OBJECTIVES : Poor medication adherence (MA) is associated with inadequate glycemic control. We assessed association between patient characteristics and patient activation (PA) and association between PA and MA among Medicare beneficiaries with type 2 diabetes.

METHODS : Data from 2009 through 2013 Medicare Current Beneficiary Survey (MCBS) was used. Inclusion criteria were type 2 diabetes and 18 years or older. Exclusion criteria were missing responses to more than half of PA items, no continuous Medicare Part A or Part D coverage, less than 2 Part D claims for an antidiabetic medication, insulin use, not community-dwelling, Alzheimer’s disease, dementia, mental retardation or mental disorders. Based on PA scores, PA was categorized as “low,” if less than the mean minus 1/2 standard deviation; “high,” if greater than the mean plus 1/2 standard deviation; or “moderate,” otherwise. Proportion of days covered (PDC) ≥ 0.80 in the 6 months after PA measurement was considered adherent. Weighted ordinal logistic regression in SAS 9.4 for Unix assessed association between patient characteristics and PA and association between PA and MA. An alpha of 0.05 was used.

RESULTS : Among 571 individuals included, 27.5% had low activation, 38.7% had moderate activation, and 33.7% had high activation. Approximately three-fourths was adherent to antidiabetic medications. Low activation was more common in males (Odds ratio, O.R. = 1.83, 95% confidence interval [1.21 – 2.78], p < 0.01) and patients without arrhythmia (O.R. = 1.55 [1.02 – 2.34], p = 0.04). Patients without high school education were 7.4 [3.24 – 16.88] times more likely to have low activation than those with post-graduate degree (p < 0.01). In multivariable analysis adjusting for race, gender, osteoporosis, Charlson Comorbidity Index score, and number of prescribed medications, PA level was not significantly associated with MA.

CONCLUSIONS : No association was found between PA level and MA among Medicare beneficiaries with type 2 diabetes.