# Medication adherence in hypertension and diabetes comorbidity: implications for disease control in a population-based study

## TL;DR

This study explores how education, income, and diabetes comorbidity affect medication adherence in hypertension patients, revealing nonlinear patterns and the need for targeted interventions.

## Contribution

The study identifies nonlinear education effects and subgroup differences in medication adherence among hypertension patients with diabetes comorbidity.

## Key findings

- Medication adherence strongly predicts hypertension control (aOR = 1.35).
- Adherence is higher in patients with comorbid diabetes (aOR = 1.52).
- Education has inverted U-shaped associations with adherence, peaking around 4–8 years of schooling.

## Abstract

Medication adherence is essential for effective hypertension management, yet its social and behavioral determinants remain incompletely understood. In particular, the influence of education and comorbid diabetes on adherence is unclear.

To examine factors associated with blood pressure control and medication adherence among patients with hypertension, with emphasis on nonlinear educational effects and subgroup differences by sex and comorbidity.

Using a multistage random sampling design, we recruited 40,037 adults with physician-diagnosed hypertension across all 11 prefecture-level cities in Zhejiang. Participants were identified through electronic chronic disease registries and completed standardized questionnaires. Blood pressure control was defined according to national guidelines, and medication adherence was measured by self-reported consistent use of prescribed antihypertensive medication in the past 2 weeks. Multivariable logistic regression models were fitted to identify associated factors, and restricted cubic spline and quadratic models were applied to assess nonlinear education–adherence relationships.

Medication adherence strongly predicted hypertension control (aOR = 1.35, 95% CI: 1.25–1.46). Adherence was higher among patients with comorbid diabetes (aOR = 1.52, 95% CI: 1.39–1.66) and those with higher income, but declined at the highest education levels. Nonlinear analyses revealed inverted U-shaped associations, with thresholds around 4–8 years of schooling. These patterns differed by sex and comorbidity status.

Education exerts threshold-dependent effects on adherence, modified by sex and diabetes comorbidity. By uncovering nonlinear and subgroup-specific patterns, this study extends prior evidence and underscores the need for tailored interventions to reduce disparities in hypertension management.