Despite the frequency of medication non-adherence in clinical trials and its adverse consequences on study interpretability and costs, there has been very little research on ways to improve adherence in clinical trials. We propose to evaluate the effectiveness of two medication adherence improvement interventions for clinical trials that are based on computer technology. Both the Telephone-Linked Computer (TLC) system and the Medication Event Monitoring System (MEMS) Feedback have been shown to be effective in improving medication adherence in clinical settings. In this proposed study, they will be evaluated a national multi-center VA clinical trial that will assess antiplatelet (aspirin) and anticoagulant (warfarin) therapy for preventing myocardial re-infarction and death in patients who have had a myocardial infarction. TLC and MEMS Feedback will each be compared to the usual procedures (UP) employed in the clinical trial to improve adherence with patients randomly assigned to TLC, MEMS Feedback or to UP. TLC users will converse weekly with TLC, an automated system that speaks to patients in their homes over the telephone. TLC monitors medication adherence by patient self-report and gives feedback to motivate medication adherence. TLC sends reports on patient adherence to the patient, their physician and the study coordinator. MEMS Feedback entails collection and storage of daily adherence information obtained by a special bottle cap that automatically records when the bottle is opened. MEMS Feedback patients will return monthly to have the adherence data in the bottle cap displayed on a computer monitor. The study coordinator will show the patient the rate and pattern of non-adherence and suggest alternative cues to help the patient remember to take the medication. The patient will receive a printout of the adherence data and their cue. In this proposal (1) TLC and MEMS Feedback will be adapted for clinical trial use, (2), adherence study evaluation methods will be developed and pre-tested with 40 patients at two VA hospitals, and (3) the adherence study will be conducted at 25 VA sites involving 330 patients. All 1,670 patient subjects accrued to the VA study at these sites through 1/94 will be screened for eligibility for the adherence study by measurement of their adherence using MEMS bottles. Those who take less than 80% of their daily aspirin doses over a 2 month period will be eligible. They will be randomized to TLC, MEMS Feedback or UP and followed for a six month period during which their adherence will be measured by MEMS. The end-point will !be changes in aspirin adherence measured during the 2 months prior to randomization (baseline) compared to Months 5-6 of the intervention. Subgroup analyses will be performed on demographic, comorbidity, side effects, and medication variables. The practicality and acceptability of the two interventions to patients and study coordinators will be measured by questionnaires. The cost effectiveness of the two interventions will be measured by relating annual operating costs to percent change in medication adherence.
