scale of Gerneral Healh survey

Baseline values intervention group
Baseline values control group

Results in intervention group
70% excluding drop-outs, 63% including drop-outs
Results in control group
19% excluding drop-outs, 17% including drop-outs

Comments
Drop-outs classified as not improved.
Difference between groups = 51% (95% CI: 28–74), excluding drop-outs, 46% (95% CI: 24–68) including drop outs, p<0.001 for both comparisons
Outcome measured
Functioning
Physical functioning scale of Medical Outcomes
Study Short-Form General Health Survey

Baseline values intervention group
25.5 (18.9)
Baseline values control group
27.8 (27.1)

Results in intervention group
71.6 (28.0)
Results in control group
38.4 (26.9)

Comments
p for the difference between groups >0.50
Outcome measured
Work
Work and Social adjustment scale

Baseline values intervention group
6.0 (1.2)
Baseline values control group
6.1 (1.3)

Results in intervention group
3.3 (2.2)
Results in control group
5.4 (1.8)

Comments
p for the difference between groups <0.001
Outcome measured
Goals
Long-term goals rating (mean of two)

Baseline values intervention group
7.0 (0.7)
Baseline values control group
6.8 (1.0)

Results in intervention group
2.9 (1.9)
Results in control group
5.9 (1.8)

Comments
p for the difference between groups <0.001
Outcome 5Outcome 6Outcome 7Outcome 8
Outcome measured
Fatigue
Fatigue problem rating

Baseline values intervention group
7.0 (0.9)
Baseline values control group
6.3 (1.2)

Results in intervention group
3.4 (2.2)
Results in control group
5.5 (1.9)

Comments
p for the difference between groups <0.001
Outcome measured
Fatigue
Fatigue questionnaire

Baseline values intervention group
10.2 (1.3)
Baseline values control group
9.5 (2.6)

Results in intervention group
4.1 (4.0)
Results in control group
7.2 (4.0)

Comments
p for the difference between groups <0.01
Outcome measured
Depression
BDI score

Baseline values intervention group
14.5 (7.2)
Baseline values control group
14.2 (6.1)

Results in intervention group
10.1 (6.9)
Results in control group
12.3 (8.5)

Comments
p for the difference between groups >0.30
Outcome measured
General health
General health questionnaire

Baseline values intervention group
6.2 (3.6)
Baseline values control group
6.0 (4.2)

Results in intervention group
3.4 (3.7)
Results in control group
4.3 (3.9)
Comments
p for the difference between groups >0.70
Additional comments: Results presented are at 6 month follow-up, results presented as mean (sd) unless otherwise stated
Outcome 9Outcome 10Outcome 11Outcome 12
Outcome measured: Employment
Proportion employed

Baseline values intervention group
Baseline values control group

Results in intervention group
56%
Results in control group
39%

Comments
p=0.05
Outcome measured
Work
Mean hours worked per week

Baseline values intervention group
Baseline values control group

Results in intervention group
19.92 (sd=15.82)
Results in control group
9.89(sd=15.82)

Comments
p<0.05
Outcome measured
Global improvement

Baseline values intervention group
Baseline values control group

Results in intervention group
Results in control group

Comments
Logistic regression analysis of predictors of global improvement showed that age showed a significant relationship with global improvement, age and illness duration showed significant association with MOS physical functioning score and illness duration showed significant association with fatigue questionnaire. Pre-treatment fatigue score or psychiatric disorder showed no association with any measure of global improvement.
Outcome measured: Global improvement Self rating

Baseline values intervention group
Baseline values control group

Results in intervention group
70% better or much better
Results in control group
31% better or much better

Comments
p for the difference between groups <0.01
Outcome 13Outcome 14Outcome 15Outcome 16
Outcome measured
Satisfaction
Patient satisfaction with treatment outcome

Baseline values intervention group
Baseline values control group

Results in intervention group
78% satisfied or very satisfied
Results in control group
50% satisfied or very satisfied

Comments
p for the difference between groups <0.05
Outcome measured
Usefulness
Patient assessment of usefulness of treatment

Baseline values intervention group
Baseline values control group

Results in intervention group
96% useful or very useful
Results in control group
85% useful or very useful

Comments
p for the difference between groups >0.10
Outcome measured
Functioning
Blinded assessor rating of physical functioning at 3 month follow-up

Baseline values intervention group
Baseline values control group

Results in intervention group
80% better or much better
Results in control group
26% better or much better

Comments
p for the difference between groups <0.001
Outcome measured
Fatigue
Blinded assessor rating of fatigue at 3 month follow-up

Baseline values intervention group
Baseline values control group

Results in intervention group
72% better or much better
Results in control group
17% better or much better

Comments
p for the difference between groups <0.001
Results at 5 year follow up (Withdrawals: 25 CBT patients and 28 relaxation patients followed up at 5 years)
Outcome 1Outcome 2Outcome 3Outcome 4
Outcome measured: Global improvement Proportion much or very much better

Baseline values intervention group
Baseline values control group

Results in intervention group
64%
Results in control group
36%

Comments
p<0.05
Outcome measured
Functioning
MOS physical functioning scale, proportion with score>83

Baseline values intervention group
0
Baseline values control group
0

Results in intervention group
48%
Results in control group
32%

Comments
p=0.272
Outcome measured
Fatigue
Fatigue questionnaire, proportion with score <4

Baseline values intervention group
0%
Baseline values control group
7%

Results in intervention group
32%
Results in control group
25%

Comments
p=0.571
Outcome measured
General health
GHQ score < 4

Baseline values intervention group
30%
Baseline values control group
33%

Results in intervention group
48%
Results in control group
54%

Comments
p=0.579
Outcome 5Outcome 6Outcome 7Outcome 8
Outcome measured
Symptoms
Course of symptoms over time

Baseline values intervention group
Baseline values control group

Results in intervention group
absent: 68%, fluctuated markedly 28%, worsened or consistently severe 4%

Results in control group
Steadily improved or absent: 43%, fluctuated markedly 36%, worsened or consistently severe 21%

Comments
p=0.05
Outcome measured
Relapses

Baseline values intervention group
Baseline values control group

Results in intervention group
None:36%, 1/2:12%, 3/4 20%, 5+: 32%
Results in control group
None:7%, 1/2:11%, 3/4 21%, 5+: 61%

Comments
p=0.05
Outcome measured
CFS
Proportion that no longer meet UK criteria

Baseline values intervention group
Baseline values control group

Results in intervention group
52%
Results in control group
39%

Comments
p=0.415
Outcome measured
Status
Completely recovered

Baseline values intervention group
Baseline values control group

Results in intervention group
24%
Results in control group
5%
Comments
p=0.05
Study IDParticipantsInterventions/comparatorsWithdrawals and adverse events
Friedberg (1994)152

Study design
Controlled trial

Level of evidence
2−
Number: 44
Adults or children?: Not stated

Inclusion criteria: Not stated

Exclusion criteria:

Diagnosis/ case definition: CDC (1988)

Age: mean 35.7 in treatment group, 39.7 in control

% Female: 95.5% women in treatment group, 67.2 in control (p<0.02)

Duration of illness: 32.5 months in treatment group, 74 in control

Baseline functioning: Both groups had significantly elevated fatigue severity scores compared to depression control group (p<0.002)

Further details:
17/22 participants had a current psychiatric condition, major depression in 10 cases, 11/22 in control group had diagnosed psychiatric illness, major depression in 6 cases.
Patients recruited from neurology clinic and through local CFS support group. No significant differences between two groups with respect to demographic variables or severity of illness. Patients offered CBT those that refused assigned to no-treatment group
Not stated
CBT
Patients either treated with CFS or untreated CBT modelled for chronic pain, used group therapy format, structured on following interventions: shared coping, relxation training and guided imagery, cognitive therapy techniques, and behavioural presecription

Number of participants in each group
22 in treatment, 22 in control
Withdrawals: 2 patients who did not want CBT refused to participate in control group.

