Assessment of the Utilization and Effectiveness of Alzheimer's Disease and Related Dementia Treatments

This dissertation investigates the real-world utilization patterns and effectiveness of symptomatic treatments for Alzheimer’s Disease and Related Dementias (ADRD). Despite the introduction of disease-modifying therapies, symptomatic treatments remain the primary strategy for managing ADRD symptoms. These include acetylcholinesterase inhibitors (AChEIs), including donepezil, galantamine, and rivastigmine, and the N-methyl-D-aspartate receptor antagonist or memantine. However, there are significant gaps in understanding their treatment utilization patterns, long-term effectiveness among these treatments, and comparative effectiveness among different treatment durations. The dissertation comprises six chapters. The first chapter reviews the background literature on ADRD, discussing the disease's epidemiology and the current evidence of symptomatic treatments, including the current guidelines, benefits, and harms from using them. The review identifies several gaps in the literature, such as the limited exploration of treatment patterns beyond initial adherence, the lack of comparative effectiveness among the drugs, and consensus on optimal treatment duration. These gaps informed the three primary aims of this dissertation: (1) to describe the utilization patterns of symptomatic treatments; (2) to examine the long-term effectiveness of symptomatic treatments on neuropsychiatric and functional outcomes; and (3) to assess the impact of treatment duration on neuropsychiatric and functional outcomes. The second chapter addresses the first aim by describing the utilization patterns of symptomatic treatments among patients with ADRD. The study revealed low persistence and high reinitiation rates among ADRD patients, with healthier individuals with mild dementia more likely to adapt advanced treatments. These findings highlighted the dynamic use of ADRD symptomatic treatments, which may be related to the temporal treatment effectiveness, treatment side effects, and disease progression over time. The third chapter addresses the second aim regarding the long-term effectiveness of symptomatic treatments on neuropsychiatric and functional outcomes, using proxies of the initiation of antipsychotics and incident falls or related injuries. To effectively account for time-varying confounding, inverse probability weighted marginal structural model was used. The study found no significant differences in the risks of initiating antipsychotic between memantine and AChEI users, but memantine users exhibited a higher risk of falls compared to those using donepezil. Additionally, significant differences were noted between rivastigmine formulations: patients using the rivastigmine patch had a higher risk of antipsychotic initiation than those using donepezil, while oral rivastigmine users showed a reduced risk. These findings informed the necessity for careful selection of treatment regimens based on their differential effectiveness, and further research is warranted to validate these observations. The fourth chapter addresses the third aim which examined the impact of different treatment durations on neuropsychiatric and functional outcomes. To correctly accounted for immortal time bias, the clone-censor-weight approach was utilized. The study identified a 4-6 month treatment period as potentially optimal for balancing therapeutic benefits and side effects. Shorter treatment durations may be associated with higher risks of antipsychotic initiation and falls, due to insufficient treatment effects, while longer durations may lead to increased side effects and diminishing treatment effects. However, the absence of data on reasons for discontinuation limited the depth of insight we could derive from the results. The final chapter of this dissertation summarizes the main findings from each study aim and discusses their clinical implications. This research findings narrowed the gap in understanding ADRD management by providing evidence-based insights into the utilization and effectiveness of symptomatic treatments. The findings reflect the complexity of ADRD care and highlight the necessity for personalized, patient-centered approaches that balance therapeutic benefits, side effects, and personal needs. Future research should prioritize longitudinal studies with comprehensive datasets to evaluate a wider range of outcomes and gather detailed information on reasons for treatment discontinuation.