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

Meis L, Griffin J, Greer N, et al. Family Involved Psychosocial Treatments for Adult Mental Health Conditions: A Review of the Evidence [Internet]. Washington (DC): Department of Veterans Affairs (US); 2012 Feb.

Cover of Family Involved Psychosocial Treatments for Adult Mental Health Conditions: A Review of the Evidence

Family Involved Psychosocial Treatments for Adult Mental Health Conditions: A Review of the Evidence [Internet].

Show details


1. Are the objectives, scope, and methods for this review clearly described?
Yes. On page 9 Introduction PL 110-387 signed in Oct 2008 added Marriage and Family Counseling and dropped the contingency on non-service connected Veterans. The May 2010 PL just added primary caregivers to eligible individualsWe have revised this part of the introduction to clarify that PL 110-387 expanded coverage and that PL 111 added primary caregivers to eligible folks.
No. There is a good measure of ambiguity about the goals and scope of this review. This ambiguity is generated from the following factors:1) We have clarified in the introduction the rationale for the review.
1) The background of the review highlights the legislation that expands the services to family members (the 2010 Caregiver legislation, but also applies to 2008 legislation that provides for marriage counseling as a VA service). I believe it was fair to say that the primary impetus for this legislation was the national pressure on VA to provide expanded services to family members, in reference to greater mental health needs of family members, and the impact that both medical and mental health issues of Veterans have on families. The focus of the review, however, is on the treatment of individual disorders, and not on outcomes of family members as individuals or the marital/ family unit. This may create a disconnect in the reader’s mind about the rationale for the review.
The rationale for the change in services, however, does include that family members constitute important members of the treatment team. This is a key part to the rationale that a review is needed to examine the evidence that family member involvement does improve outcomes. A more nuanced and spelled out rationale would help set the reader’s expectations a bit better.
2) The definitions of different types treatments defined by Baucom et al. were described as part of the background, but no systematic differentiations regarding these classifications of how family members are involved in treatment were made in this review (only brief occasional mentions). Thus, the review is not really a proper follow-up to Baucom et al. One consequence of this is that the review did not place marital distress or family dysfunction as clinical syndromes, unlike Baucom et al., where the authors treated those outcomes as treatable entities in and of themselves. This would be expected given the background/introduction of the review. Although the Limitations section discuss this point, it should be highlighted in the beginning of the review2) We have clarified the scope of the review in the introduction and highlighted the review is not intended as a strict update to Baucom and colleagues’ review.
3) There was very little emphasis was made on relationship distress as a moderator in the review, with only a mention in the sections on couples therapy interventions for ED and also for depression. This is potentially highly relevant in that findings in the pre-1995 period of time prior was that couples therapy for depression may not be effective, and perhaps ill-advised in couples who do not consider themselves maritally distressed, only with a partner with depression. This finding may be relevant for other disorders, and although few studies have addressed the issue in their designs, it should be part of the dialogue from the beginning of the review and part of the discussion and recommendations for future research.3) We agree this is an important issue. We have highlighted throughout the results section when this information is available and included a discussion of findings relevant to this question in the discussion.
4) Behavioral Couples Therapy (BCT) versions as treatments for substance abuse and alcohol use disorders were referenced often in the review with no qualifier that these are variants of BCT specifically designed to treat these disorders, except in the more detailed descriptions of the Appendices (which may not get read by many readers). They include procedures never used in standard BCT or expanded Integrative Behavioral Couples Therapy (IBCT) designed to treat marital distress. IBCT being disseminated throughout VA currently would very likely not be effective for substance abuse or alcohol use disorders. This ambiguity could be very misleading to readers unfamiliar with the literature.4) Thank you for your suggestion. We have clarified this in the results section for substance use disorders and refer to BCT as a ‘disorder specific couple/family treatment’ in additional places for clarity.
Yes. All methods are clearly described. Methodology is rigorous and effectively implemented. Outcomes of interest were well selected and decisions to include and exclude studies seem sensible given the intent to extrapolate findings to U.S. Veteran populations.Thank you.
Yes. Objectives, scope and methods are clearly articulated and findings are clearly summarized in multiple formats. Tables which include main findings are particularly facilitative (e.g., Table 8).Thank you.
2. Is there any indication of bias in our synthesis of the evidence?
No. There is no indication of bias.