Adverse events: Not stated
Results
Outcome 1Outcome 2Outcome 3Outcome 4
Outcome measured: Depression
Depression symptom score. CES-D scale, 20 item self-report scale scored from 0–60

Baseline values intervention group
Baseline values control group

Results in intervention group
lower than pre-treatment score, p=0.058
Results in control group
No significant difference

Comments
Outcome measured
Stress symptom score
Brief symptom inventory, 53 item self-report scale

Baseline values intervention group
Baseline values control group

Results in intervention group
No significant difference
Results in control group
No significant difference

Comments
Outcome measured
Fatigue fatigue severity score, 9 items on 7 point
Likert scale

Baseline values intervention group
Baseline values control group

Results in intervention group
No significant difference
Results in control group
No significant difference

Comments
Outcome measured
Cognition
Fatigue related cognition scale, 14 item self-report scale developed by one of trial authors

Baseline values intervention group
Baseline values control group

Results in intervention group
Significant reduction, p<0.023
Results in control group
No significant difference

Comments
Additional comments: Those with higher CES-D scores at baseline improved more than those with low CES-D scores (median split), high scores improved in depression (p<0.001), stress (p<0.01), fatigue severity (p<0.05), and fatigue related thinking (p<0.04)
Study IDParticipantsInterventions/comparatorsWithdrawals and adverse events
Fulcher (1997)157

Study design
RCT

Level of evidence
1++
Number: 66
Adults or children?: Adults

Inclusion criteria: Patients excluded who had a current psychiatric disorder or symptomatic insomnia as assessed by DSM-III-R (Diagnostic and Statistical Manual of Mental Disorders, third edition, revised)

Exclusion criteria:

Diagnosis/case definition: Oxford

Age: mean = 37.2 (sd=10.7)

% Female: 74% women

Duration of illness: Median duration = 2.7 years (range 0.6 – 19 years)

Baseline functioning:

Further details:
Not stated
Mean BMI= 23.8 (sd=4.6). Twenty patients were taking full dose anti=-depressants, 10 were taking low dose tricyclic antidepressants as hypnotics, 44 patients blamed viruses for their illnesses
Physical screening investigations were carried out or, when appropriate, full recent records were obtained from referring doctors to ensure other disorders had been discounted.
GET
Patients randomly assigned to either graded aerobic exercise or flexibility treatment weekly for 12 weeks
Patients attended for supervised treatment and given next week’s exercise prescription, home exercise was prescribed on at least 5 days a week with initial sessions lasting between 5 & 15 mins with intensity of 40% of peak oxygen consumption (roughly 50% max heart rate), daily exercise prescription increased by 1 or 2 minutes up to a maximum of 30 minutes, intensity increased to 60% peak oxygen consumption, patients given heart rate monitors to ensure did not exceed level prescribed. Main exercise was walking but also encouraged to take other forms of exercise, advised not exceed prescribed exercise during a good phase, if patients complained of increased fatigue were advised to continue with same level of exercise for extra week and increase when fatigue had lessened. Control subjects were taught stretching routine and relaxation techniques building up to longer sessions like exercise group, specifically told to avoid doing any extra physical activities

Number of participants in each group
33 in each group
Withdrawals: 7 patients dropped out: 4 in exercise group and 3 in control, 1 from each group dropped out as said treatment made them worse

Adverse events:
Results
Outcome 1Outcome 2Outcome 3Outcome 4
Outcome measured: General health CGI-I scale. Self rated global impression change scores after treatment range from 1 (very much better), 2 (Much better), 3 (A little better), 4 (no change), 5 (a little worse), 6 (much worse) to 7 (very much worse)

Baseline values intervention group
Baseline values control group

Results in intervention group
1: 9 (31%); 2:7 (24%); 3:11 (38%); 41: (3%); 5: 1 (3%); 6:0; 7:0
Results in control group
1: 2 (7%); 2:6 (20%); 3:18 (60%); 4: 3 (10%); 5: 0; 6:1(3%); 7:0

Comments
Analysis by intention to treat showed that 17/33 patients improved with exercise and 9/33 improved with flexibility treatment (chi2=4.06, p=0.04)
Outcome measured
Physical
Physiological variables

Baseline values intervention group
Baseline values control group

Results in intervention group
Results in control group

Comments
Exercise group showed significant increase in: peak oxygen consumption and maximum ventilation but not in any other physiological measures compared to control.
Outcome measured
Symptom measure
Various symptomatic and functional measures

Baseline values intervention group
Baseline values control group

Results in intervention group
Results in control group

Comments
Chalder fatigue score, total fatigue score, physical fatigue score, SF36 total score, SF36 physical function score and SF-36 general health score were significantly better in the exercise than in the flexibility groups. No difference in mental fatigue score, depression score, anxiety score or sleep total score
Outcome measured

Baseline values intervention group
Baseline values control group

Results in intervention group
Results in control group

Comments
Study IDParticipantsInterventions/comparatorsWithdrawals and adverse events
Moss-Morris (2005)160

Study design
RCT

Level of evidence
1+
Number: 49
Adults or children?: Adults

Inclusion criteria: Aged between 18 and 65, meeting CFS criteria.

Exclusion criteria: Patients who were unable to undergo exercise testing for medical reasons or who were already taking part in a regular and consistent exercise programme.

Diagnosis/case definition: CDC 1994

Age: mean 36.7 years GET group, 45.5 yrs control group

% Female: 60% GET group, 79% control group

Duration of illness: median 2.67 years GET group, 5.00 years control group

Baseline functioning: 22% unemployed and unable to work due to disability

Further details: Recruited from specialist CFS private general practice in New Zealand. Around 25% of participants suffered from anxiety/depression.
12 week graded exercise programme versus standard care.

Programme consisted of CBT rationale, developing individual plan for exercise programme, target heart rate set at 40% VO2max, attained on treadmill test, to be maintained for 1–15 minutes 4–5 times per week. Researchers and participants met weekly to reassess goals.

Standard care included advice on managing diet, stress and CFS symptoms.

Number of participants in each group: 25 in GET group and 24 in control group
Withdrawals: 3/25 dropped out of treatment and 3/24 did not return follow-up questionnaires at 12 weeks

Adverse events: Not stated
Results
Outcome 1Outcome 2Outcome 3Outcome 4
Outcome measured: Clinical Global Impression

Results in intervention group
12/22 were much or very much better

Results in control group
5/21 were much or very much better

Comments
Statistically significant difference (p=0.04)
Outcome measured: Physical fatigue (14 item fatigue scale)

Baseline values intervention group
14.55 (5.40)
Baseline values control group
14.61 (4.86)

Results in intervention group
7.91 (7.06)
Results in control group
14.27 (5.75)

Comments
p=0.02
Outcome measured: Mental fatigue (14 item fatigue scale)

Baseline values intervention group
9.90 (3.74)
Baseline values control group
10.74 (3.90)

Results in intervention group
6.00 (4.06)
Results in control group
10.14 (4.27)

Comments
p=0.03
Outcome measured: total fatigue score (14 item fatigue scale)

Baseline values intervention group
24.45 (8.79)
Baseline values control group
25.35 (8.05)

Results in intervention group
13.91 (10.88)
Results in control group
24.41 (9.69)

Comments
p=0.02
Outcome 5Outcome 6Outcome 7Outcome 8
Outcome measured

Baseline values intervention group
Baseline values control group
G
Results in intervention group
Results in control group

