SERVE Model: Is Adherence to Five Key Evidence-Based Lifestyle Behaviors Associated with Changes in BMI Among Veterans?

OBJECTIVES: To examine whether: 1) adaptation of the SERVE (Sugar-Sweetened Beverages (SSB), Exercise, Restaurants, Vegetables and fruits (V/F), and Eating small portions) model help veterans to lose weight?; 2) Do single or aggregate of lifestyle changes in the SERVE model predict weight loss in veterans? METHODS: This study is a secondary longitudinal analysis of data collected from veterans with obesity or overweight with a comorbidity participating in the Goals for Eating and Moving (GEM) RCT. The SERVE model consists of 5 key evidence-based lifestyle behaviors collected at the baseline and 1 year visits: SSB intake, eating small portions, and restaurants collected via the Latino Dietary Behaviors Questionnaire, V/F intake via the Food Behavior Checklist, and exercise via the International Physical Activity Questionnaire. Logistic regression analyses were conducted to assess associations between single or aggregate of lifestyle behaviors in the SERVE model and changes in BMI controlling for the group assignment, gender, age, and race/ethnicity. RESULTS: Of the 240 participants, 72% were male with mean age of 51.2 years and mean BMI of 33.3 at baseline. After 1 year, the mean BMI of the veterans decreased (−0.43), more participants reported to exercise ≥150 minutes/week (23% vs 42%), eat small portions (26% vs 40%), and eat at a restaurant <3 times/week (60% vs 74%). No changes in V/F and SSB intakes were found. After controlling for confounders, those with higher portion control (many times/all the time vs never/sometimes: B = −1.8) and those with exercise level ≥ 150 minutes/week (B = −1.5) had significantly lower BMI after 1 year. Although not significant, similar trends were seen for eating at a restaurant < than 3 times/week and intakes of V/F (2 ≥ cups/day vegetables and ≥1.5 cup/day fruit) and SSBs. Those with at least 2, 3, and 4 positive lifestyle behaviors had decreases in BMI after 1 year (B = −0.8 [SERVE2]; B = −0.6 [SERVE3]; B = −0.8 [SERVE4]; all p < 0.05). CONCLUSIONS: Our findings support that the five key evidence-based lifestyle behaviors in the SERVE model are important determinants of changes in BMI and suggest that the SERVE model may be useful for dietary counseling. Results from this study may help inform future planning of interventions and policies targeting prevention of obesity and related chronic diseases in veterans. FUNDING SOURCES: National Institutes of Health.