Although I understand ESP’s rationale, I believe that given the undeveloped nature of this literature, limiting the review only to RCTs may have been overly limiting to understand the relevant clinical issues, trends, or promising practices.We certainly agree with the need to disseminate information on those promising interventions underdevelopment that are currently or soon to be subjected to more rigorous RCTs to evaluate their efficacy. Given the size of this review as it currently stands, limited to RCTs, it was beyond the scope of the project to expand our search to other study designs (e.g., open trials; quasi-experiments). We have added this to the limitation section.
3. Are there any published or unpublished studies that we may have overlooked?
Please refer to reviews by Shirley Glynn and Lisa DixonThese reviews have both been integrated into the discussion section specific to findings for schizophrenia.
No. I am not aware of studies that have been overlooked.
No. Review appears extensive and literature search process is clearly displayed in Figure 2.Thank you.
I was surprised to see that none of Candice Monson’s work on couples therapy for PTSD was included. I don’t have the studies in front of me, so it may be that is because they were not RCTs. If so, see my comment above.You are correct. Dr. Monson’s currently published work did not meet our inclusion criteria (i.e., currently she has no published RCTs). We referenced this work in our limitations.
4. Please write any additional suggestions or comments below. If applicable, please indicate the page and line numbers from the draft report.
Page 5 and 62 Recommendation for Future Research – PL lists eligible individuals for family services and that does not include close friends or intimate partner unless they are residing with the Veteran.
Page 5 and 62 Family Services and Caregiver Services are administered from two different Program Offices and are conceptualized as different – perhaps introducing caregivers brings in a different topic?
We have revised the introduction to better describe the two laws that have expanded services. In this explanation we also describe that PL 111-163 is only for a select group of family members. We have also highlighted the issue of who is eligible for these services in the discussion.
Overall, the review was comprehensive and inclusive, providing a critical snapshot of the state of the evidence for family-involved psychosocial treatments for mental health conditions of relevance to Veterans.Thank you.
There was very little integration of the findings of this review with the findings from Baucom et al. (1998). Combining the findings from this review with the previous is important since many interventions showing strong evidence of effectiveness (e.g., Family Psychoeducation for schizophrenia spectrum disorders), have not been as extensively examined in the period from 1995 forward. As stated above, this review did not continue with the classification of types of family involvement, which significantly weakens our understanding of the actual interventions being examined.We have taken better care to highlight the specific interventions that are reviewed and which category of intervention they fall under throughout the document (results and discussion section).
There was only a brief final mention in the recommendations of comorbidity as a factor examined in very few of the studies. This issue should be mentioned earlier and in greater detail since comorbidity is the norm for Veterans and indeed many older adults, Veterans or not. This recommendation should be front and center.We have included a more explicit review of the types of co-occurring problems that were inclusion and exclusion criteria in the trials reviewed. We have also expanded the discussion of this issue in our future research section.
The term “slower rate of relapse” was used consistently in the section on family involved treatments of substance and alcohol abuse. I believe the authors mean “lower rate of relapse” since most or all of the findings are rates at various endpoints and do not describe a slope or growth curve of relapse across time.Following the review of this draft, we conducted pooled analyses of the BCT studies which allowed us to draw more definitive conclusions about the efficacy of BCT compared to individual treatment. See results section.
On page 10, the authors state “Most prior reviews have focused on specific conditions (i.e., depression or substance use disorders), limiting the ability of past work to generalize to family-involved mental health care more broadly.” It is unclear what “more broadly“ means: Comorbidity? Special populations? Non-symptom outcomes?We have clarified this in the Introduction
On page 33, the authors state “For studies of AUD, all trials report better outcomes for BCT or BMT than IBT post-treatment and all follow-up time points, but many of these differences were not statistically significant.” The authors should allow that only the statistically significant findings are actually reportable as “better outcomes.”We have removed discussion of non-significant differences between conditions.
On page 34, the authors discussion the controversy over Fals-Stewart’s findings very economically and fairly. They need to provide a citation for the public charges of fabrication and of his death, a reputable news source, for example (a Google search will yield one fairly quickly).We have included a citation of both the NY State Attorney General’s press release and a copy of the felony complaint filed by the AG’s office.
Page 39, last line “(Reference)” appears in the text when it likely [should list the author/year citation].Corrected.
On page 58, the authors refer to Table XX, when the next table is 15.Corrected.
