NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

O'Neil ME, Peterson K, Low A, et al. Suicide Prevention Interventions and Referral/Follow-Up Services: A Systematic Review [Internet]. Washington (DC): Department of Veterans Affairs (US); 2012 Mar.

Cover of Suicide Prevention Interventions and Referral/Follow-Up Services: A Systematic Review

Suicide Prevention Interventions and Referral/Follow-Up Services: A Systematic Review [Internet].

Show details

APPENDIX AAPEER REVIEW COMMENTS/AUTHOR RESPONSES

ReviewerCommentResponse
Question 1: Are the objectives, scope, and methods for this review clearly described?
1No. There was no discussion of the methods used for evaluating the strength of the evidence in a publication for drawing inferences about suicide prevention. There was a “boiler plate” discussion of the methods used for rating evidence. However, the draft did not provide adequate information about the way that this was applied for evaluating the strength of a paper for drawing inferences about suicide as an outcome, rather than for the primary outcome. More specifically, the literature reviewed included a number of papers reporting on studies conducted to evaluate treatments for other conditions or other outcomes (e.g., antidepressants for depression, or other interventions for suicidal ideation). Apparently, these were included because the papers included finding on suicide or suicide attempts, even though the studies were not designed to test hypotheses regarding suicide-related outcomes. There is a clear need to separate evaluations of the quality of the research as designed to test the primary hypotheses from the quality of the same studies for drawing inferences about suicide-related outcomes. The draft should have included information about methods for evaluating the quality of the studies for contributing to the literature on suicide. The absence of this information is a serious drawback.We added a table to the final report which now provides this information to readers. This table also lists sample sizes for the various studies so readers can see how sample size compares for studies with different primary outcomes (i.e., those studies designed to prevent suicide versus studies in which this was not a primary or pre-specified outcome of interest).
1As a related issue, the draft did not include a clear discussion of statistical power. Based on the discussion that was provided, statistical power did not appear to be included in the rates of the quality of research and the strength of the evidence. The methods section of the draft should have included a discussion of sample sizes and power, specifically for suicide prevention.We discuss this issue throughout many of the sections of the report (e.g., stating “The majority of trials did not involve the necessary sample sizes (mean, 284.4 patients; standard deviation, 177.8) or follow-up durations (median, 8 weeks; range, 4 weeks to 2.5 years) required to adequately evaluate risk of suicide attempts or suicides. Therefore, these trials generally provided inadequate to low-strength evidence for drawing conclusions about risk of suicide attempts and suicides.” in the section on pharmacotherapy. We also include a specific discussion of this issue pertaining to the table referenced in the above comment.
2Yes. The objectives and scope are clearly described. Methods are clearly articulated and documentation re: process is provided. The authors state that the goal is to update work by reviewing literature that was not reviewed by Gaynes et al or Mann et al. It may be helpful for the reader to know the main findings from these reviews. The authors may also want to provide more detail about further support or lack thereof for Gaynes et al and Mann et als' assertions based on this review. Some of this is provided later in the document – but seems to be missing from the beginning of the review and is not consistently presented throughout.We have updated the report to include more information on results from the Mann et al. (2005), Gaynes et al. (2004) and NICE (2011) systematic reviews throughout the report, and this information is also presented in tabular format.
3Yes; no comment.
4Yes; no comment.
5No. Overall I think this is very well written. The objectives, scope and methods are fairly well described, but I do have several comments:Thank you. Noted.
5
  1. In the Key Questions 3 and 4 themselves, it needs to be made clearer what referral and follow-up services are. How are these approaches not subsumed under KQs 1 and 2—if they are a subset of the interventions covered in KQ 1 and 2, why are they being looked at separately? How is some change in referral or follow-up process not an intervention?—this needs to be clarified for Exec Summary and in introduction. Perhaps general access to mental health care may be a better/clearer construct than referral and follow-up??
We have clarified differences between studies cited in the “psychotherapy” versus “referral/follow-up services” sections of the report, which describes why these studies were discussed in two different sections when the treatments were similar.
5
2.

