Evaluating the role of multimorbidity in modulation medication effects in older adults

Multimorbidity, defined as the co-occurrence of two or more medical conditions, impacts two-thirds of older individuals over 65 – corresponding to 36 million U.S. adults, and is a major driver of healthcare spending, polypharmacy, and mortality. However, the routine exclusion of older and more multimorbid patients from clinical trials has resulted in the paucity of data regarding the risks and benefits of medications in this population, or an understanding of how multimorbidity alters treatment effects. To address this unmet need, this proposal will evaluate the role of multimorbidity in modulating medication effects and identify the optimal approach that best quantifies its impact on medication outcomes. Our central hypothesis is that (a) by attenuating drug-related benefits and amplifying drug-related harms, multimorbidity should be a key consideration when making treatment decisions, and that (b) approaches that incorporate the cumulative burden of illness – especially the multi- morbidity weighted index [MWI] – can better characterize these alterations in medication effects (preliminary analysis). The proposal will use Medicare fee-for-service data from >23 million patients and replicate findings in two large external databases. We will focus on cardiometabolic therapies as: older adults have the highest burden of these conditions, and since 2010, more than 20 new cardiometabolic therapies have been approved, highlighting the immense need to study these medications. We will identify patients with: (a) type 2 diabetes initiating sodium glucose co-transporter 2 inhibitors vs established antidiabetic therapies; (b) atrial fibrillation initiating direct oral anticoagulants vs warfarin; and (c) atherosclerotic cardiovascular disease [CVD] initiating newer antiplatelet drugs (e.g. ticagrelor) vs clopidogrel. Aim 1 will evaluate how clinical (e.g. cognitive impairment) and non-clinical (e.g. social deprivation) factors interact with multimorbidity to influence medication prescribing of cardiometabolic therapies in the real world. Aim 2 will elucidate the role of multimorbidity in modulating the risks and benefits for newer compared to established cardiometabolic medications by estimating the adjusted rates of disease specific benefits (i.e. reduction in CVD events), harms (e.g. major bleeding) and universal outcome measures (e.g. home-time, loss of functional independence) by levels of multimorbidity. We will also validate multimorbidity measures (e.g. MWI, Elixhauser index) and frameworks (e.g. disease dyads) against medication outcomes. The impact of this proposal is significant as it will establish a rigorous and readily scalable framework to study the effects of multimorbidity on drug outcomes in older adults. It will also represent the first effort to systematically evaluate and validate multimorbidity indices and approaches against medication outcomes, beginning a new and exciting line of research that has potential to expand to other populations (e.g. middle-aged adults) and clinical areas. Given the paucity of data from clinical trials, study findings will serve as the primary source of information for patients, caregivers, and clinicians to make individualized evidence-based decisions.