Background/Purpose:

Low body mass index (BMI) is associated with more destructive disease in rheumatoid arthritis (RA). Paradoxically, obese patients have been shown in studies to be more likely to discontinue therapy, suggesting a refractory phenotype. The purpose of this study was to examine the association between BMI and DMARD persistence, accounting for factors that may confound this relationship.

Methods:

VA administrative databases were used to define unique initial courses of methotrexate (MTX), self-injectable TNF inhibitors (TNFi), hydroxychloroquine (HCQ), sulfasalazine, and prednisone. Discontinuation was defined as a lapse in drug refill >90 days. The closest values for CRP and BMI within 30 days of treatment start date were linked. Multiple imputation was used to address missing laboratory values. Health factor data and diagnosis codes were linked to treatment courses. Multivariable Cox proportional hazards models were used to evaluate associations between BMI category and time to DMARD discontinuation. Covariates included age, sex, race (Black vs. White), calendar year (2005-2009 vs. 2010-2014), current smoking, CRP, disease duration >5 years, anti-CCP status, other DMARDs, diabetes, hypertension, CHF, history of malignancy, depression, and anxiety, and the Rheumatic Disease Comorbidity Index (RDCI).

Results:

There were 46,970 unique initial DMARD courses (88% male) with RA between 2003-2014. Patients in low (BMI <18.5) and normal (BMI 18.5-25) categories had the greatest likelihood of discontinuing MTX, HCQ, and TNFi (Figure). Severe obesity (BMI >35) was only associated with a greater likelihood of discontinuing prednisone compared to overweight BMI (BMI 25-30) [HR 1.09 (1.02, 1.16) p=0.004]. Factors associated with earlier MTX and/or TNFi discontinuation included female sex, black race, older age, greater comorbidity, and a history of depression, malignancy, CHF, active smoking, and recent calendar year (Table). Among TNFi users, concurrent MTX use was associated with a lower likelihood of discontinuation.

Conclusion:

Among US veterans with RA, obesity was not associated with reduced persistence of DMARDs and/or TNFi’s, except for prednisone. Conversely, low BMI was associated with reduced drug persistence; this finding may be explained by greater severity of RA and/or other co-morbidities and warrants further evaluation. These data are not consistent with the hypothesis that obesity is a biologic mediator of refractory disease, resulting in failure of drug therapy. In contrast, other co-morbidities are shown to be important predictors of drug persistence.