642-P: Discovering Nonadherence to Diabetes Medications—Which Questions Are Most Effective?

Abstract

Half of patients with diabetes have difficulty taking medications as directed, creating a major barrier to glycemic control. In studies outside of diabetes, specific, negatively-framed questions (e.g. “Are you missing doses?”) were most effective in eliciting patient nonadherence disclosure. Therefore, we performed the first study examining how primary care providers (PCPs) evaluate nonadherence to diabetes medications. We analyzed audio-recorded PCP visits for 341 adults with diabetes from 4 geographically diverse Veterans Affairs medical centers from 2017-2020. This analysis focused on patients who reported substantial nonadherence on a post-visit questionnaire, defined as taking their diabetes medications “some of the time,” “hardly ever,” or “never.” Using an established framework, two coders independently classified PCP’s medication-related questions as broad (e.g. “How are you doing with your medications?”), clarifying of dose/regimen (e.g. “Is insulin still 30 units?”), positively-framed (e.g. “Are you taking your medications every day?”) or negatively-framed. We analyzed associations between question type and nonadherence disclosure using Fisher’s Exact tests. Results: there were 41 patients with substantial nonadherence who were seen by 25 PCPs. Fewer than half of these patients (19/41; 46%) disclosed nonadherence to their PCP. PCPs asked 42 medication-related questions: most were clarifying (37/42; 88%), 2 were broad, 2 positively-framed and 1 negatively-framed. Patients disclosed nonadherence more often after specific positively- or negatively-framed questions (100% disclosed) vs. other question types (18% disclosed), p=0.008. In conclusion, over half of patients with substantial nonadherence to diabetes medications did not disclose this to their PCP. While PCPs often reviewed patients’ medications and dosing, this rarely uncovered nonadherence. Asking specific questions about nonadherence was infrequent, yet is likely a more effective strategy. Disclosure S.J.Pilla: None. N.M.Maruthur: Other Relationship; Johns Hopkins HealthCare Solutions. S.Saha: None. M.T.Bugayong: None. J.A.Long: None. H.S.Gordon: None. J.T.Bates: None. D.Washington: None. B.G.Bokhour: None. A.Tuepker: None. M.Beach: None. Funding National Institute of Diabetes and Digestive and Kidney Diseases (K23DK128572); U.S. Department of Veterans Affairs (IIR 14-007)