There are a few places where there may be inconsistencies in the terminology used, and the terminology may not be consistent with the new VA DOD terminology for self harm behaviors. I would overtly acknowledge and reference this new nomenclature early on, include a table on it, and make sure that it is consistent throughout the document.

We added a description and definition of “suicidal self-directed violence,” the adopted VA/DoD nomenclature, in both the executive summary and the introduction section of the report. We have also updated the report with consistent terminology throughout.
5
3.

Because this follows up on previous reviews, I think it would be important to include some type of summary at the end of the response to each KQ that incorporates or acknowledges the previous relevant findings from those reviews. For several of the KQs, you do not have findings, but perhaps that is because you are only reviewing what was published between 2005 and 2011. Perhaps there are older findings that would provide more information or context for your findings. The findings from the previous reviews also should be addressed/integrated into the Discussion/Summary section so the reader can see if and how (or not) things may have changed.

We have updated the report to include more information on results from the Mann et al. (2005), Gaynes et al. (2004) and NICE (2011) systematic reviews throughout the report, and this information is also presented in tabular format.
5
4.

In the Exec Summ response to KQ2, the response is written as if suicide is the main or perhaps only outcome of interest. But you are also looking at other suicide behavior outcomes such as suicide attempts. In some places, like on page 3 and on page 15 this is not clear.

We have made these corrections and updated the report to consistently use the term suicidal self-directed violence in reference to outcomes.
5
5.

doing this review again so soon—have there been a lot of new studies/what is rationale? Who was the proponent for this review—can that be listed?

We have updated the methods section to more clearly describe the rationale and request for the report.
5
6.

In the Exec Summary it is striking to me that 16,502 papers were initially reviewed-these are all since 2005? Do you want to briefly describe your key or main inclusion criteria (I focus a lot on Exec Summary because this is all many readers will actually read.

