Weight Loss After Participation in a National VA Weight Management Program Among Veterans With or Without PTSD

Details regarding MOVE! participants are available elsewhere (8). Veterans with a body mass index (BMI) of ≥30 kg/m² or a BMI of 25–29.99 kg/m² and at least one obesity-related comorbidity are eligible for MOVE! We used visit and outcome data from the MOVE! database comprising 148,963 patients who had attended at least two MOVE! sessions during fiscal years 2008–2012.

A total of 29 outliers with a baseline weight of more than 500 pounds or a six- or 12-month weight change of more than 100 pounds were removed. Analyses were limited to the 20,819 (14%) participants from the full MOVE! sample with intense and sustained participation, given that these individuals received a “dose” found to be effective (8). There were no differences by mental health status (PTSD, other mental conditions, or no mental health diagnoses) in the proportion that achieved this measure of MOVE! engagement (14% of each group; p=.91). Participants whose weights were available six (N=16,551, 79.5%) or 12 (N=13,270, 63.7%) months after enrollment were included in the effectiveness analyses. They represented 11% and 9%, respectively, of the full MOVE! sample.

The Institutional Review Board (IRB) at the VHA Ann Arbor Medical Center agreed that this project was consistent with an existing IRB approval, granting a waiver of informed consent. VHA Puget Sound’s Research and Development committee also approved this project.

Outcomes for this evaluation were total weight change and clinically meaningful weight loss (≥5% body weight) (9) and were assessed at six and 12 months following a veteran’s baseline visit to MOVE! (index enrollment date). The MOVE! database captures patient weights at 180-day increments after the index enrollment date, which include data collected during the 60-day window before and after each increment (8).

The independent variable was mental health status: PTSD, other mental condition or conditions but no PTSD; and no mental health diagnoses. Mental health diagnosis was determined by the presence in VHA medical record data of at least one inpatient or outpatient visit with an ICD-9-CM code for one of the following conditions in the year of or in the year preceding the index enrollment date: PTSD; schizophrenia, schizoaffective disorder, or other psychotic disorders; bipolar disorder; depressive disorders; alcohol or other substance abuse or dependence; dementia; other anxiety disorder; and personality disorder.

Covariate selection was guided by a prior MOVE! evaluation (10), and variables were obtained from VHA medical record data. Standard sociodemographic characteristics included age, sex, race-ethnicity, and marital status. Any service connection (injury, ailment, disease, or disability due to military service) was included because it facilitates access to care, and Operation Enduring Freedom/Operation Iraqi Freedom (OEF/OIF) status, confirmed by OEF/OIF roster, was included because OEF/OIF veterans with mental conditions are at increased risk of obesity (11) but have suboptimal engagement in care (2).

We also examined exposure to obesogenic psychotropic medication (use for 30 or more days and one or more refills or two supplies of prescriptions in the six months prior to MOVE! for six-month outcomes and in the six months following the MOVE! start date for 12-month outcomes). Two recent VHA weight management evaluations guided the selection of the following medications as obesogenic (12,13): lithium, gabapentin, valproic acid, thioridazine, clozapine, olanzapine, quetiapine, risperidone, nefazodone, paroxetine, amitriptyline, clomipramine, desipramine, doxepin, imipramine, nortriptyline, and mirtazapine. MOVE! baseline weight was included, and to adjust for disease burden, the Charlson Comorbidity Index was calculated by summing the number of diagnoses for general medical conditions associated with at least one visit in the current or prior fiscal year.

SAS, version 9.3, was used for all analyses. Linear and logistic regression analyses identified associations of mental health status and outcomes. Veterans with intense and sustained participation and those with six- and 12-month weight data differed significantly from the general MOVE! sample across all baseline characteristics. We addressed this issue by adjusting for these characteristics in multivariate analyses.

Participants differed significantly on all characteristics by mental health status (Table 1). [Multivariate regression analysis results of the association between participant characteristics and weight loss are available online as a data supplement to this report.] Adjusted analyses indicated that veterans with PTSD lost significantly less weight than veterans with no mental health diagnoses at six (mean±SE=–.72±.31 pounds, 95% confidence interval [CI]=−1.32 to –.12) and 12 (mean±SE=−1.01±.46 pounds, CI=−1.91 to –.11) months (p<.05). They were also less likely to lose 5% or more of body weight at six months (odds ratio=.89, CI=.79–.99, p<.05). At six months, participants with other mental conditions lost significantly less weight than participants with no mental health diagnoses (mean±SE=–.51±.25 pounds, CI=−1.01 to –.02, p<.05).

