Treatment Adherence and Its Associated Factors in Patients with Type 2 Diabetes: Results from the Rio de Janeiro Type 2 Diabetes Cohort Study

OBJECTIVES: To investigate treatment adherence in patients with type 2 diabetes and to evaluate its associated factors. METHODS: The Summary of Diabetes Self-Care Activities (SDSCA) questionnaire was used to assess treatment adherence. Good adherence was defined as ≥5 days a week in each SDSCA item. Pain, emotional, and physical domains of the SF-36 quality of life questionnaire and the Canadian Occupational Performance Measure (COPM) were also evaluated. Multivariable logistic regressions explored the independent correlates of good general adherence and of specific items of the SDSCA (diet, exercise, and medications). RESULTS: Good adherence was 93.5% for medication use, 59.3% for foot care, 56.1% for blood glucose monitoring, 29.2% for diet, and 22.5% for exercise. Patients with general good adherence had lower BMI, better serum lipid profile, higher values of functional capacity, emotional and pain domains of SF-36, better occupational performance, and lower prevalence of pain or limitation in the upper and lower limbs than patients with worse adherence. The variables associated with good adherence were younger age, lower BMI, presence of macrovascular complications, better occupational performance and emotional domain of SF-36, and higher HDL cholesterol levels. The presence of pain/limitation in the upper limbs was associated with worse adherence. Good medication adherence was associated with longer diabetes duration, lower BMI, and lower HbA(1c) levels. Higher values of pain and emotional domains of the SF-36 and lower BMI were related to better exercise and diet adherence, while the presence of peripheral neuropathy and joint pain/limitation were associated with worse exercise adherence. CONCLUSIONS: Emotional and physical performances are important determinants of good diabetic treatment adherence. Good adherence has beneficial impact on BMI, lipid, and glycemic control.