Structured education improves medication knowledge in chronic heart failure

Background

Medication knowledge is essential for self-management in chronic heart failure (HF), yet baseline knowledge and responsiveness to structured education remain insufficiently characterized.

Purpose

To evaluate baseline medication knowledge and the effect of a brief structured educational intervention in ambulatory HF patients.

Methods

Seventy-seven ambulatory HF patients were prospectively assessed. Medication knowledge was measured using a specifically designed set of 40 cards. The cards displayed icons illustrating organ systems, bodily functions, and pharmacological actions; patients were asked to link each medication they were currently taking to the corresponding icon depicted on one of the cards. Knowledge of each medication was categorized as either a match or a mismatch. After the first round, mismatches were explained to the patient within a brief structured educational conversation. Approximately 30 minutes later, the card game was repeated and matches/mismatches were reassessed. “Medication knowledge" was operationalized as the proportion of correct matches relative to the total number of medications assessed. All chronic medications currently taken by the patient, including heart failure and non–heart failure drugs, were included. A generalized linear mixed-effects logistic regression model with random intercept for patient ID was used to assess intervention effects. Baseline group differences were evaluated using appropriate univariable tests. Predictors of percentage knowledge gain at the patient level were analyzed using multivariable linear regression models adjusted for baseline knowledge.

The convenience sample comprised 77 individuals (48 men, 29 women); baseline characteristics are detailed in the table. The median number of medications per patient was 10 (Q1–Q3: 8–13). Median baseline knowledge was 60% (Q1–Q3: 33–86). A total of 1,616 medication-level observations were analyzed across both rounds of card play. The proportion of correct matches increased to 81% (Q1–Q3: 55–100) following the intervention, corresponding to a mean increase of 20 percentage points (95% CI 16–24; p<0.001). In mixed-effects analysis, the odds of correct medication identification post-intervention were significantly higher (OR 3.40, 95% CI 2.62–4.39; p<0.001; ICC=0.33). The magnitude of improvement was consistent across drug classes (interaction p=0.591). Diabetes mellitus was independently associated with smaller knowledge gains (B = −8.1; p=0.014).

Conclusion

A simple structured educational intervention markedly improved medication knowledge across pharmacological classes in ambulatory HF patients. Whether these improvements translate into better adherence or clinical outcomes warrants further investigation.Table1