Many factors contribute to medication nonadherence including psychological and memory disorders, aging, and pill burden. The Automated Home Medication Dispenser (AHMD) is a medication management system intended to help solve unintentional medication nonadherence.
The purpose of this study was to determine if use of the AHMD improved medication adherence.
We conducted a 6-month prospective, feasibility study assessing use of the AHMD in 21 patient-caregiver dyads. Patients were referred by their physician because of poor medication adherence and included if they resided in Rochester, NY and on at least two medications in pill form. Pill counts were performed at baseline to assess previous adherence. Prospective medication adherence was assessed using AHMD recorded dosing information. A paired t-test was used to compare previous and prospective adherence.
The mean age of patients was 75.1 years. Fifteen patients (71.4%) and eight caregivers (38.1%) were women; half (47.6%) of caregivers lived with the patient. The most common patient comorbidities were hypertension (76.2%) and memory disorder (61.9%). Mean adherence increased from 49.0% at baseline to 96.8% after 6 months of AHMD use (p < .001).
In a cohort of unintentionally nonadherent patients, use of the AHMD for 6 months significantly improved medication adherence.
For more information on this article, contact Jack Brown at firstname.lastname@example.org.
The study was not funded by any sponsor. However, the devices under study (MedaCube) were supplied by PharmAdva.
Presented at the 2016 American Geriatrics Society Annual Scientific Meeting; May 18–21, 2016; Long Beach, CA.
Study concept and design were contributed by C. Hoffmann, A. Schweighardt, K. M. Conn and J. Brown. Acquisition of subjects and data collection were contributed by C. Hoffmann, A. Schweighardt, K. M. Conn, D. Nelson, R. Barbano, F. Marshall and J. Brown. Analysis and interpretation of data were contributed by C. Hoffmann, A. Schweighardt, K. M. Conn and J. Brown. Initial draft and final revision of manuscript were contributed by C. Hoffmann, A. Schweighardt, K. M. Conn, D. Nelson, R. Barbano, F. Marshall and J. Brown.
The authors declare no conflicts of interest.