Comments
Outcome measured

Baseline values intervention group
Baseline values control group

Results in intervention group
Results in control group

Comments
Outcome measured

Baseline values intervention group
Baseline values control group

Results in intervention group
Results in control group

Comments
Outcome measured

Baseline values intervention group
Baseline values control group

Results in intervention group
Results in control group
Comments
Additional comments:
Study IDParticipantsInterventions/comparatorsWithdrawals and adverse events
Powell (2001)162

Study design
RCT

Level of evidence
1++
Number: 148
Adults or children?: Both

Inclusion criteria:
Exclusion criteria:
Diagnosis/case definition: Oxford

Age: mean 34 in group 1 & 2, 32 in group 3 & 4
% Female: % female: 24 group 1, 28 group 2, 33 group 3, 31 group 4
Duration of illness: Mean (months): 48.6 group 1, 51.2 group 2, 51.5 group 3, 55.0 group 4
Baseline functioning:
Further details:
Not stated
Not stated
Graded exercise and discussion of symptoms

Number of participants in each group
34 in control, 37 in group 2, 39 in group 3, 38 in group 4
Withdrawals: 21 dropped out, 19 in intervention groups, dropped out during treatment: 8 for medical reasons, 7 for psychiatric reasons, 4 gave no reason, 1 emigrated, 1 was dissatisfied with treatment

Adverse events: Not stated
Results
Outcome 1Outcome 2Outcome 3Outcome 4
Outcome measured: Physical functioning SF 36 (range 10–30, 30 is best functioning).

Baseline values intervention group
Group 2: 16.00 (14.99, 17.01) Group 3: 15.77 (14.57, 16.97), Group 4: 15.95 (14.84, 17.05)
Baseline values control group
Group 1: 16.32 (15.15, 17.50)

Results in intervention group
Group 2: 25.08 (23.34, 26.81), Group 3: 24.26 (22.54, 25.98), Group 4: 24.89 (23.35, 26.43)
Results in control group
Group 1: 16.94 (15.44, 18.44)

Comments
p<0.001 for each intervention group compared to control, no difference between interventions
Outcome measured
Fatigue
Measured on scale from 0–11, 11 is most severe

Baseline values intervention group
Group 2: 10.35 (9.98, 10.72), Group 3: 9.92 (9.22, 10.63), Group 4: 10.24 (9.85, 10.62)
Baseline values control group
Group 1: 10.61 (10.36, 10.88)

Results in intervention group
Group 2: 3.24 (1.78, 4.71), Group 3: 3.47 (2.05, 4.87), Group 4: 3.11 (1.84, 4.37)
Results in control group
Group 1: 10.06 (9.31, 10.81)

Comments
p<0.001 for each intervention group compared to control, no difference between interventions
Outcome measured
Depression
Measured on HAD scale: range 0–21, >10 = clinical depression

Baseline values intervention group
Group 2: 9.27 (8.03, 10.51), Group 3: 9.03 (7.81, 10.24), 9.03 (7.84, 10.21)
Baseline values control group
Group 1: 10.35 (8.93, 11.78)

Results in intervention group
Group 2: 4.24 (3.00, 5.49), Group 3: 4.62 (3.22, 6.01), Group 4: 4.21 (2.92, 5.50)
Results in control group
Group 1: 10.06 (8.39–11.72)

Comments
No measure of significance presented
Outcome measured
Anxiety
Measured on HAD scale as outcome 3

Baseline values intervention group
Group 2: 10.62 (9.13, 12.12), Group 3: 10.03 (8.40, 11.65), Group 4: 10.21 (8.75, 11.67)
Baseline values control group
Group 1: 11.18 (9.55, 12.80)

Results in intervention group
Group 2: 7.14 (5.79, 8.48), Group 3: 6.51 (5.13, 7.90), Group 4: 7.71 (6.14, 9.29)
Results in control group
Group 1: 10.06 (8.40–11.72)

Comments
No measure of significance presented
Outcome 5Outcome 6Outcome 7Outcome 8
Outcome measured
Sleep
Sleep problems measured on scale of Jenkins et al, range 0–20, 20 indicated maximum problems

Baseline values intervention group
Group 2: 12.43 (10.82, 14.05), Group 3: 13.54 (12.10, 14.97), Group 4: 13.03 (11.39, 14.66)
Baseline values control group
Group 1:12.79 (11.13, 14.45)

Results in intervention group
Group 2: 6.70 (4.98, 8.43), Group 3: 8.56 (6.80, 10.33), Group 4: 7.13 (5.55, 8.71)
Results in control group
Group 1: 11.53 (9.67–13.39)

Comments
No measure of significance presented
Outcome measured
Improvement
Clinically significant improvement as assessed by authors

Baseline values intervention group
Baseline values control group

Results in intervention group
Group 2: 26/37, Group 3: 27/39, Group 4: 26/38
Results in control group
Group 1: 2/34

Comments
p<0.001 using a chi-squared test
Outcome measured
Improvement
Patients report of being very much or much better

Baseline values intervention group
Baseline values control group

Results in intervention group
84%
Results in control group
12%

Comments
No measure of significance presented
Outcome measured

Baseline values intervention group
Baseline values control group

Results in intervention group
Results in control group
Comments
Additional comments: Results given are at 12 month follow-up. Also presented results after 3 and 6 months. Results presented as mean (95% CI). Patients rated physiological explanations offered for their symptoms as very important.
Study IDParticipantsInterventions/comparatorsWithdrawals and adverse events
Powell (2004)159

Study design
RCT

Level of evidence
1++
Number: 148
Adults or children?: Both

Inclusion criteria: Patients aged 15–55, scored <25 on physical functioning subscale of SF36. Excluded if undergoing further physical investigations or other treatments including antidepressant therapy, had psychotic illness, somatisation disorder, eating disorder or history of substance abuse, if confined to wheelchair or bed

Exclusion criteria:

Diagnosis/case definition: Oxford

Age: mean 34 in group 1 & 2, 32 in group 3 & 4

% Female: % female: 24 group 1, 28 group 2, 33 group 3, 31 group 4

Duration of illness: Mean (months): 48.6 group 1, 51.2 group 2, 51.5 group 3, 55.0 group 4

Baseline functioning: Between 11 and 15% were working, 15–17% were receiving disability benefits, 3–10% were taking antidepressants, 17–20% believed in physical cause of illness

Further details:
not stated
Recruited from consecutive referrals to CFS and infectious diseases clinic. Randomisation was stratified by scores on HAD depression scale
Same study as Powell 2001162, followed up at 2 years. 32 patients from the control group were offered the intervention after 1 year and assessed 1 year later
Graded exercise and discussion of symptoms
Group 1: standardised medical care, given pack without medical explanation but which engouraged regular activity and positive thinking.
Group 2 (minimum education): patients received 2 individual treatment sessions over 2 weeks, causal explanations given for symptoms, graded exercise programme designed for each patient, given comprehensive educational pack, followed up with phone calls at 3 and 6 months. Group 3 (telephone intervention): same as group 2 but also received 7 planned telephone contacts lasting 30 mins each, rationale for treatment reiterated and problems with exercise discussed, Group 4 (maximum educational intervention): same as group 2 but also received 7 one hour face-to-face treatment sessions, similar to phone calls.

Number of participants in each group
34 in control, 37 in group 2, 39 in group 3, 38 in group 4
Withdrawals:

Adverse events:
Results
Outcome 1Outcome 2Outcome 3Outcome 4
Outcome measured: Physical functioning SF 36 (range 10–30, 30 is best functioning).

