Abstract Body (Do not enter title and authors here): Background: Medication nonadherence is associated with adverse outcomes in observational studies. However, whether that association persists in a clinical trial designed to improve adherence is uncertain.
Question: We describe factors associated with low adherence among patients enrolled in a mHealth intervention utilizing text messages to improve cardiovascular (CV) medication refills. The relationship between adherence and adverse outcomes was then assessed.
Methods: The NUDGE study was a pragmatic clinical trial randomizing patients to usual care versus different types of text messages (generic, behavioral nudge, behavioral nudge plus an AI chatbot) to improve refill rates when patients had a 7-day delay in a CV medication refill. For the current analysis, the primary exposure of low adherence was determined from pharmacy data and was defined as a proportion of days covered by medication (PDC, as calculated by prescription fill data) ≤0.80 in the first 3 months after enrollment. Adverse outcomes were defined as time to emergency department visit, hospitalization or death. The association between low adherence and adverse outcomes was evaluated using Cox proportional hazards models.
Results: Of the 9268 patients enrolled in the study with prescription fill data, the median PDC was 0.74 (0.19-0.96, 25%-75% percentile). Patients with low adherence comprised 53.5% of the cohort. In this clinical trial, all 3 text messaging intervention arms increased refill adherence at similar rates (~5%) at 3-months compared to usual care. In multivariable analyses, low adherence in the first 3 months was associated with adverse outcomes (HR 1.13; 95% CI 1.06-1.19) regardless of study arm enrollment.
Conclusion: Among patients enrolled in a randomized clinical trial to improve medication refill adherence, low adherence remains common and was associated with adverse outcomes. While this mHealth intervention increased CV medication refills, additional solutions are needed to improve medication adherence.