Relationships between beliefs about medications and use of prescribed chronic medications

Hemant M Phatak, Purdue University


The purpose of this study was to assess associations between medication beliefs and medication adherence among users of prescribed chronic medications. Under a modified self-regulatory framework, medication adherence is considered as a coping strategy and medication beliefs as factors influencing medication adherence. Patients' medication beliefs were assessed using the Beliefs about Medicines Questionnaire (BMQ) and medication non-adherence was assessed using the Morisky Medication Adherence Scale. Associations between medication beliefs and medication non-adherence were assessed using Pearson correlation and regression analysis. Strengths of associations between beliefs and non-adherence were compared across patients using a different number of chronic medications using Fisher-transformed Pearson correlation estimates and pairwise Tukey's comparison test. Path analysis with latent variables was used to assess fit of the factor structure of associations between medication beliefs and medication non-adherence. Data were collected at outpatient pharmacy of a primary care clinic affiliated with a hospital in Indianapolis, IN. Study participants were sought from patients waiting to see their pharmacists. Eligibility criteria were continuous use of prescribed medications for at least two months prior to completion of the survey and ability to comfortably read and complete a survey in English. A response rate for this study was 78.3 percent. Specific necessity to use medications had a negative but insignificant association with medication non-adherence; whereas, specific concerns related to medications, perceived general overuse of medications by prescribers and general harmful nature of medications had significant positive associations with medication non-adherence. When relative strengths of associations between medication beliefs and non-adherence were assessed, only specific-necessity and specific-concerns exhibited significant associations with medication non-adherence. Strengths of associations between beliefs and non-adherence were similar across patients using different numbers of chronic medications, which indicated that the BMQ effectively assessed medication beliefs across patients using different number of chronic medications. Path analysis indicated that specific-necessity, specific-concerns and general-overuse were able to predict 26.3 percent of the variance in medication non-adherence. General-harm and specific-necessity accounted for 54.8 percent of variance in specific-concerns. Medication beliefs assessment can be used to detect problems related to medication use.




Thomas, Purdue University.

Subject Area

Pharmaceuticals|Public health|Biostatistics

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server