Patient, facility, and environmental factors associated with obesity treatment in US Veterans.

Abstract

Background: Identifying patient-, facility-, and environment-level factors that influence the initiation and retention of comprehensive lifestyle management interventions (CLMI) for urban and rural Veterans could improve obesity treatment and reach at Veterans Affairs (VA) facilities.

Aims: This study identified factors at these various levels that predicted treatment engagement, retention, and weight management among urban and rural Veterans.

Methods: A retrospective cohort study of 631,325 Veterans was designed using VA databases to identify Veterans with class II and III obesity during 2015-2017. Primary outcomes were initiation of CLMI, bariatric surgery, or obesity pharmacotherapy within 1 year of index date. Secondary outcomes included treatment retention and successful weight loss. Generalized linear mixed models were used to evaluate the relationships between factors and obesity-related outcomes, with rurality differences assessed through interaction terms.

Results: Patient characteristics associated with increased odds of initiating CLMI included female sex (p < 0.001), black race (p < 0.001), sleep apnea (p < 0.001), mood disorder (p < 0.001), and use of medications associated with weight loss (p < 0.001) or weight gain (p < 0.001). Facility use of telehealth was associated with greater odds of CLMI initiation in urban Veterans (p < 0.001) but lower retention in both populations (p = 0.003). Routine consideration of pharmacotherapy was associated with higher CLMI initiation. Environmental characteristics associated with increased odds of CLMI initiation included percent of population foreign born (OR = 1.03 per 10% increase; p < 0.001), percent black (p < 0.001), and high walkability index (p < 0.001). The relationship between total population and CLMI initiation differed by rurality, as greater population was associated with lower odds of CLMI initiation in urban areas (OR: 0.99 per 1000 population; p < 0.001), but higher odds in rural areas (OR:1.01, p = 0.01). Veterans in the south were less likely to initiate CLMI and had lower retention (p < 0.001).

Conclusion: Treatment and retention of CLMI among Veterans remain low, highlighting areas for improvement to expand its reach both urban and rural Veterans.