We have added this brief description to the executive summary.
6Yes. The questions appear sound, but an explanation of what “suicidal self-directed violence” means would be helpful.We have defined and cited this terminology.
7Yes; no comment.
8Yes. The objectives, scope and methods are clear. I think the focus on RCTs is key as these studies have greater internal validity, and many other syntheses have made the choice to combine RCTs with observational studies. Keeping the focus on RCTs makes it clear how few high-quality data are available regarding preventative interventions for suicide, particularly in Veteran and military populations.Thank you. Noted, and we agree that the focus on RCTs helps limit the report to the highest quality research available on suicidal self-directed violence outcomes.
8Since this review explicitly uses the Mann review as a starting point, I would recommend that the synthesis build even more upon the Mann review. (The report already does do this, in part, in the more detailed sections.) Specifically I would recommend acknowledging where there is sufficient evidence to confirm the conclusions of this prior synthesis, where there is insufficient new evidence to comment on prior conclusions, and where there is sufficient new evidence that conflicts with prior conclusions. This should be done in addition to findings from the recent literature in new areas of intervention. At this point, there are several statements emphasizing contrasts to the Mann report which seem more a function of insufficient new evidence rather than new evidence that conflicts with prior conclusions. (For example, no new literature on clozapine was reviewed and there are insufficient data from studies of other antipsychotics to make a statement about other antipsychotic medications or the group of antipsychotic medications.) Insufficient new evidence would not seem to overturn prior findings, unless there was further synthesis of both the older and the newer findings.We have updated the report to include more information on results from the Mann et al. (2005), Gaynes et al. (2004) and NICE (2011) systematic reviews throughout the report, and this information is also presented in tabular format.
8I would also recommend mentioning the Bagley VA Evidence Based Synthesis earlier on as he and his colleagues also reviewed literature on psychotherapy and pharmacotherapy in addition to larger public health interventions.We have included information about this report and scope differences in the methods section of both the executive summary and the body of the report.
8On a minor note, this review started with studies published as of January 2005 and the Mann review covered until June 2005. Thus there is some overlap in the dates covered in the two reviews. Given review inclusion criteria, this results in an important 2005 publication being included in the Mann report (Brown, JAMA 2005) and not this report. Knowledgeable readers will likely be looking for this paper.We clarified that articles included in this report are only those not previously included in the Mann et al. report, and hence articles such as the Brown (2005) paper were excluded (pages 1 and 9).
2. Is there any indication of bias in our synthesis of the evidence?
1Possible. I used what I viewed as potentially positive findings that were published during the relevant period as markers for evaluating the draft. One was the Lauterbach study discussed in my response to question 3. This was a report of a study designed to determine whether adjuctive lithium prevented suicide reattempts in patients with depression or bipolar disorder who survived and initial attempt. Unfortunately, the investigators were unable to achieve the planned sample size. However, a finding based on post-hoc analyses suggested that lithium may have reduced deaths from suicide. I would have been interested in seeing how the draft evaluated this claim. However, the article was not included.Noted. As you state, this article was not included. This is because the study was conducted in Germany, a country outside the scope of this review, per initial scoping agreement with the stakeholders/CPG group requesting the report.
1Another was the Hatcher article on the effect of problem-solving therapy for suicide prevention. The article reports that there was no significant effect of problem-solving therapy in the entire sample. However, they report that a planned subgroup analysis demonstrated that the intervention was effective in the subsample of the subjects who had survived a suicide attempt prior to the index attempt that led to study entry. The article was included in the review but the planned subgroup analysis was not mentioned.We have updated this section, making specific reference to these findings.
1The two articles I mentioned represent two of the most significant potential advances of the past few years. Personally, I was looking to the Evidence Synthesis for guidance about the evaluation of the reported findings. However, neither of the salient findings were addressed. It is possible that this reflects poor implementation. However, it is also possible that this may reflect a bias towards negative findings.Noted. We have attempted to use the most current, objective, and stringent methods for preventing bias in this report, and have addressed the comments about these two studies above.
2No; No.Noted.
3No; no comment.
4No; no comment.
5No; no comment.
6Yes; no comment.
7No; no comment.
8No. The study selection criteria, quality assessment criteria, and rating of the strength of the evidence are clearly described. I agree with the focus on RCTs, given the limitations of the evidence from other study designs.Thank you. Noted, and we agree that the focus on RCTs helps limit the report to the highest quality research available on suicidal self-directed violence outcomes.
8Please include a table for 317 studies and reasons for exclusion.Appendix W is a table of excluded studies, and contains information on reasons for exclusion.
3. Are there any published or unpublished studies that we may have overlooked?
1Yes, I used what I viewed as important papers published during the relevant period as markers to evaluate the literature that was reviewed. Specifically, I searched for, “Lauterbach E. Felber W. Muller-Oerlinghausen B. Ahrentos B. Bronisch T, et al Adjunctive lithium treatment in the prevention of suicidal behaviour in depressive disorders: a randomised, placebo-controlled, 1-year trial. Acta Psychiatrica Scandinavica. 118(6):469-79, 2008”. Its absence from the literature that was identified raises serious questions about the process for identifying relevant literature.Noted. As you state, this article was not included. This is because the study was conducted in Germany, a country outside the scope of this review, per initial scoping agreement with the stakeholders/CPG group requesting the report. We have, however, included this article in a non-systematic addition to the review, per your suggestion. This information is now contained within the pharmacotherapy section of the report.
1In addition, I have heard verbal reports that findings were available, but not yet published, from a second study of cognitive behavioral therapy for suicide prevention in attempt survivors to determine whether it decreased the rate of reattempts. I may be useful to contact Dr. Gregory Brown from the University of Pennsylvania ( ude.nnepu.dem.liam@worbgerg) to get more information.Noted. Per follow-up discussions with the CPG and stakeholders, the decision was made not to include unpublished studies or data analysis in this report.
2Yes.
  1. In terms of intervention, several key studies seem to have been left out of the psychotherapy results section- perhaps because they were published prior or after – this makes it seem like even less work has been down in this area. See comment 1A above.
Noted; see responses below for each study.
2
1.

