Impact of social determinants of health (SDOH) measures on prescribing advanced androgen blockade (AAB) in patients with metastatic hormone sensitive prostate cancer (mHSPC).

Abstract

e17086 Background: Use of AAB (abiraterone acetate, apalutamide or enzalutamide) has been shown to improve survival in mHSPC, but many patients are not prescribed this treatment. There is substantial interest in the broader role of SDOH in cancer. We explored how SDOH influences prescribing patterns of AAB in patients with mHSPC. Methods: Patients diagnosed with mHSPC between 1/1/2017 and 12/31/2021 were identified using the iKnowMed electronic health record database from The US Oncology Network of community oncology practices. Records were searched for prescriptions for AAB (based on intent to treat). Individual level measures were age, diagnosis year, ECOG, race and type of health insurance. Area level measures were Area Deprivation Index (ADI) at national and state level, and rural status. ADI is a validated metric based on demographic variables from census block groups; high ADI scores for state ( > 8) and national ( > 80) level are markers of low socioeconomic status. Logistic regression models were run on each SDOH variable and adjusted for confounding variables, including a joint distribution model of African American (AA) and ADI (significance at p < 0.05). Results: There were 3,855 patients identified with mHSPC: 40% had a prescription for AAB. Summary measures overall were mean age: 71 years, AA: 10%, Medicaid: 5%, ECOG 0-1 and 2+: 60% and 14% respectively; and diagnosis year 2017-21: 15%, 20%, 21%, 22%, and 21% respectively. Area level scores: National ADI high: 8%, State ADI high: 18%, rural: 5%. Interaction variable: AA + State ADI (high): 2%. Statistical differences between those with/without AAB (not shown) were diagnosis year, ECOG and age. Multivariate logistic models were adjusted for these 3 variables with the single SDOH measure as the primary independent variable and the binary variable AAB as the dependent variable (Table). Conclusions: There was no significant association between AAB prescriptions and AA race, rural status, ADI (state or national), or Medicaid status. These data indicate that SDOH measures do not appear to influence prescribing of these treatments. We believe this study is among the first to examine these particular SDOH measures and their relationship to the prescribing of oncology treatments. Further research should be conducted into the impact of SDOH measures on the fulfillment and compliance of these drugs. [Table: see text]