Abstract
RATIONALE & OBJECTIVE: Nonadherence to medical regimens increases the risk of graft loss among adolescent and young adult recipients of kidney transplants. Interventions that improve adherence may decrease rejection rates, but their perceived costs are a barrier to clinical implementation. We developed a model to assess the cost-effectiveness of an adherence promotion strategy, the Medication Adherence Promotion System (MAPS). STUDY DESIGN: Simulation-based. Data sources included published articles indexed in Medline or referenced in bibliographies of relevant English-language articles. Data on costs and outcomes were taken from a single clinical center. SETTING & POPULATION: US adolescent patients after their first kidney transplant. INTERVENTION: Usual posttransplant care versus usual care plus MAPS. OUTCOME: Effectiveness measured in quality-adjusted life years (QALYs) and costs measured in 2020 US dollars. MODEL, PERSPECTIVE, & TIMEFRAME: Markov state transition decision model. We used a health care system perspective with a lifelong time horizon. RESULTS: In the base-case analysis, MAPS was more effective and less costly than usual care. MAPS cost $9,106 per patient less than usual care and resulted in a gain of 0.32 QALYs. In probabilistic sensitivity analyses, MAPS was cost saving 100% of the time. Extending results to a program level with 100 patients, any adherence promotion intervention similar in effectiveness to MAPS would cost less than $50,000/QALY if the start-up costs were <$2.5 million and annual costs <$188,000. Strategies with costs similar to MAPS that reduce the risk of rejection by as little as 3% would also have similar cost-effectiveness. LIMITATIONS: Estimates of components and costs for MAPS were based on a single center. CONCLUSIONS: Adherence promotion strategies with costs similar to MAPS can be cost-effective as long as they reduce rejection rates by at least 3%. This model can be applied to study the cost-effectiveness of adherence promotion strategies with varying costs and outcomes.