Effect of mortality in cost-effectiveness modeling of disease-modifying treatment for Alzheimer's disease

Abstract

INTRODUCTION: We examined (1) the magnitude of mortality attributed to Alzheimer's disease (AD), and (2) the effect of mortality in cost-effectiveness modeling of hypothetical disease-modifying treatment (DMT) in AD. METHOD: Data were derived from Swedish Dementia Registry (N = 39,308). Mortality was analyzed with survival analysis and multinomial logistic regression. A Markov microsimulation model was used to model the cost effectiveness of DMT using routine care as a comparator. Three scenarios were simulated: (1) indirect effect, (2) no effect on overall mortality, (3) indirect effect on AD-related mortality. RESULTS: Overall mortality increased with cognitive decline, age, male sex, number of medications used, and lower body mass index. Nearly all cause-specific mortality was associated with cognitive decline. DMT increased survival by 0.35 years in scenario 1 and 0.14 years in scenario 3. DMT with no mortality effect is the least cost effective. DISCUSSION: The results provide key mortality estimates and demonstrate influences on the cost effectiveness of DMT. Highlights: We describe cause-specific mortality in relation to disease severity in Alzheimer's disease (AD). We model different assumptions of disease-modifying treatment (DMT) on AD survival. DMT was the least cost effective when assuming no effect on AD survival. Cost effectiveness is mainly influenced by the relative cost of staying in each disease state.