Baseline values intervention group
Group 2: 16.00 (14.99, 17.01) Group 3: 15.77 (14.57, 16.97), Group 4: 15.95 (14.84, 17.05)
Baseline values control group
Group 1: 16.32 (15.15, 17.50)

Results in intervention group
Group 2 24.11 (5.94) Group 3: 23.64 (6.39)
Group 4: 25.45 (4.72)
Results in control group
Group 1: not reported

Comments
no significant difference between groups
Outcome measured
Fatigue score
Measured on scale from 0–11, 11 is most severe

Baseline values intervention group
Group 2: 10.35 (9.98, 10.72), Group 3: 9.92 (9.22, 10.63), Group 4: 10.24 (9.85, 10.62)
Baseline values control group
Group 1: 10.61 (10.36, 10.88)

Results in intervention group
Group 2: 4.46 (4.78) Group 3: 3.59 (4.69) Group 4: 2.84 (3.67)
Results in control group
Group 1, not reported

Comments
no significant difference between groups
Outcome measured
Depression score
Measured on HAD scale: range 0–21, >10 = clinical depression

Baseline values intervention group
Group 2: 9.27 (8.03, 10.51), Group 3: 9.03 (7.81, 10.24), 9.03 (7.84, 10.21)
Baseline values control group
Group 1: 10.35 (8.93, 11.78)

Results in intervention group
Group 2: 5.11 (5.12) Group 3: 4.77 (4.67)
Group 4: 4.08 (4.33)
Results in control group
Group 1: not reported

Comments
no measure of significance presented
Outcome measured
Anxiety
Measured on HAD scale as outcome 3

Baseline values intervention group
Group 2: 10.62 (9.13, 12.12), Group 3: 10.03 (8.40, 11.65), Group 4: 10.21 (8.75, 11.67)
Baseline values control group
Group 1: 11.18 (9.55, 12.80)

Results in intervention group
Group 2: 7.65 (4.78) Group 3: 7.03 (5.07);
Group 4: 7.13 (4.47
Results in control group
Group 1: not reported

Comments
no measure of significance presented
Outcome 5Outcome 6Outcome 7Outcome 8
Outcome measured
Sleep
Sleep problems measured on scale of Jenkins et al, range 0–20, 20 indicated maximum problems

Baseline values intervention group
Group 2: 12.43 (10.82, 14.05), Group 3: 13.53 (12.10, 14.97), Group 4: 13.03 (11.39, 14.66)
Baseline values control group
Group 1:12.79 (11.13, 14.45)

Results in intervention group
Group 2: 7.62 (5.30) Group 3: 8.15 (5.59)
Group 4: 7.92 (5.50)
Results in control group
Group 1: not reported

Comments
no measure of significance presented
Outcome measured
Clinical Global Impression Scale

Baseline values intervention group
Baseline values control group

Results in intervention group
70/90 (78%) reported being much better or very much better at 2 years
Results in control group

Comments
Outcome measured
Clinically significant outcome improvement assessed by authors (relates to SF36 physical functioning scores)

Baseline values intervention group
Baseline values control group

Results in intervention group
Group 2: 20 Group 3: 23 Group 4: 26
Results in control group

Comments
Outcome measured
No longer fulfilled trial criteria

Baseline values intervention group
Baseline values control group

Results in intervention group
Group 2: 17 Group 3: 22 group 4: 24
Results in control group
Comments
Additional comments: At the end of the trial 14/30 patients who crossed over at one year from the control group achieved a clinically significant outcome. 7/30 no longer met trial criteria and 17/25 who completed the educational intervention reported being much better or very much better.
Study IDParticipantsInterventions/comparatorsWithdrawals and adverse events
Prins (2001)151

Study design
RCT

Level of evidence
1++
Number: 270
Adults or children?: Adults

Inclusion criteria: Aged 18–60, no previous or current engagement in CFS research, not pregnant or engaged in pregnancy stimulating techniques and living within one and a half gours travelling time of the 3 centres. Patients in CFS group could not undergo further medical examinations of other treatments for CFS during study period

Exclusion criteria:

Diagnosis/case definition: CDC (1994)

Age: Mean (sd): CBT 36.2 (9.4), Support: 37.1 (10.6), control: 36.7 (10.3)

% Female: 19–24% female

Duration of illness: Mean (sd) years: CBT: 4.9 (4.8), support: 6.6 (6.4), control: 5.3 (5.4)

Baseline functioning: Not stated

Further details:
Not stated
Recruited from outpatient clinics at departments of internal medicine
Participants did not have to meet the CDC criteria of 4/8 additional symptoms. Score of 40+ on subscale fatigue severity of Checklist of individual strength and score of 800+ of Sickness Impact Profile
CBT
CBT group: 16 sessions of 1 hour over 8 months, basic elements cognitive restructuring, building up activity, returning to work and relapse prevention
Guided support groups: 11 group meetings of one and a half-hours during 8 months, treatment orientation non-directive and client-centered.
Natural course (control): no interventions offered and no further requirements, patients could attend other examinations or treatments

Number of participants in each group
92 in CBT group, 90 in support group, 88 in no treatment
Withdrawals: 6 patients excluded (not included in overall number): 5 developed other diseases during trial, one was pregnant at pre-test. 2 patients did not meet criteria for CFS due to pre-morbid anorexia nervosa. 37 in CBT group, 29 in support group and 18 in control group dropped out. 10 patients in CBT did not start treatment, 8 in support group did not start. 23 CBT group, 17 support group and 9 control group stopped treatment. During follow-up 4 in CBT, 4 in support and 9 in control group dropped out (dropped out of treatment or did not attend assessments)

Adverse events: Not stated, but very large number of drop-outs
Results
Outcome 1Outcome 2Outcome 3Outcome 4
Outcome measured: Fatigue CIS fatigue score. Results presented as change from baseline to follow-up and mean (SE). Results presented on ITT basis

Baseline values intervention group
Baseline values control group

Results in intervention group
CBT: −11.8 (1.4)
Results in control group
Support: −6.5 (1.2), Control: −6.6 (1.0)

Comments
Outcome measured
Psychological well-being
Measured on SCL90. Results presented as mean(sd). Results presented on ITT basis

Baseline values intervention group
CBT: 170 (38.5)
Baseline values control group
Support: 169 (41.5), Control: 166 (36.0)

Results in intervention group
CBT: 138 (35.1)
Results in control group
Support: 153 (33.9), Control: 147 (32.8)

Comments
F=4.96, p=0.001 for differences between groups (group x time)
Outcome measured
Quality of life
Measured on EuroQol scale. Results presented on ITT basis

Baseline values intervention group
CBT: 46 (17)
Baseline values control group
Support 43 (16), Control: 40(14)

Results in intervention group
CBT: 57 (22)
Results in control group
Support: 44 (19), Control: 49 (19)

Comments
F=3.92, p=0.004 for differences between groups (group x time)
Outcome measured
Work
Number of hours at work during 12 days.
Results presented on ITT basis

Baseline values intervention group
CBT: 16.3 (21.1)
Baseline values control group
Support: 12.8 (19.1), Control: 13.5 (18.6)

Results in intervention group
CBT: 23.1 (28.1)
Results in control group
Support: 11.0 (15.4), Control: 16.8 (21.8)

Comments
F=2.60, p=0.036 for differences between groups (group x time)
Outcome 5Outcome 6Outcome 7Outcome 8
Outcome measured
Fatigue
Proportion of participants with a clinically significant improvement in fatigue on CIS fatigue score

Baseline values intervention group
Baseline values control group

Results in intervention group
CBT: 20/58=35%
Results in control group
Support: 8/62=13%, Control: 13/76=17%

Comments
p=0.009 comparing CBT to support and 0.026 comparing CBT to control
Outcome measured
Functional
Proportion of participants with a clinically significant improvement in Karnofsky score

Baseline values intervention group
Baseline values control group

Results in intervention group
CBT: 28/57=49%
Results in control group
Support: 12/62=19%. Control: 17/75=23%

