Relationships between Zip Code Level Social Determinants of Health and Type 2 Diabetes Medication Adherence
Author(s)
Nelson J1, Thiel E1, Marlin T1, Palmer L2
1Merative, Ann Arbor, MI, USA, 2Merative, Cambridge, MA, USA
Presentation Documents
OBJECTIVES: Traditional real-world data sources may be missing Social Determinants of Health (SDoH) that impact health behaviors and outcomes. This study compares geographic location-based SDoH between patients adherent and non-adherent to oral antidiabetes medications.
METHODS: This analysis utilized Merative’s MarketScan Commercial and Medicare Databases supplemented with 5-digit zip code level SDoH data. SDoH and patient attitude/behavior data were collected via surveys administered to individuals across the U.S. and results were then summarized for the respondents’ 5-digit zip code as a normalized number where 100 is the average for all zip codes (>100=higher than average;<100=lower than average). These scores were then linked to the patients’ 5-digit zip codes in MarketScan that is available to Merative researchers. Patients with type 2 diabetes who newly initiated 2nd-line, oral antidiabetes monotherapy between 1/1/17 and 12/31/21 were selected and stratified into adherent vs. non-adherent cohorts based on the Medication possession ratio (≥0.8) during a 12-month follow-up. Cohorts were compared via 2-sided t-test.
RESULTS: A total of 86,970 diabetes patients were analyzed, of which 85,627 lived in zip codes for which SDoH data was available. Adherent and non-adherent sub-cohorts, which represented 53% and 47% of the patients, respectively, had similar average ages (54.7yrs vs. 53.3yrs). The mean relative score for respondents answering “somewhat hard to pay for basic needs” was 99.4 for adherent patients’ zip codes and 101.3 for non-adherent (p<0.001). Similar results were seen for those reporting that it was “highly likely” to delay/not fill a prescription in the next 3 months” (adherent: 99.9; non-adherent: 104.0; p<0.001). Neighborhood safety was more of a concern for zip codes of non-adherent patients (adherent: 94.0; non-adherent: 99.5; p<0.001).
CONCLUSIONS: Adding survey response SDoH data, summarized to the 5-digit zip code level, adds additional insight into the differences between patients who remain adherent to chronic condition medication versus those who do not.
Conference/Value in Health Info
Value in Health, Volume 26, Issue 6, S2 (June 2023)
Code
EPH108
Topic
Study Approaches
Topic Subcategory
Surveys & Expert Panels
Disease
Diabetes/Endocrine/Metabolic Disorders (including obesity), No Additional Disease & Conditions/Specialized Treatment Areas