Brown GK, Ten Have T, Henriques GR, Xie SX, Hollander JE, Beck AT. Cognitive therapy for the prevention of suicide attempts: a randomized controlled trial. JAMA. 2005 Aug 3;294(5):563-70. PubMed PMID: 16077050.

Though our search did, indeed, capture this article, we did not include it because of its inclusion in the previously published Mann et al. (2005) report. We note this exclusion criterion in the report on pages 1 and 9.
2
2.

Bruce ML, Ten Have TR, Reynolds CF 3rd, Katz II, Schulberg HC, Mulsant BH, Brown GK, McAvay GJ, Pearson JL, Alexopoulos GS. Reducing suicidal ideation and depressive symptoms in depressed older primary care patients: a randomized controlled trial. JAMA. 2004 Mar 3;291(9):1081-91. PubMed PMID: 14996777.

This study was published prior to the beginning of our search dates, and was therefore not included in the report. Information published prior to June, 2005 was addressed in the Mann et al. (2005) report, and so as to avoid duplication, we did not include any such studies in this current review.
2
1.

Simpson GK, Tate RL, Whiting DL, Cotter RE. Suicide prevention after traumatic brain injury: a randomized controlled trial of a program for the psychological treatment of hopelessness. J Head Trauma Rehabil. 2011 Jul-Aug;26(4):290-300. PubMed PMID: 21734512.

This study did not include reports on outcomes included in this report (i.e., suicide and suicidal self-directed violence); therefore it was excluded from the review.
2
B.

For TBI among veterans may want to include Brenner LA, Ignacio RV, Blow FC. Suicide and traumatic brain injury among individuals seeking Veterans Health Administration services. J Head Trauma Rehabil. 2011 Jul-Aug;26(4):257-64. PubMed PMID: 21734509