Comments
p=0.001 comparing CBT to support and 0.001 comparing CBT to control
Outcome measured
Improvement
Proportion of participants with self-rated improvement

Baseline values intervention group
Baseline values control group

Results in intervention group
CBT: 29/58=50%
Results in control group
Support: 9/62=15%, Control: 24/76=32%

Comments
p<0.001 comparing CBT to support and 0.034comparing CBT to control
Outcome measured
Functional Impairment
Measured using Sickness Impact Profile.
Results presented as change from baseline to follow-up and mean (SE). Results presented on ITT basis

Baseline values intervention group
Baseline values control group

Results in intervention group
CBT: −590 (80)
Results in control group
Support: −320 (80), Control: −390 (80)
Comments
Additional comments: All results presented are at follow-up after 14 months. Results also presented at post-test (8 months), similar to follow-up so not presented here. In CBT group predictors for post-test fatigue severity were pre-test score, type of activity pattern and focusing on bodily symptoms (R2=20)
Study IDParticipantsInterventions/comparatorsWithdrawals and adverse events
Sharpe (1996)150

Study design
RCT

Level of evidence
1+
Number: 60
Adults or children?: Adults

Inclusion criteria: Consecutive patients aged 18–60, with major complaint of fatigue. Patients excluded if currently receiving psychotherapy or antidepressant drugs (unless taking same dose for at least 3 months without improvement), were unwilling to accept randomisation or unavailable for follow-up, met criteria for severe depression or had histroy of bipolar affective disorder, schizophrenia, or substance misuse or were at significant risk of suicide or in need of urgenet psychiatric treatment

Exclusion criteria:

Diagnosis/case definition: Oxford

Age: 18–60

% Female: M:F: 12:18 in CBT group, 7:23 in standard care group

Duration of illness: In months: Median 17 in CBT group, 20 in control, mean 33.6 in CBT, 29.7 in control, range 6–91 months

Baseline functioning: Groups did not differ on functional impairment, or psychiatric diagnoses. Patients in CBT group spent more days in bed (3.3 vs 1.6), and fewer were actively employed.

Further details:
Not stated
Treatment groups did not differ substantially with respect to age, sex, educational level, marital status. 20% reported infection onset in CBT group, 22% in control
Also fulfilled CDC 94 criteria
CBT
Medical care alone compared with medical care plus CBT Patients with medical care alone told to increase their level of activity as much as they felt able, and reassured that hthere was no organic cause. CBT group given 16 1 hour individual sessions over 4 monthsFinal assessment was at 12 months.

Number of participants in each group
30 in each group
Withdrawals: Complete data not available for one patiend, did not attend 12 month follow-uo. Phone call to patient indicated no substantial change since previous evlauation, so these data used for both. 7 patients (3 in CBT group) refused to do walking test on one or more occasions so previous test results used.

Adverse events: 2 participants in CBT group attributed deterioration in symptoms to treatment
Results
Outcome 1Outcome 2Outcome 3Outcome 4
Outcome measured: Functioning
Proportion of patients with normal functioning at 12 months follow-up (achieved Karnofsky score of 80 or more)

Baseline values intervention group
Baseline values control group

Results in intervention group
73%
Results in control group
27%

Comments
Difference in proportion = 47 (95% CI: 24–69) %, p<0.001, difference increased over time
Outcome measured
Functioning
Proportion of patients with at least 10 point improvement on Karnofsky scale at 12 months follow-up

Baseline values intervention group
Baseline values control group

Results in intervention group
73%
Results in control group
23%

Comments
Difference in proportion = 50 (95% CI: 28–72) %, p<0.001, difference increased over time
Outcome measured
Work status
Improvement in work status

Baseline values intervention group
Baseline values control group

Results in intervention group
63%
Results in control group
20%

Comments
Outcome measured
Global improvement
Proportion of patients reporting much improved or very much improved, or worse or very much worse, measured on CGI scale (7 point patient rated scale)

Baseline values intervention group
Baseline values control group

Results in intervention group
Imprved: 60%, Deteriorated: 13%
Results in control group
Improved: 23%, Deteriorated: 10%%

Comments
Outcome 5Outcome 6Outcome 7Outcome 8
Outcome measured
Illness beliefs
Proportion of patients reporting reduction in strength of illness beliefs, measured on Likert type scales

Baseline values intervention group
Baseline values control group

Results in intervention group
Illness mainly physical:33%, cause is a virus, 48%, illness is ME 17%, avoidance of exercise 60%
Results in control group
Illness mainly physical:7%, cause is a virus, 20%, illness is ME 27%, avoidance of exercise 30%

Comments
All differences in proportions were significant (p<0.05), exetp for the belief that illness is ME
Outcome measured: Activitites
Percentage interference with activities

Baseline values intervention group
65
Baseline values control group
64

Results in intervention group
50
Results in control group
37

Comments
Difference in change between the groups = 14(95% CI: 3 to 25), p<0.05
Outcome measured
Rest
Number of days in bed per week

Baseline values intervention group
3.3
Baseline values control group
1.6

Results in intervention group
0.9
Results in control group
2.0

Comments
Difference in change between the groups = 2.8(95% CI: 1.7 to 4.0), p<0.05
Outcome measured
Exercise
Distance walked in 6 minutes (m)

Baseline values intervention group
437
Baseline values control group
435

Results in intervention group
481
Results in control group
424

Comments
Difference in change between the groups = 55(95% CI: 17 to 94), p<0.05
Outcome 9Outcome 10Outcome 11
Outcome measured
Fatigue
Fatigue severity, graded 0–10

Baseline values intervention group
7.8
Baseline values control group
7.9

Results in intervention group
4.3
Results in control group
6.3

Comments
Difference in change between the groups = 1.9(95% CI: 0.5 to 3.3), p<0.05
Outcome measured
Anxiety
Measured on hospital anxiety and depression scale

Baseline values intervention group
6.3
Baseline values control group
8.4

Results in intervention group
4.4
Results in control group
6.8

Comments
Difference in change between the groups = 0.3(95% CI: −1.6 to 2.2), p>0.05
Outcome measured
Depression
Measured on hospital anxiety and depression scale

Baseline values intervention group
6.7
Baseline values control group
6.8

Results in intervention group
3.6
Results in control group
5.8

Comments
Difference in change between the groups = 2.0 (95% CI: 0.0 to 4.1), p<0.06
Additional comments: All results presented are after 12 months follow-up
Study IDParticipantsInterventions/comparatorsWithdrawals and adverse events
Stulemeijer (2004)218

Study design
RCT

Level of evidence
1−
Number: 69
Adults or children?: Children

Inclusion criteria: All consecutive patients with major complaint of fatigue aged 10 to 17.2 years, meeting CDC 1994 criteria, referred to outpatient clinic between 1999 and 2002

Exclusion criteria: Patients with psychiatric comorbidity.

