Relationships between beliefs about medications and nonadherence to prescribed chronic medications

Ann Pharmacother. 2006 Oct;40(10):1737-42. doi: 10.1345/aph.1H153. Epub 2006 Sep 19.


Background: Medication beliefs of patients with a specific medical condition have been associated with nonadherence to drugs used to treat that condition. However, associations between medication beliefs and nonadherence of individuals on chronic, multiple medications have not been studied.

Objective: To investigate associations between patients' medication beliefs and nonadherence to chronic drug therapy.

Methods: A cross-sectional, self-administered survey of patients waiting to see pharmacists at an outpatient pharmacy in a primary care clinic was conducted. Participants' medication beliefs were assessed using the Beliefs about Medicines Questionnaire, and nonadherence was assessed using the Morisky Medication Adherence Scale. Pearson correlation analysis was used to assess bivariate associations between medication beliefs and nonadherence. Regression was used to assess relative strength of associations between various medication beliefs and nonadherence and also to assess the significance of the interactions between those beliefs and nonadherence.

Results: There were positive bivariate associations between specific concerns about medications (p < 0.001), perceived general harmful effects of medications (p < 0.001), and perceived overprescribing of medications by physicians (p < 0.001) and medication nonadherence. When relative strength of associations between each medication belief and nonadherence was assessed, while controlling for other medication beliefs, specific-necessity (p = 0.02) and specific-concerns (p = 0.01) exhibited significant negative and positive associations with nonadherence, respectively. All two-way interactions between variables in the model were insignificant. A model consisting of age, total number of drugs used, and medication beliefs, that is, specific-necessity, specific-concerns, general-overuse, and general-harm, accounted for 26.5% of variance. Medication beliefs alone explained 22.4% of variation in nonadherence to chronic drug therapy.

Conclusions: Patients' medication beliefs explained a significant portion of variation in medication nonadherence.

Publication types

  • Comparative Study

MeSH terms

  • Adult
  • Cross-Sectional Studies
  • Culture*
  • Data Collection
  • Drug Prescriptions*
  • Female
  • Humans
  • Male
  • Middle Aged
  • Patient Acceptance of Health Care
  • Patient Compliance*
  • Pharmaceutical Preparations / administration & dosage


  • Pharmaceutical Preparations