Home Care Medication Management for the Frail Elderly

For many older adults, living independently at home depends on managing complicated medication regimens. Older adults are the most frequent users of prescription drugs; however, physiologically they are most vulnerable to adverse drug reactions. Recently developed devices are designed to assist older adults with medication administration using cost effective methods. The MD.2 is one such device. It includes both audible and visual prompts for medication administration, stores and dispenses medications, and provides a monitoring device to promote client adherence, including notification of a provider if a dose is missed. The purpose of this study is to determine whether a home care medication management program which includes nurse coordination and use of the MD.2 medication-dispensing machine will affect older adults health outcomes, satisfaction, use of health care services, and health care costs over a one year period. We propose a longitudinal three group repeated measures design, enrolling, and randomly assigning, clients who are discharged from a home health care agency with documented problems in medication management. One group will receive the MD.2 medication dispensing device and nurse coordination, the second group will receive a Mediset: a simple box that has separate compartments for individual medication times over the course of a week plus nurse coordination, and the final group will receive Usual Care. A sample size of 339 is needed (113 per group): Up to 453 subjects will be enrolled over 27 months. Health outcomes to be measured at enrollment and every three months include general physical and mental health, depression, cognition, and functional status. Client satisfaction will also be assessed. Health care utilization, costs of care, and cost-effectiveness will be assessed using Medicare claims data. Statistical analyses will involve several regression methods including Poisson regression for count data, Cox proportional hazards regression for time-to-event data, and ordinary least squares. It is important to evaluate new methods of helping older adults to manage their medications. If medication management machines can be shown to be safe and effective, quality of life will be improved, and unnecessary nursing home admissions, hospitalizations, and emergency department visits may be prevented.