This study is not a RCT and was therefore excluded from the review; however, this study was considered for the companion review conducted by our research team on suicide risk and assessment.
3No; At least none that I am aware ofNoted.
4No; no comment.
5No; Not that I know ofNoted.
6No; no comment.
7No; no comment.
8I am not aware of any additional RCTs in this area during this time frame.Noted.
4. Please write additional suggestions or comments below. If applicable, please indicate the page and line numbers from the draft report.
1The Mann article reviewed articles through June, 2005. Please provide more specifics about how you ensured that there were neither gaps, nor overlaps with the Mann article.We clarified that articles included in this report are only those not previously included in the Mann et al. report, and hence articles such as the Brown (2005) paper were excluded (pages 1 and 9).
1The methods suggest that the review process did not distinguish articles on the basis of the goals or the aims of the research that was reported. This is important. Research that was conducted, for example, to determine whether a specific intervention led to reductions in suicidal ideation, may have been well designed and adequately powered to address that question. However, even if it reported on the number of subjects who attempted or died from suicide, it would probably be underpowered to address these outcomes. This distinction should be considered in evaluating the quality of the studies reviewed.We added a table to the final report which now provides this information to readers. This table also lists sample sizes for the various studies so readers can see how sample size compares for studies with different primary outcomes (i.e., those studies designed to prevent suicide versus studies in which this was not a primary or pre-specified outcome of interest).
1The first sentence under ”Pharmacotherapy results” states, “Studies evaluated antidepressants …. For their efficacy in prevention of suicide …. “ In fact, none of the studies were designed or intended to evaluate efficacy for suicide prevention. The sentence is incorrect.We have updated the sentence to read: Studies evaluated antidepressants, atypical antipsychotics, mood stabilizers, and omega-3 supplements and reported their efficacy in prevention of suicidal self-directed violence in non-Veteran/military populations.
1The first paragraph under “Pharmacotherapy results” states that 9 studies were reviewed, then it cites 10. This should be clarified.We have clarified that these were 9 studies published in 10 publications.
1Among the citations in the first paragraph under Pharmacology results, the citations numbered 15-20 and 24 were conducted to evaluate depression as an outcome. They were not designed or powered for suicide-related outcomes. This should be stated.See above comment re: providing this information in tabular format.
1The conclusions stated at the end of the first paragraph refer only to suicide as an outcome. This is inconsistent with the definition of the scope of the review that addresses “suicidal self-directed violence.” As written, the conclusions are confusing and misleading.This sentence has been updated to read: “Therefore, they are felt to be of low strength, and are insufficient for determining the effectiveness of various combinations of antidepressant medications for reducing suicidal self-directed violence.”
1The reports cited as 21 and 22 were written to report on outcomes related to suicidal ideation. This should be acknowledged.See above comment re: providing this information in tabular format.
1The report cited as 23 included suggestive, apparently post-hoc, analyses of greater self-harm with certain medications. It may be misleading to conclude only that it said nothing about deaths from suicide.Added results of subgroup analyses showing increased risk in patients taking venlafaxine vs SSRIs and those taking benzodiazepines.
1Studies cited as 25-27 in the second paragraph were not conducted to evaluate the effectiveness of antipsychotic medications in reducing suicide deaths. This should be acknowledged.See above comment re: providing this information in tabular format.
1The point of reference 28 was that there were no drug related increases in suicidal ideation as an adverse drug effect. It should be acknowledged that the study was not conducted to test for decreases in death from suicide.See above comment re: providing this information in tabular format.
1The last sentence of the second paragraph say there is a contrast between the cited papers and findings of an effect of clozapine. In fact, the findings on clozapine reflect a difference between that medication and another atypical antipsychotic. There is no contrast.This sentence has been updated to read: “Notably, the previous review by Mann and colleagues reported an antisuicidal effect of clozapine, an atypical antipsychotic medication.”
1The text in the first sentence of the third paragraph is incorrect. Reference 29 was a 2.5 year study. Reference 30 was an 8 week study.This correction has been made.
1Reference 29 found no significant differences between lithium and valproate, but it is not correct to say that it did not have suicide or suicide attempt outcomes.This has been clarified in the report.
1Reference 30 focused on ideation and related symptoms; this should be acknowledged.We did not include ideation as an outcome in this report, and therefore those results are not reported. However, we do report primary outcome information from studies in tabular format.
1The reference for citation 31 is incomplete. It is from the British Journal of Psychiatry.This correction has been made.
1The review should acknowledge that 31 was intended to report on outcomes related to ideation and related measures rather than attempts or deaths from suicide.We did not include ideation as an outcome in this report, and therefore those results are not reported. However, we do report primary outcome information from studies in tabular format.
1Reference 32 reported that problem solving was effective for decreasing repeated self-harm in a subsample of patients with multiple previous episodes. This should be acknowledged.See above comment re: the Hatcher paper.
1Citation 39 referred to a study evaluating 64 adults with history of self harm. It may have reported on deaths, but it was conducted primarily to look at other outcomes.See above comment re: providing this information in tabular format.
1Citation 45 reported a decreases in ideation. This should be acknowledged. In its critique, it is not clear what was meant by the phrase “had methods that suggested an unclear risk of bias.”We did not report ideation outcomes in this report. The latter sentence has been clarified to read: “…used methods resulting in an unclear risk of bias.”
1Citation 52 is a secondary analyses of a study of an intervention similar to that reported in 45. It should be acknowledged that the outcome of interest was total mortality, not suicide. The study was conducted to test for decreases in suicidal ideation. Moreover, it is not clear why 52 is discussed in a section separate from 45 when the interventions were so similarSee above comment re: providing outcome information in tabular format. We have clarified differences between studies cited in the “psychotherapy” versus “referral/ follow-up services” sections of the report, which describes why these studies were discussed in two different sections when the treatments were similar.
2
  1. Page 1 – 20% of veterans – believe this number originally came from the work of Kaplan et al. It is somewhat problematic in that Veteran was broadly defined and likely included other cohorts (e.g. active duty).
This data came from NVDRS, which does include anyone who has served in the armed forces. We have modified this sentence to be more clear about who the 20% represent.
2
B.