“Baucom (1998)” many times was cited when the correct citation is Baucom et al. (1998).Corrected.
The evidence base bearing on the questions of interest was, unfortunately, very limited. The studies reviewed covered a wide range of interventions but the number of trials for the same interventions was very few. This means that although there were a number of promising findings from single trials, but evidence in these cases was of low quality, given lack of replication. One finding with moderate strength of evidence, that behavioral couples therapy can slow the rate of relapse for substance abuse disorders, appears to overstate the impact of the intervention, given that findings related to more important outcomes such as abstinence rates were mixed. In the Conclusions section starting on page 60, it is stated that Behavioral Couples Therapy is superior to individual therapy for substance abuse disorders, but this conclusion does not seem warranted given the mixed findings across studies. Behavioral Family Therapy did seem to have a consistently positive effect on family functioning outcomes across all four studies that reported outcomes in this domain; possibly, this finding should be emphasized more in the report. Given the lack of the research base, it may be worth expanding the Future Research section; potentially this report can prompt more methodologically strong research on family interventions within VA research organizations.Regarding the strength of evidence of BCT, since the initial peer review, we have conducted pooled analyses comparing BCT to individual treatment in improving rates of abstinence and improving family adjustment. These findings are more supportive of BCT then our previous narrative review of the number of studies finding significant versus non-significant differences.
1. A paragraph (pg. 34) is included regarding work by Fals-Stewart – it may be helpful to provide this background information prior to presenting data regarding studies (Fals-Stewart – 1996, 2002, 2003 etc…)Thank you for the suggestion. This has been done.
2. Table 15 – may be useful to add borders (gridlines) to facilitate ease of reading.Done.
3. Cost related outcomes did not appear to be a focus of studies presented. Wonder about this as an outcome for future studies (particularly within VA), and whether it would be useful to include discussion regarding this in the Recommendations for Future Research.Thank you for your suggestion. We have addressed this in Future Research.
4. Several small typos noted (e.g., page 34 line 2 – Fals-Stewart, 1996, 200, 2002…) – also Higgins 2009 reference appears to be missing from list (this reviewer was interested in this publication so it was looked for all references were not checked).Thank you for your attention to detail. We have attended closely to these issues in the final report draft.
The exclusive focus on RCT’s and patient outcomes is a limitation. Not clear why previous reviews such as meta-analyses were not considered. Numerous sophisticated quantitative reviews have been published.Our literature search identified systematic reviews and meta-analyses in additions to RCTs. Several recent reviews are mentioned in the report. We also looked at reference lists of recent reviews to identify primary studies our literature search might have missed. We have taken care to be more explicit in integrating these reviews into our results discussion for each set of mental health conditions reviewed.
It is not clear to me what “drug treatment” or “no treatment” means in the comparison condition for KQ1. Does that mean the absence of any alternative active treatment? The reason for asking is that drug treatment would typically come with some kind of support, and that might be mentioned.We were interested in reviewing the evidence of the efficacy of family involved interventions (compared to no intervention or non-psychosocial interventions), as well as the degree to which family involved interventions are superior to an alternative individually-focused or family involved intervention (i.e., specificity). The ‘medication only’ conditions involve interventions that were solely pharmacological including medication and monitoring of medication use, but where the medication condition was not intended as a psychosocial treatment or psychotherapy. This has been clarified in the introduction and the wording of the Key Questions, We have also clarified what additional provider contact was included in intervention conditions we considered ‘medication only’
I am not sure what this means: “Overall, the studies reviewed appeared to favor comparisons between a family-intervention and an active treatment, limiting our conclusions for this key question. (page 3).” Does that mean that the review didn’t consider many of the landmark studies? The review’s findings regarding schizophrenia are puzzling given the extensive number of studies and meta-analyses supporting the effectiveness of family psychoeducation.This is due to the scope of our review. We did not include non-US studies or studies published prior to 1996. However, we highlighted the work prior to our review that established the efficacy of these treatments in Table 1 and discussed our findings within the context of other reviews throughout the document in the executive summary, results, and discussion sections
One issue for consideration is the “lumping” vs “splitting” issue. This review splits studies by diagnosis. However, in practice family interventions are not narrowly offered, and they share techniques. Miklowitz’s FFT is similar to FPE for schizophrenia; an alternative way to understand the literature is across diagnoses.We have addressed this in the limitations section.
