ABSTRACT Concern about long-term cognitive impairment following exposure to surgery and anesthesia has persisted over the past two decades, in part because there is no conclusive evidence; large-scale studies have not definitively demonstrated a lack of association between exposures to anesthesia and surgery and subsequent occurrences of Alzheimer’s disease and related disorders (ADRD), nor have they demonstrated strong evidence suggesting there is increased risk. This proposal seeks to better explore the association between operative exposure and neurocognitive decline through a new analysis of existing Medicare data that has the potential to overcome limitations of previous studies. This study uses a new solution for accessing and analyzing Medicare claims (the Virtual Research Data Center (VRDC)), allowing researchers, in an economi- cally feasible manner, to analyze all CMS data since 1999 through a secure remote server. By (1) utilizing a natural experiment that reflects an exposure to surgery (the event of appendectomy) but is not associated with lifestyle or disease related treatments that may themselves be associated with long-term changes in cognition, and (2) through the use of multivariate matching to account for potential observable differences in cohorts, we have the ability to compare an exposed cohort of 80,000 appendectomy cases with matched controls that are closely balanced for both observable and potentially unobservable risk factors. This enables us to better conduct a large-scale study determining if there is a long-term association between operative exposure and the rate of subsequent neurocognitive decline. Aim 1 will refine claims-based definitions of diagnoses that identify neurocognitive decline through validation of Medicare claims data against neurocognitive test results collected on samples of Medicare beneficiaries by the Health and Retirement Study, the National Health and Aging Trends Study, and clinical assessment data in the Long Term Care Minimum Data Set. Aim 2 will use these definitions to conduct a multivariate matched cohort study to compare the hazard for developing neurocognitive decline in patients who experienced the random event of appendicitis leading to an appendectomy versus controls who did not undergo an appendectomy. We hypothesize that cases with appendectomy will have an elevated hazard of neurocognitive decline as compared to controls. We will identify all patients who underwent appendectomy between ages 68 to 77 and match each case to 5 controls who were the same age and very similar on risk factors for neurocognitive decline such as diabetes, stroke and hypertension, as well as an extensive list of comorbidities, at the same point in time as the case developed appendicitis, while exactly matching patients who already show diagnostic codes for neurocognitive abnormalities and also controlling for anesthesia exposure prior to the match. Outcomes that will be examined by studying the hazard ratio of exposed to control groups will include: diagnoses of neurocognitive decline, nursing home admission, neuro- cognitive testing at the nursing home, drugs that specifically indicate neurocognitive decline, and mortality. 1

Project Narrative: This proposal seeks to better explore the association between operative exposure and subsequent incidence of Alzheimer’s Disease and Related Disorders (ADRD) through a new analysis of existing Medicare data that has the potential to overcome limitations of previous studies. By taking advantage of the random nature of appendicitis, we can reduce bias associated with indications for surgery that are also potential confounders for either increased risk of dementia (such as cancer surgery) or decreased risk (such as shoulder surgery). By fully incorporating the entire national experience of elderly Medicare beneficiaries, we can analyze a very large population to convincingly contribute to answering this worrisome question that has gained much attention in the medical community and the general public. 1