Psychosocial factors predict medication adherence in young adults with youth‐onset type 2 diabetes: Longitudinal results from the TODAY2 iCount study

Abstract Aim To identify psychosocial predictors of medication adherence in young adults with youth‐onset type 2 diabetes in the Treatment Options for type 2 Diabetes in Adolescents and Youth (TODAY2) cohort. Methods Participants (mean age: 26 years) completed validated psychosocial measures. Unannounced telephone pill counts were completed at T1 (baseline) and T2 (follow‐up, approximately 1 year later) to assess adherence to oral hypoglycaemia agents (OHAs). Adherence to insulin was assessed by self‐report. Logistic and linear regressions identified factors that predicted ‘low adherence’ (<80% of pills/insulin) and per cent adherence, adjusted for potential confounders. Results Of 171 participants with OHA adherence scores at T1 and T2 (65% women, 43% Hispanic and 35% non‐Hispanic Black), 65.4% were low adherent. After adjustment (including T1 adherence), concerns about diabetes medicines (adverse effects, dependence) at T1 predicted higher odds of being low adherent (categorical) at T2 ( p = 0.019). Housing insecurity ( p = 0.045) and reporting ≥2 need insecurities ( p = 0.027) at T1 predicted lower per cent adherence (continuous) at T2. Of 157 participants with insulin adherence scores at T1 and T2 (69% women, 38% Hispanic and 38% non‐Hispanic Black), 36.3% were low adherent. After adjustment (including T1 adherence), beliefs that medicines are overused predicted higher odds of insulin low adherence at T2 ( p = 0.013), and beliefs that medicines are harmful ( p = 0.004) and overused ( p = 0.010) predicted lower per cent insulin adherence at T2. Conclusions Suboptimal medication adherence, common in young adults with youth‐onset type 2 diabetes, is predicted by interfering beliefs about medicines and social factors. We must address these beliefs and unmet needs to develop tailored interventions for this vulnerable group.