Leveraging electronic health records to examine differential clinical outcomes in people with Alzheimer's Disease

BACKGROUNDAlzheimers disease (AD) carries a high societal burden inequitably distributed across demographic groups. OBJECTIVETo examine demographic differences and drivers of AD decline using real-world electronic health record (EHR) data with accurate AD population identification. METHODSLeveraging EHR data (1994-2022) from two large independent healthcare systems, we applied a novel unsupervised phenotyping algorithm to predict AD diagnosis and validated using gold-standard chart-reviewed and registry-derived diagnosis labels. Among patients with [&ge;]24 months of EHR data not living in nursing homes pre-AD diagnosis, we performed healthcare system-specific competing risk survival analyses to estimate the time to two readily ascertainable AD decline outcomes (i.e., nursing home admission, death), stratified by demographic groups and accounting for baseline covariates (e.g., age, gender, race, ethnicity, healthcare utilization, and comorbidities). We then performed a covariate-adjusted fixed-effects meta-analysis using inverse variance weighting to estimate the time to AD decline stratified by demographic groups. RESULTSThe algorithm achieved robust prediction in identifying AD patients across both healthcare systems (AUROC score range: 0.835-0.923) and demographic groups. Of the 29,262 AD patients in both healthcare systems (61% women, 90% non-Hispanic White, 79.52{+/-}9.39 years of age at AD diagnosis), 49% entered nursing homes and 48% died during follow-up. In covariate-adjusted fixed-effects meta-analysis, women had a higher risk of nursing home admission (HR[95% CI]=1.061[1.024-1.100], p=.001) and lower risk of death (HR[95% CI]=0.856[0.811-0.904], p<.0001) than men. Non-Hispanic White individuals had similar nursing home risk (HR[95% CI]=1.006[0.952-1.063], p=0.831) but higher death risk (HR[95% CI]=1.376[1.245-1.521], p<.0001) than racial and ethnic minorities. Older age at AD diagnosis and greater pre-existing comorbidity burden increased both nursing home admission and death risk. CONCLUSIONFindings from two large EHR cohorts add to the real-world evidence of demographic differences in clinical AD decline and its drivers, which could potentially inform individual clinical management and future public health policies.