Diagnosis/case definition: CDC (1994)

Age: mean 15.6 yrs CBT, 15.7 yrs control

% Female: 89% CBT, 91% controls

Duration of illness: median 16 months CBT, 18 months controls

Baseline functioning: Fatigue severity (checklist individual strength) CBT 52.5 (3.8), control 51.6 (4.3). Physical functioning (SF36) CBT 42.1 (16.5), control 45.3 (17.0). Full school attendance CBT 4/35, control 6/34

Further details:
not stated
10 in CBT group and 7 in control group had a passive activity pattern (spend most time lying down and go out infrequently)
Detailed history and physical and laboratory examinations were undertaken. Severe fatigue and severe functional impairment were defined as a score of 40 or more on the fatigue severity subscale of the checklist individual strength and a weighted score of 65 or less on the SF36 physical functioning subscale.
CBT
Ten individual sessions over 5 months. For relatively active patients, treatment began with recognition and acceptance of limitations and reduction of activity. Activity levels were then increased. For inactive patients, a programme of activity building was started as soon as possible. Parents were involved in both CBT groups and return to full time education was a goal. Control group = waiting list for CBT

Number of participants in each group
36 CBT, 35 waiting list
Withdrawals: 6 patients dropped out during treatment. 7 were missing from CBT group and 2 from control group at final assessment

Adverse events: none reported
Results
Outcome 1Outcome 2Outcome 3Outcome 4
Outcome measured: Fatigue severity checklist individual strength

Baseline values intervention group
52.5 (3.8)
Baseline values control group
51.6 (4.3)

Results in intervention group
30.2 (16.8). Treatment effect = 14.5 (95% CI: 7.4, 21.6), p=0.001
Results in control group
44.0 (13.4)

Comments
Outcome measured
Physical functioning
SF-36

Baseline values intervention group
42.1 (16.5)
Baseline values control group
45.3 (17.0)

Results in intervention group
69.4 (28.0). Treatment effect 17.3 (95% CI: 6.2, 28.4), p=0.003
Results in control group
55.3 (21.1)

Comments
Outcome measured
School attendance hours attended/hours that should have been attended

Baseline values intervention group
46.2 (38.9)
Baseline values control group
56.4 (38.6)

Results in intervention group
74.7 (37.8). Treatment effect 18.2 (95% CI: 0.8, 35.5), p=0.040
Results in control group
66.7 (36.0)

Comments
Outcome measured
Additional symptoms
unrefreshing sleep; muscle pain; impaired concentration; tiredness after exercise; headache; impaired memory; multijoint pain; sore throat; sensitive lymph nodes. Rated on a 4 point Likert scale.

Baseline values intervention group
Baseline values control group

Results in intervention group
Results in control group

Comments
At 5 months in CBT group, significantly greater decrease in prevalence of: feeling ill after exercise, impaired concentration, unrefreshing sleep, muscle pain, headache.
Study IDParticipantsInterventions/comparatorsWithdrawals and adverse events
Taylor (2004)153

Study design
RCT

Level of evidence
1+
Number: 47
Adults or children?: Adults

Inclusion criteria: see diagnosis details

Exclusion criteria: exclusionary medical conditions (e.g. hyperthyroidism)

Diagnosis/case definition: CDC (1994)

Age: mean 49.0 yrs immediate, 44.9 yrs delayed programme

% Female: 91% immediate, 100% delayed group

Duration of illness: not stated

Baseline functioning: not stated

Further details:
none stated
Recruited from local CFS self-help groups, Chicago area physicians specialising in CFS treatment, advertisements on CFS newsletters, Chicago-area newspapers, CFS web sites and listservs and local TV Screening process to confirm self-diagnosis: CFS screening questionnaire (Jason et al 1997111), semistructured psychiatric interview (SCID for DSM-IV), collection of medical records documenting CFS diagnosis
Rehabilitation programme Integrative consumer driven rehabilitation programme consisting of: 8 sessions of illness-management group (bi-weekly over 4 months), followed by 7 months of peer counselling, focusing on goal attainment control group received ‘delayed’ programme (assume usual care while waiting?)

Number of participants in
each group
23 immediate programme, 24 delayed programme
Withdrawals: none

Adverse events: none reported
Results
Outcome 1Outcome 2Outcome 3Outcome 4
Outcome measured: Symptom severity
CFS symptom rating form (Jason et al 1997111): 0 no problem, 100 severe problem

Baseline values intervention group
15.1 (3.0)
Baseline values control group
14,2 (2.8)

Results in intervention group
after group phase 14.4 (3.5); after one on one phase 13.9 (3.5)
Results in control group
after group phase: 14.3 (2.7); after one on one phase 14.8 (2.8)

Comments
significant interaction p<0.05
Outcome measured
Overall quality of life
Quality of Life Index, final scores 0–30, higher scores indicate higher life quality

Baseline values intervention group
13.1 (4.3)
Baseline values control group
14.0 (3.9)

Results in intervention group
after group phase 13.2 (3.8); after one on one phase 15.7 (3.7)
Results in control group
after group phase: 14.6 (4.8); after one on one phase 14.6 (4.1)

Comments
significant interaction p<0.05
Outcome measured
QoL health and functioning

Baseline values intervention group
12.9 (1.6)
Baseline values control group
13.1 (1.7)

Results in intervention group
after group phase 12.8 (1.8); after one on one phase 14.1 (1.7)
Results in control group
after group phase 13.6 (2.1); after one on one phase 13.6 (1.8)

Comments
Outcome measured
QoL social and economic

Baseline values intervention group
15.0 (1.2)
Baseline values control group
15.4 (0.7)

Results in intervention group
after group phase 15.2 (0.8); after one on one phase 15.6 (0.8)
Results in control group
after group phase 15.5 (1.0); after one on one phase 15.5 (0.9)

Comments
Outcome 5Outcome 6Outcome 7Outcome 8
Outcome measured
QoL psychological and spiritual

Baseline values intervention group
15.0 (1.2)
Baseline values control group
15.0 (1.1)

Results in intervention group
after group phase 15.0 (1.1); after one on one phase 15.5 (1.1)
Results in control group
after group phase 15.2 (1.3); after one on one phase 15.1 (1.2)

Comments
Outcome measured
QoL family

Baseline values intervention group
15.4 (0.9)
Baseline values control group
15.7 (1.0)

Results in intervention group
after group phase 15.4 (1.0); after one on one phase 15.6 (0.8)
Results in control group
after group phase 15.5 (1.0); after one on one phase 15.5 (0.9)

Comments
significant interaction p<0.05
Outcome measured

Baseline values intervention group
Baseline values control group

Results in intervention group
Results in control group

Comments
Outcome measured

Baseline values intervention group
Baseline values control group

Results in intervention group
Results in control group
Comments
Additional comments: Linear growth models were estimated comparing programme and control conditions for each outcome using random effects regression analyses
Study IDParticipantsInterventions/comparatorsWithdrawals and adverse events
Viner (2004)217

Study design
Controlled trial

Level of evidence
2+
Number: 56
Adults or children?: Children

Inclusion criteria: children (aged 9–17) who met CDC diagnostic criteria

Exclusion criteria: Those with treatable medical causes of fatigue, those on drugs known ot cause fatigue, those in whom somatoform disorder or school refusal was considered to be the diagnosis

Diagnosis/case definition: CDC (1994)

Age: mean 13.9 yrs rehab, 14.4 yrs supportive care group

% Female: 58% rehab, 59% supportive care

Duration of illness: mean 25.7 months rehab, 28.1 months supportive care

Baseline functioning: Of 78 children in the initial assessment, 62% had severe CFS, 29% moderate and 9% minimal CFS/ME

Further details:
a depressed mood was noted in 33 of the 78 children initially assessed
Young people with CFS/ME seen by joint GOSH and UCL Hospitals Adolescent Medicine service between June 1998 and December 2002.
modified for use with children and adolescents by using a three month duration of fatigue. Severity defined as follows: minimal: wellness score and school attendance both >= 75%, moderate: either or both scores >=50% but <75%, severe: either or both scores <50%.
Rehabilitative treatment Outpatient rehabilitative treatment (supportive care plus graded activities/exercise programme and family sessions) compared with supportive care alone. Followed up for 3–24 months.

Number of participants in each group
26 rehabilitation group, 22 supportive care alone
Withdrawals: data was available on outcome of treatment in 56 (remainder too early in follow-up). 22 had supportive care and 26 entered rehabilitation programme. The remaining 8 were prescribed SSRI either with supportive care or the programme, and they are not included in the analysis.