Recent research has focused more on warning signs vs. risk factors as a prevention strategy. May want to consider including this. May also want to include warning signs in Analytic Model. 1: Rudd MD, Berman AL, Joiner TE Jr, Nock MK, Silverman MM, Mandrusiak M, Van Orden K, Witte T. Warning signs for suicide: theory, research, and clinical applications. Suicide Life Threat Behav. 2006 Jun;36(3):255-62. Review. PubMed PMID: 16805653.

Any intervention RCTs meeting inclusion criteria were included in this report, and a separate report completed by our research team is addressing risk factors and warning signs. We have included both risk factor and warning sign terminology in the analytical model.
2
C.

A number of studies are currently underway in the VA – this seems worth mentioning – information could be found on clinical trials.gov or VA websites.

We have included this information in the discussion.
2
D.

As Analytic model 1 and 2 appear to be identical may way to combine.

We agree, and have combined the analytical models as you suggest.
3This is an excellent review of RCTs to date and is an important followup to the Mann review. Essentially it states that we are a long way from where we would like to be in understanding what is and is not an effective intervention. As it pointed out, the base rate of suicide is so low, the phemenon itself so complex and the interventions so diverse that it is difficult to put together an RCT, particularly a blinded RCT, with sufficient power while appropriately limiting the variables being studied. This may account for the dearth of RCTs. In the end, it may be that other forms of evidence, albeit lower level evidence (e.g. aggregated performance improvement data), will be necessary to identify successful interventions.Noted, and we agree. We have made this suggestion in the discussion section of the report.
4None.Noted.
5In the executive summary, on page 1 I would include the reference number for the Gaynes review right after you write “Gaynes and colleagues”, not at the end of the sentence.We have made this change.
5On page 3, you write about two studies on mood stabilizers that did not have any suicide or suicide attempt outcomes—if not, why were these included in the review?We have clarified this sentence to read: “These trials reported no instances of suicidal self-directed violence for the duration of either study.” The trials did collect information on these outcomes, though no such events occurred.
5Intro section on pages 6 and 7 is nicely writtenNoted. Dr. Denneson was responsible for much of this section.
5In the inclusion criteria section you don't include the specific dates for including the studies you are reviewing (ie from 2005 to 2011)We have made this addition.
5The limitations section should also acknowledge that your search strategy specifically looked for keywords and terms related to suicide/suicide behaviors. There may have been potentially relevant manuscripts published about various interventions which did not have those terms attached to them. There may be a bit more to say about the limitations of the search strategy itself in the Limitations sectionWe have included this information in the appropriate sections.
6This is a very thorough review of suicide prevention interventions. The authors make a good case for focus on RCTs only and the four questions appear sound. I have a few suggestions for this report.Noted.
6First, it might be helpful to justify why other forms of violent death/ behavior were not included. There might be public health interventions that are relevant to suicide prevention that address other causes of death such as homicide, accidents, “suicide by cop” or accidental overdose.We have included comments to this effect in the discussion section.
6Second, more discussion on the heterogeneity of studies is warranted. Many of the RCTs reported might have had stringent exclusion criteria (as the authors noted that patient who are suicidal are often excluded from trials), often leading to minimal changes in outcome. Some recommendations on how such criteria should be modified would help in the development of more generalizable studies in the future.We have added more information on this topic in the results and discussion sections of the report.
7Page (i) wrong headerThis correction has been made.
7p. 3, p. 19 should be “usual care alone” rather than “along”This correction has been made.
7p. 5 Bruce et al 2004 is the primary reference for the PROSPECT study. Gallo et al was designed to examine all-cause mortalityAgreed; however, we report as the primary citation the article which reports on our outcome of interest (i.e., suicidal self-directed violence), and therefore we cite Gallo et al., whereas Bruce et al. is cited for inclusion as a companion article which was reviewed for methods information about the study. Similar examples are also cited as such in the report (e.g., the TADS trial papers).
7p. 6 “suicide screen” is problematic language since suicide does not meet many clinical epidemiologic criteria for appropriate screening targets, nor is there an evidence base to commend a particular technique to assess suicidalityNoted. We have removed this terminology from the sentence.
7p. 14 change from number of articles to number of publications is confusingWe agree that the paragraph can be confusing due to the fact that some studies are published in more than one paper. We hope the Literature Flow Chart can provide clarification.
