Background: Reliable approaches for measuring antihypertensive medication compliance in the outpatient setting are not readily available. The objective of the current study was to determine the reliability of the Hill-Bone Compliance Scale among elderly hypertensive patients.
Methods: We conducted a cross-sectional survey of community-dwelling patients attending the hypertension section of the Internal Medicine Clinic in a large multispecialty group practice. Participants (n=239) completed a self-administered questionnaire consisting of demographic questions and the Hill-Bone Compliance to High Blood Pressure Therapy Scale, which includes a nine-item medication compliance subscale.
Results: The mean age of respondents was 69 years; 51% of patients were men, 73% were white, 86% had at least a high school education, and 61% were married. The Cronbach alpha was 0.68 for the medication compliance subscale. All nine items of the medication compliance subscale maintained higher correlations with their own subscale total than with the salt intake and appointment keeping subscale totals. After adjusting for other demographic variables, the odds ratio (95% confidence interval) of perfect medication compliance as reported on the medication compliance subscale was 1.71 (0.95-3.07) for participants 65 years of age and older versus those younger than 65 years of age, 2.53 (1.37-4.66) for whites versus nonwhites, 1.27 (0.73-2.20) for males versus females, 1.30 (0.73-2.29) for married versus unmarried participants, and 1.63 (0.74-3.62) for those with at least a high school education versus those with less education.
Conclusion: The medication compliance subscale of the Hill-Bone Compliance Scale appears reliable and may be a useful tool for detecting noncompliant patients in outpatient settings.