Adverse events: none reported.
Results
Outcome 1Outcome 2Outcome 3Outcome 4
Outcome measured: Global wellness score Self-rated - asked to provide an average score for the previous month on a scale between 100 (best) and 0 (worst)

Baseline values intervention group
50
Baseline values control group
50

Results in intervention group
85 (p<0.01)
Results in control group
67

Comments
Mean change in wellness score: 31% controls, 71% treatment group (p<0.05)
Outcome measured
School attendance
Average school attendance in the previous three months.

Baseline values intervention group
20
Baseline values control group
40

Results in intervention group
90 (p<0.05)
Results in control group
40

Comments
change in school score 25% control, 182% rehabilitation group (p<0.01). Those in the programme had higher school attendance from nine months after beginning treatment, with this difference reaching significance after 12 months (p=0.02).
Outcome measured
CFS severity

Baseline values intervention group
Baseline values control group

Results in intervention group
43% resolved (35% of severe cases resolved)
Results in control group
4.5% resolved (64% of severe cases remained severe)

Comments
Resolution defined as wellness score >=90% and school attendance of >=95%
Outcome measured

Baseline values intervention group
Baseline values control group

Results in intervention group
Results in control group

Comments
Additional comments: No major individual factors were associated with response to treatment, however improvement in wellness score in the whole group was correlated with older age at onset (p<0.05) and shorter duration of illness (p<0.001) irrespective of treatment.
Study IDParticipantsInterventions/comparatorsWithdrawals and adverse events
Wallman (2004)161

Study design
RCT

Level of evidence
1−
Number: 61
Adults or children?: Adults

Inclusion criteria: Aged between 16 and 74 years, diagnosed with CFS (CDC 1994 criteria).

Exclusion criteria:

Diagnosis/case definition: CDC (1994)

Age: 16–74 years (mean not reported)

% Female: 84% graded exercise, 69% controls

Duration of illness: not stated

Baseline functioning:

Further details:
six diagnosed with major depressive disorder in the previous 12 months. Not stated which intervention group they were in.
Recruited from notices placed in medical surgeries and advertisements in local newspapers Written confirmation of CFS diagnosis provided by doctor
Graded exercise with pacing
Initial exercise 5–15 mins based on mean HR during submaximal exercise tests. Walking, cycling or swimming. Instructed to exercise every 2nd day unless they had a relapse (exercise reduced instead). Duration 12 weeks.
comparator: relaxation/flexibility therapy every second day over 12 weeks

Number of participants in each group
32 graded exercise, 29 controls
Withdrawals: One excluded after randomisation because BMI too high to participate in exercise test. None reported during the study

Adverse events: none reported
Results
Outcome 1Outcome 2Outcome 3Outcome 4
Outcome measured: Resting heart rate and blood pressure

Baseline values intervention group
75 (71–78) bpm, 79 (76–82)/117 (112–121) mmHg
Baseline values control group
74 (70–78) bpm, 80 (76–84)/119 (114–124) mmHg

Results in intervention group
72 (69–75) bpm, 74 (71–76)/112 (108–116) mmHg
Results in control group
74 (70–78) bpm, 76 (74–79)/120 (115–125) mmHg

Comments
comparisons seem to have been made within groups rather than between groups
Outcome measured
Exercise test values
Oxygen uptake (mL/kg/min), Respiratory exchange ratio, net blood lactate production (mmol/L)

Baseline values intervention group
15.6 (13.3 – 17.7), 0.97 (0.93 – 1.01), 1.7 (1.4 – 1.9)
Baseline values control group
15.8 (13.7 – 17.9), 0.98 (0.94 – 1.02), 1.6 (1.4 – 1.9)

Results in intervention group
17.1 (14.9 – 19.2), 1.03 (0.99 – 1.06), 1.8 (1.5 – 2.1)
Results in control group
14.4 (12.4 – 16.4), 1.00 (0.96 – 1.04), 1.4 (1.1 – 1.7)

Comments
comparisons seem to have been made within groups rather than between groups
Outcome measured
Achievement of target heart rate during the exercise test

Baseline values intervention group
70%
Baseline values control group
64%

Results in intervention group
74%
Results in control group
53%

Comments
Outcome measured
Psychological results
HADS depression, HADS anxiety, mental fatigue, physical fatigue

Baseline values intervention group
6.5 (5.3 – 7.6), 7.3 (5.8 – 8.7), 6.3 (5.6 – 7.0), 11.6 (10.1– 13.0)
Baseline values control group
7.1 (5.9 – 8.2), 8.7 (7.5 – 9.9), 5.6 (5.0 – 6.1), 11.4 (10.4 – 12.3)

Results in intervention group
4.8 (3.6 – 5.9), 5.7 (4.4 – 6.9), 4.5 (3.9 – 5.2), 8.1 (6.9 – 9.4)
Results in control group
6.5 (5.5 – 7.6), 7.8 (6.5 – 9.2), 4.8 (4.2 – 5.5), 9.6 (8.3 – 10.9)

Comments
scores significantly lower in the exercise group (p=0.027)
Outcome 5Outcome 6Outcome 7Outcome 8
Outcome measured
Cognitive results
Stroop test 82 questions; Stroop test 95 questions

Baseline values intervention group
73.7 (68.0 – 79.3), 80.1 (73.1 – 87.0)
Baseline values control group
70.0 (61.3 – 78.9), 75.8 (64.6 – 87.0)

Results in intervention group
79.4 (78.0 – 80.8), 87.5 (81.4 – 93.6)
Results in control group
71.1 (63.3 – 78.9), 73.1 (60.3– 85.9)

Comments
significantly in favour of the exercise group on the more difficult level of the test (p=0.029)
Outcome measured
Clinical global impression
self-rated

Baseline values intervention group
Baseline values control group

Results in intervention group
5 very much better, 14 much better, 10 a little better, 3 no change
Results in control group
2 very much better, 10 much better, 10 a little better, 6 no change, 1 a little worse

Comments
no significant difference between the two groups
Outcome measured

Baseline values intervention group
Baseline values control group

Results in intervention group
Results in control group

Comments
Outcome measured

Baseline values intervention group
Baseline values control group

Results in intervention group
Results in control group
Comments
Study IDParticipantsInterventions/comparatorsWithdrawals and adverse events
Wearden (1998)158

Study design
RCT

Level of evidence
1++
Number: 136
Adults or children?: Adults

Inclusion criteria: Aged 18+. Pre-menopausal women required to take precautions against pregnancy. Excluded: those with schizophrenia, bipolar disorder, eating disorder, alcohol or illicit drug misuse, current suicidal ideation, history of ischaemic heart disease, inability to read and write English. Those on antidepressants underwent a 2 weeks washout.

Exclusion criteria:

Diagnosis/case definition: Oxford

Age: mean 38.7 (10.8)

% Female: 97 F 39 M

Duration of illness: median (!QR): 28.0 (39.5) months

Baseline functioning: 62 fulfilled DSM-III-R criteria for a current psychiatric diagnosis, 14 had major depression, 32 had either dysthymia or non-specific depressive disorder, 14 had various anxiety disorders and 2 had somatisation disorder.

Further details:
none stated
114 had changed their occupation. 35 were members of a self-help group.
GET & fluoxetine
  1. Fixed daily dose 20mg fluoxetine plus graded exercise.
  2. Graded exercise and placebo drug.
  3. Exercise control (activity diaries) and fluoxetine.
  4. Exercise control and placebo drug.
placebo controlled AND controlled for the amount of therapist contact. Treatment by physiotherapist on 8 occasions over 6 months. Graded exercise: subjects instructed to carry out preferred aerobic activity (walking/jogging, swimming or cycling) for 20mins at least 3x per week. Activity intensity initially set at a level which utilised oxygen at 75% of subject’s tested functional maximum. Exercise intensity was increased when there was a consistent recorded reduction of 10 beats per minute in post-exercise heart rate for one week and two points on the perceived exertion scale. Exercise control groups: subjects not offered specific advice on how much exercise to take but told to do what they could when they felt capable and rest when they felt they needed to. All trial participants kept activity diaries which were reviewed every 4 weeks.