We reselected our sample so that those with alcohol dependence or abuse were not included in the “other mental health condition” category and performed interaction testing. This was done to examine whether alcohol misuse explained the differences in weight loss between those without mental health conditions and those with PTSD or other mental health conditions. The interactions were nonsignificant.

Among veterans with intense and sustained MOVE! participation, those with PTSD experienced less weight loss than veterans with no mental health diagnoses, both at six and 12 months. Veterans with PTSD also experienced less weight loss than veterans with other mental conditions, but the differences were not statistically significant in adjusted analyses. In addition, at six months, veterans with other mental conditions lost less weight than those without a mental health diagnosis. Although differences between groups were generally small, there was a more substantial difference in the proportion of veterans with PTSD versus those without a mental health diagnosis who lost a clinically meaningful amount of weight at six months (23.7% versus 28.7%). However, this difference disappeared at 12 months. Whereas weight loss outcomes were poorer for veterans with any mental conditions, MOVE! appeared to be less effective for veterans with PTSD, and alcohol misuse did not explain this difference.

We do not know why veterans with PTSD and other mental conditions lost less weight than veterans without mental health diagnoses after intense and sustained participation in MOVE! Although there were sociodemographic and health status differences by mental health status, differences in weight loss among the three groups persisted in adjusted analyses. Moreover, the difference in weight loss cannot be explained by poorer participation because intense and sustained participation was equivalent across categories. In fact, the average number of MOVE! visits was lowest for veterans without mental health diagnoses (data not shown).

It is possible that the observed difference in weight loss was due to poor health behaviors among veterans with PTSD (6,7) and that core PTSD symptoms interfered with making behavioral changes needed to achieve weight loss. Such symptoms include behavioral and emotional avoidance, anhedonia, and sleep disturbance.

System-level factors may also play a role. In the VHA, mental conditions are primarily treated within specialty mental health clinics. Prior research has suggested that integrating health behavior change programs within mental health clinics may help to reduce disproportionate morbidity facing individuals with mental illness. For example, a recent review recommended integrated physical activity interventions for those with serious mental illness because mental health providers are best equipped to provide relevant reinforcement and address unique barriers to behavior change (14). Such interventions might also address psychological functioning, given the benefits of weight loss for patients with PTSD and depression (15).

Although these data have potential implications for VHA health care delivery, it is important to note that the findings may not generalize to other patient populations, including the general MOVE! participant population who did not have intense and sustained participation. Measures were derived from administrative data, so we were not able to examine the severity of PTSD and other psychiatric conditions or potential barriers to weight loss for those with PTSD and other mental conditions.

This evaluation introduces novel findings about MOVE!’s effectiveness for veterans with PTSD and other mental conditions who were engaged at an intense and sustained level. Future studies are needed to explore potential contributors to poorer outcomes for veterans with mental conditions and whether tailored weight management interventions are needed to help such veterans lose weight. Finally, few veterans, regardless of mental health status, had intense and sustained participation in MOVE! Thus although behavioral weight management interventions tailored for individuals with mental conditions may be warranted, increased efforts are also needed to enhance MOVE! participation broadly.

This material is based on work supported by the Office of Research and Development, U.S. Department of Veterans Affairs (VA). Funding for this evaluation came from Quality Enhancement Research Initiative (QUERI)–funded locally initiated project grants (QLP 55-017) provided by the VA Health Services Research & Development Diabetes and Mental Health QUERI Programs. Dr. Littman’s time was also supported by a VA Rehabilitation Research and Development Career Development Award (6982). Dr. Klingaman’s contributions were supported by the VA Office of Academic Affiliations Advanced Fellowship Program in Mental Illness Research and Treatment. This quality improvement evaluation was conducted at the request of the VA National Center for Health Promotion and Disease Prevention. The views expressed in this article are those of the authors and do not necessarily represent the views of the VA.

The authors report no competing interests.