8For clarity in the narrative, it may help to group studies (e.g. those assessing psychotherapy) into those with sufficient sample size and duration to actually have a chance of addressing the outcomes of suicides and suicide attempts and those that do not. Currently the narrative is organized primarily around the specific psychotherapeutic intervention which often have been examined in only a single study with quality issues.Though we did not group studies in this manner, we have added this information in tabular format to address this point, comparing sample size and commenting on statistical power in the studies.
8Would recommend a short discussion section in the executive summary.
5. Are there any clinical performance measures, programs, quality improvement measures, patient care services, or conferences that will be directly affected by this report? If so, please provide detail.
1I am concerned that the quality of the draft report as it is currently written could represent a barrier to implementation of new advances. I do not think the report should be released in its current form.Noted. The report was revised per reviewer feedback, and is released to the public after suggested changes have been made.
2The lack of evidence-based treatments would be expected to impact care for suicidal veterans.Yes, it could. It is our hope that future research will continue to inform evidence-based treatment research and implementation so that Veterans and members of the military may have access to effective, evidence-based care.
3This report will be viewed with interest by many in Patient Care Services, particularly those involved in suicide prevention. I think that the report validates what many believe – that suicidal behavior is complex, difficult to predict and can be difficult to prevent.Noted.
4Not that I am aware ofNoted.
6Office of Mental Health Operations, Canandaigua COE, National Center on Homelessness among Veterans, VA Cooperative Studies Program, local policeNoted.
7Suicide prevention coordinator programsNoted.
8In my reading of this summary, there were no RCTs at all for interventions for military populations or Veterans. There was insufficient evidence for specific pharmacotherapies or psychotherapies in general English speaking populations in eligible countries. The strongest evidence for psychotherapies was moderate evidence for no benefit of problem solving therapy for patients with suicide attempts. Therefore, this review suggests no evidence to support changes in or new clinical performance measures or mandated programs that emphasize these interventions for suicide prevention. The synthesis does outline an important research agenda for the VA.Noted, and we agree that, in the case of an absence of evidence, particularly for the populations of interest, this report provides information related to areas of research in need of further investigation.
6. Please provide any recommendations on how this report can be revised to more directly address or assist implementation needs.
1The organization of the report is generic, and as such, it does not appear to have been designed specifically to address suicide prevention. It may have been better to organize the report around the clinical ecology of suicide, where low numbers demand larger studies, and where information about surrogate endpoints (e.g., suicidal ideation) may be important, but where they may not translate directly into the prevention of suicide-related behaviors.Though we did not group studies in this manner, we have added a table to address this point, comparing sample size and statistical power in the studies.
1The evaluation and discussion should, perhaps, focus on studies that had adequate power to detect effects, and those where claims of effects were made. It should acknowledge that there may be promising findings regarding suicide ideation as an outcome, but that these were outside of the scope of the review. It should also be acknowledged that there were promising interventions, some that have been the subject of recent research, but where adequately powered clinical trials have not yet been conducted.We have added a section on this topic, and included a table to present information on statistical power and primary outcomes in the trials. We have acknowledged that there may be promising findings regarding suicide ideation as an outcome, but that these were outside of the scope of the review as part of the discussion section.
2
  1. Are there common elements of the most promising interventions that could be incorporated into current practice? May be helpful to review:
    Oxford Text of Suicidology and Suicide Prevention, DOI: 10.1093/med/9780198570059.003.0058,
    Chapter 58 The psychological and behavioural treatment of suicidal behavior
    What are the common elements of treatments that work?
    M David Rudd, Ben Williams and David RM Trotter This chapter provides a review of all currently available clinical trials targeting suicidal behaviour. In contrast to some previous available reviews, the focus of the current chapter is on identifying common elements of treatments that work. More specifically, we attempted to answer the question, what do treatments that work have in common? A number of psychological treatments have emerged as effective or potentially effective at reducing suicidal behaviour (i.e. suicide attempts). There now appear to be a number of identifiable core elements for treatments that have proven effective at reducing suicide attempts, all with direct and meaningful implications for day to day clinical practice. We also point out limitations in current science, including problematic follow-up periods and questions about the high-risk nature of some study samples.
We have reviewed this chapter, and agree that this non-systematic review could contain guidance for future research directions in the area of suicidal self-directed violence prevention interventions. We have added this citation to the discussion section.
2
B.

