(B-186) Impact of social determinants of health on receipt of guideline-concordant infection prophylaxis among Medicare beneficiaries with antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis

Background: Infections are the leading cause of hospitalization and mortality in patients with ANCA vasculitis (AAV), a relapsing-remitting autoimmune disease treated with strong immunosuppressants. Although clinical practice guidelines recommend antimicrobial prophylaxis to prevent Pneumocystis jirovecii pneumonia (PJP) during AAV treatment with cyclophosphamide and/or rituximab, only about one-fourth of patients receive it. It is unknown how social determinants of health (SDOH) impact receipt of prophylaxis.

Objectives: To examine the impact of individual and community-level SDOH on receipt of guideline-recommended PJP prophylaxis among Medicare beneficiaries with AAV.

Methods: From Medicare fee-for-service claims linked to the Area Health Resources File (AHRF) we identified a national cohort of new episodes of immunosuppressant treatment for adults with AAV involving cyclophosphamide and/or rituximab over 2016-18. New episodes were identified using a 90-day washout and required ≥1 diagnosis code for AAV in the prior year. We measured SDOH as individual-level variables in Medicare data (i.e., sex, race/ethnicity, insurance coverage with Medicaid or Part D low-income subsidy [LIS],) and community-level factors for the beneficiary’s county (i.e., low education, persistent poverty, primary care shortage area, urban/rural, US census region) as defined in the AHRF. PJP prophylaxis was defined as receiving ≥14 days’ supply of trimethoprim-sulfamethoxazole, pentamidine, dapsone, or atovaquone overlapping the first 30 days of the immunosuppressant episode. We used log-binomial regression to examine associations of individual SDOH with prophylaxis receipt, using a likelihood ratio (LR) test to determine if SDOH collectively improved model fit when added to clinical covariates (age, disease characteristics, comorbidity, infection history, season of year).

Results: Among 4,870 immunosuppressant treatment episodes (83% ≥65 years, 60% female, 88% non-Hispanic White), PJP prophylaxis was used in 26%. The LR test (Chi-Square=81.05, DF=15; p< 0.001) indicated that SDOH measures explained significant variation in prophylaxis. Prophylaxis was more common in patients in the Midwest vs. South (34% vs. 24%; adjusted prevalence ratio [aPR]: 1.39; 95% CI: 1.12, 1.73) and less common in females vs. males (24% vs. 30%) (aPR: 0.80; 95% CI: 0.70, 0.91). Prophylaxis was also less common in patients from counties with persistent poverty (22% vs. 32%), but with borderline statistical significance (aPR: 0.67; 95% CI: 0.44, 1.02).

Conclusions: SDOH were individually and collectively associated with guideline-recommended PJP prophylaxis in AAV patients. Addressing sex and socio-economic disparities could improve use of appropriate infection prophylaxis.