Abstract
Background:
Medication nonadherence is a worldwide issue that can lead to poor clinical outcomes and increased health-care costs.
Objective:
To determine the predictors of medication nonadherence.
Methods:
A retrospective chart review was conducted for patients who received prescription medications from Cleveland Clinic outpatient pharmacies. Prediction variables consisted of demographics, socioeconomic status, number of medications, and number of daily administrations. These variables were analyzed using a logistic regression to determine independent predictors of medication adherence.
Results:
Between January and September 2015, over 300 000 eligible prescriptions were filled, corresponding with over 70 000 unique patients. Of these, 29 134 patients were included. After multivariable regression, increasing age (odds ratio [OR]: 1.01), household income (OR: 1.03), and medication count (OR: 1.05) were found to be associated with adherence. Male gender (OR: 0.88), African American (OR: 0.45), Hispanic (OR: 0.62), or other race (OR: 0.87), being single (OR: 0.92), and increasing frequency of administrations per day (OR: 0.76) were associated with nonadherence.
Conclusion:
Medication nonadherence was associated with nonwhite race, single status, male gender, low socioeconomic status, and increasing frequency of medication administration. Based on these results, a risk prediction tool could be created to determine which patients are at the highest risk of medication nonadherence.
Keywords
Get full access to this article
View all access options for this article.