A trial of the PST (Hatcher et al. 2011) among Veterans may be indicated – wonder if this recommendation should be made.

We have included a more in depth discussion of this trial in the results section, and have highlighted this trial in the discussion section.
3Related to the comments in #4. Suicide prevention efforts have been under way in VHA and DoD for a number of years. Many interventions have a great deal of face validity, and, for reasons already cited, it may be difficult to generate RCT data to test them in traditional ways. The report could be enlarged to include a section summarizing the efforts to date, along with population data spanning the last X number of years. It need not make any statements about any particular intervention, since doing so would not be consistent with the approach taken in this review.We have now included more such information on earlier trials found in other systematic reviews such as Mann et al. (2005) and Gaynes et al. (2004) to provide a more comprehensive discussion of this point. Population data was included in the background sections, and will be covered in additional detail in a companion report by our research team on Suicide Risk and Assessment.
4Given the findings, there is little to implement.Noted.
5As per above, would flesh out Exec Summary a bit more since this is what most people will readWe have expanded this section per your recommendations.
6It would be helpful to shorten the executive summary into a one-page synopsis of the available evidence, what more needs to be researched, and from the available research, what is actionable for VA leaders to implement as public health, practice-based, or provider level interventions. For example, the Office of Mental Health Services uses a reporting tool in which key findings and progress updates are presented in tabular form.We agree that a brief summary format is beneficial for some readers. In addition to the executive summary, we report findings in a “management brief” single page format which is electronically disseminated following the final report publication.
7Less mechanical and repetitive approach to organizing the manuscript.We have attempted to organize the report in a clear manner, consistent with standard systematic review reporting criteria.
8Please see above. The evidence does not support immediate implementation of any suicide prevention program per se.Noted.
7. Please provide us with contact details of any additional individuals/stakeholders who should be made aware of this report.
1The report should be revised extensively before it is disseminated.Noted. We have made revisions as recommended by the peer reviewers.
5Jan Kemp, Ira Katz, Toni Zeiss from VA Office of Mental Health Services should just have a bit of a heads upThank you for these recommendations.
6Jan Kemp and Rob Bossarte, Canandaigua COE/VACO; Ira Katz, MD, Office of Mental Health Operations/VACO. DODThank you for these recommendations.
7Jan KempThank you for these recommendations.
8Drs. Zeiss, Kemp, Katz, Schohn in Central Office. VISN 19 MIRECC, VISN 2 Center of Excellence for Suicide Prevention, Defense Centers of Excellence for Psychological Health.Thank you for these recommendations.

Views

  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (1.7M)

Other titles in this collection

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...