Number of participants in each group
GET+F 33; GET+P 34
Withdrawals: 22 dropped out by 3 months and 40 by 6 months. More dropouts in exercise vs non-exercise groups (25/68 vs 15/69, p<0.05). No sig difference in dropout rates fluoxetine vs placebo (24/68 vs 16/69). 11 dropped out due to side effects (9 F, 2 P), 16 due to lack of efficacy (which groups not stated) and 13 for other reasons or no reason. Dropouts significantly more likely to be members of self help orgs (15/39 vs 20/95, ExP+F 35; ExP+P 34 p=0.04), have changed/given up job (38/40 vs 76/96, p=0.02) and have worse baseline scores on MOS health perception scale.

Adverse events: not stated: 11 dropped out due to them though.
Results
Outcome 1Outcome 2Outcome 3Outcome 4
Outcome measured: Fatigue
Chalder’s 14 item fatigue scale, self-rated questionnaire. Primary outcome = change in score and % of subjects scoring below case level on the fatigue scale.

Baseline values intervention group
Ex+P 33.7(33.0 to 36.9); Ex+F 35.9 (34.4 to 37.5); ExP+F 34.4(32.0 to 36.7)
Baseline values control group
ExP+P 34.0(32.3 to 35.7)

Results in intervention group
ex+P −5.7(−9.5 to −1.9); Ex+F −6.0(−9.7 to −2.3); ExP +F −3.0(−5.9 to −0.2)
Results in control group
ExP+P −2.7(−5.4 to 0.01)

Comments
there were trends for exercise to improve fatigue scale scores at wk12 (mean change 2.1(−0.6 to 4.8, p=0.13) and at wk26 (mean change 2.9(−0.2 to 6.1, p=0.07). Fluoxetine had no effect on fatigue scale at week 12 or wk26. At the beginning of the study no subjects in any group were in the non-case range for fatigue. At 26 weeks results were as follows: Ex+F 6, Ex+P 6, ExP+F 2, ExP+P 2. Interim (12wks) scores and non-case numbers are given in the paper but I haven’t extracted them - do you think I should?
Outcome measured
General health
MOS short form scales: physical function, role or occupation function, social function, social function, pain, health perceptions, mental health. Secondary outcome measure = change in score.

Baseline values intervention group
Baseline values control group

Results in intervention group
Results in control group

Comments
No significant changes on any MOS scale.
Values not reported.
Outcome measured
Depression
Hospital anxiety and depression scales (HAD). Secondary outcome = change in score.

Baseline values intervention group
Ex+F 9.4(3.6), Ex+P 8.5(2.9). ExP+F 9.1(4.2)
Baseline values control group
ExP+P 8.1(3.3)

Results in intervention group
Mean change: Ex+F −2.0(−3.3 to −0.7); Ex+P − 1.2(−2.5 to 0.2); ExP+F −1.7(−3.0 to −0.5)
Results in control group
Mean change ExP+P −1.3(−2.3 to −0.3)

Comments
no significant effects of exercise or fluoxetine on HAD scores at 26 weeks. IN complete analysis F reduced score at 12 weeks but in ITT analysis there were no differences. No effects of exercise on HAD case level of depression but fluoxetine treated group reduced from 13 to 5 with one new case arising. Placebo group cases reduced from 5 to 0 but 5 new cases arose.
Outcome measured
Physical
functional work capacity. Calculated as mL of oxygen consumed in the final minute of exercise per kg body weight.

Baseline values intervention group
Ex+F 23.1(9.3); Ex+P 19.9(6.5); ExP+F 22.7(8.7)
Baseline values control group
ExP+P 26.0(9.9)

Results in intervention group
mean change: Ex+F 2.0 (0.4 to 3.5); Ex+P 2.8(0.8 to 4.8); ExP+F 1.0(−0.9 to 3.0)
Results in control group
mean change ExP+P −0.1 (−1.7 to 1.6)

Comments
there was a significant effect of exercise on functional work capacity at week 26 (and at week12) n=132 mean change = 1.9(0.15 to 3.69) p=0.03. Fluoxetine had no significant effect on fwc at either time point.
Additional comments: 21 dropouts were reassessed at the end of the trial There was no worsening of scores on the fatigue scale, functional work capacity, HAD depression scale and MOS health perception scale.
Study IDParticipantsInterventions/comparatorsWithdrawals and adverse events
Whitehead (2002)163

Study design
RCT

Level of evidence
1−
Number: 65
Adults or children?: Adults

Inclusion criteria: All GP practices in 2 health authorities in NW England were eligible to take part.

Exclusion criteria: not stated

Diagnosis/case definition: CDC (1994)

Age: mean 36 yrs CBT, 41 yrs control

% Female: 54% CBT, 64% control

Duration of illness: mean 21 months CBT, 33 months control

Baseline functioning:

Further details:
not stated
Randomised by GP practice (35% of practices agreed to participate, 50% recruited participants), but analysis by participant only.
CBT
GP-delivered “brief CBT” consisting of patient information booklet (explanatory models of CFS), recoridng levels of activity and encouraging gradual increase at appropriate level and rate. Discussion of beliefs and behaviours around CFS. Control group: usual care (including referral to secondary care)

Number of participants in each group
26 CBT, 39 control group
Withdrawals: 2 patients moved away and 6 were found not to meet diagnostic criteria. At 6 months, follow up data was available for 18 people in the intervention group and 28 people in the control group. At 12 months data was available for 9 people in the intervention and 21 people in the control group.

Adverse events:
Results
Outcome 1Outcome 2Outcome 3Outcome 4
Outcome measured: Fatigue
11 item self-completion scale: score of >=3 indicates severe or disabling fatigue. Likert scoring, max fatigue = 33.

Baseline values intervention group
25.58
Baseline values control group
24.26

Results in intervention group
6 months 21.89, 12 months 19.11
Results in control group
6 months 20.04, 12 months 19.57

Comments
no significant difference between intervention and control groups
Outcome measured
Disability
London Handicap Scale (LHS) 0 (worst) to 100 (best)

Baseline values intervention group
58.25
Baseline values control group
62.77

Results in intervention group
6 months 65.03, 12 months 59.2
Results in control group
6 months 63.52, 12 months 65.62

Comments
no significant difference between intervention and control groups
Outcome measured
Anxiety and Depression
Hospital Anxiety and Depression Scale (HAD)

Baseline values intervention group
Baseline values control group

Results in intervention group
Results in control group

Comments
no significant differences between intervention and control groups
Outcome measured

Baseline values intervention group
Baseline values control group

Results in intervention group
Results in control group

Comments
Additional comments: 31 GPs who used the management package were asked about patients’ use of it. 21 GPs replied, of these, all but one person started to use the diaries. 5 patients used diaries for one month or less. Eight used diaries for 6 months or more and 4 for 12 months or more.

From: Appendix 1, Systematic Evidence Review to support the development of the NICE clinical guideline for CFS/ME in adults and children

Cover of Chronic Fatigue Syndrome/Myalgic Encephalomyelitis (or Encephalopathy)
Chronic Fatigue Syndrome/Myalgic Encephalomyelitis (or Encephalopathy): Diagnosis and Management of Chronic Fatigue Syndrome/Myalgic Encephalomyelitis (or Encephalopathy) in Adults and Children [Internet].
NICE Clinical Guidelines, No. 53.
National Collaborating Centre for Primary Care (UK).
Copyright © 2007, Royal College of General Practitioners.

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.