The name of the office is Office of Mental Health Services, not just Office of Mental HealthThis has been corrected. Thank you.
I appreciated that in the summary of areas for future research in two areas in the paper, the role of nontraditional family constellations was highlighted. In the substance use disorder section, I appreciated that the results were broken into different types of effectiveness re: initiation, attendance, and adherence. On pg 34, although it is a touchy subject, I think it is a good thing that the issues around the work of Drs. Fals-Stewart are addressed.Thank you for your positive feedback.
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.Thank you – we will share these suggestions with the people responsible for dissemination of the report.
Findings should be of direct relevance to the mission of the VA’s Family Services Program
Every major VA medical center will be affected by this report in that the effectiveness of family involved services, especially in reducing relapse for substance abuse and alcohol disorders
The report appears to indicate that evidence for most couples and family-based interventions is largely insufficient to warrant widespread implementation within VHA. The intervention that does appear to be supported by consistent evidence, CRAFT, is not very well suited to implementation within VHA because it is delivered by a mental health professional to a family member whose loved on is not seeking treatment. It may have important training implications for community-based providers and possibly staff members of Vet Centers. The other finding with moderate strength of evidence, that behavioral couples therapy can slow the rate of relapse for substance abuse disorders, is not very impressive given the lack of impact of this intervention on arguably more important outcomes such as abstinence rates.Regarding the strength of evidence of BCT, since the initial peer review, we have conducted pooled analyses comparing BCT to individual treatment in improving rates of abstinence and improving family adjustment. These findings are more supportive of BCT then our previous narrative review of the number of studies finding significant versus non-significant differences.
Would expect that findings would have implications in terms of future VA research funding. May also have implications for current evidence-based treatment rollouts.
The Office of Mental Health Operations should review to determine if there is any relevance of the information in this report to their Mental Health Information System, which monitors a variety of practices in the field.
6. Please provide any recommendations on how this report can be revised to more directly address or assist implementation needs.
I am still struggling with the bottom line – probably effective – won’t cause harm? How does the research supporting family interventions compare to the research supporting other interventions currently being used in the VA?To adequately address how family interventions compare to the population of interventions currently provided by the VA, a systematic review of individually-oriented interventions would be required. This is beyond the scope of the review.
However, we have taken care to better highlight the primary take home points in the executive summary and in our final discussion section. We have included additional pooled analyses of the BCT studies comparing BCT to individual therapy, which provide greater clarity to our conclusions regarding the comparative effectiveness of BCT to individual therapy.
A potential conclusion from the findings of the report is that the state-of-the-science is that more efficacy and effectiveness research is needed on Veteran-focused family-involved psychosocial treatments to inform dissemination and implementation.We agree and have highlighted these issues in the discussion section.
The review’s scope would have to be expanded significantly to discuss effective implementation strategies, but this would indeed be highly valuable for VA.We agree that identifying and evaluating effective implementation strategies would be valuable; however, it is outside the scope of this report.
As mentioned above, comorbidity is the rule, rather than the exception, and very few studies address comorbidity. Clinicians have very little guidance as to how to proceed in these circumstances. A brief (and very common) clinical scenario that illustrates the problem: A 34 yo Veteran with PTSD, depression, and TBI violently pushes his wife after weeks of arguments over money, his at-risk alcohol use, and discipline of their children. He recently entered VA care and is open to treatment. Possible interventions include individual alcohol treatment, BCT for alcohol abuse, IBCT, anger management, and cognitive rehabilitation. The couple is asking for couples counseling for their arguing because they realize it upsets their 4 yo son. The Veteran is unconvinced he has a drinking problem.See above.
It may be helpful if the authors would recommend research priorities related to the area. Several interventions are promising, but research is very limited and trials with Veterans are lacking. A set of recommendations about which interventions might be prioritized for investigation within VHA research mechanisms might be helpful.We have included a more expansive future research section and address these issues there.
I think the report could benefit from greater consideration of how family interventions might be used in clinical care and the gap between the research parameters and what is found clinically.We have included a more direct discussion of the need for studies examining patients with multiple problems (e.g., substance use, TBI, intimate partner violence) in the Future Research section.
See my comments in response to question #3 and #4. I am afraid that the super rigorous limitation of the review to just RCTs may cut off possibilities for identifying promising practices for pilot projects in the field.See above.


  • 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...