Logo image
Back
Leveraging Electronic Health Dataset to Identify Co‐existing Multimorbidity in Individuals with Alzheimer's Disease in the Midwestern United States
Abstract   Open access   Peer reviewed

Leveraging Electronic Health Dataset to Identify Co‐existing Multimorbidity in Individuals with Alzheimer's Disease in the Midwestern United States

Nai‐Ching Chi, Patrick Lozier, Kathleen Buckwalter, Nick Street, Barbara Rakel, Sue Gardner, Yelena Perkhounkova, Maria Hein, Pui Ying Yew, Chih‐Lin Chi, …
Alzheimer's & dementia, Vol.21(S6), e100675
12/2025
DOI: 10.1002/alz70860_100675
PMCID: PMC12726421
url
https://doi.org/10.1002/alz70860_100675View
Published (Version of record) Open Access

Abstract

Background Alzheimer's Disease (AD) frequently co‐occurs with multiple chronic conditions (MCCs). This study aimed to identify specific combinations of MCCs, utilizing detailed ICD‐10 groupings to inform targeted prevention strategies. Method We analyzed de‐identified electronic health records (EHR) of 2,629 AD patients from the University of Iowa Hospitals and Clinics (2015‐2021). Patient diagnoses were grouped into their first 3‐level ICD‐10 codes. The Apriori algorithm was employed to identify co‐occurring ICD‐10 groups, iteratively analyzing singular, dyad, triad, and tetrad combinations. Support (frequency) and Leverage (correlation) metrics were used to identify significant patterns. Result 46 common ICD‐10 groups were identified. Hypertension (I10; 43.3%), lipoprotein metabolism disorders (E78; 31.5%), major depressive disorder (F33; 23.6%), and diabetes (E11) were the most frequent singular co‐occurrences. Hypertension consistently appeared in the most common dyads (e.g., I10‐E78: 21.4%, I10‐F33: 13.1%, I10‐E11: 10.9%), with I10‐E78 exhibiting the highest correlation (7.8% more frequent than expected). Other highly correlated dyads included major depression‐anxiety (F33‐F41; 4.5%) and lipoprotein metabolism‐ischemic heart disease (E78‐I25; 3.9%). These dyads formed the most common/correlated triads (e.g., I10‐E78‐F33: 7.7%/4.5%, I10‐E78‐E11: 6.7%/4.2%, I10‐E78‐I25: 6.6%/4.5%). Notably, triads involving hypertension‐ischemic heart disease (I10‐I25) with gastroesophageal (K21), lipoprotein metabolic (E78), and sleep disorders (G47) demonstrated frequencies more than two‐fold higher than expected (high correlation). The most common and correlated tetrads included hypertension‐lipoprotein metabolism‐major depression (I10‐E78‐F33) with anxiety (F41), gastroesophageal reflux (K21), sleep disorders (G47), and other hypothyroidism (E03), as well as hypertension‐lipoprotein metabolism‐gastroesophageal reflux‐ischemic heart disease (I10‐E78‐K21‐I25). Conclusion This study identified highly prevalent and correlated combinations of MCCs in AD patients, including well‐established risk factors. These findings have crucial implications for developing targeted disease prevention strategies in this vulnerable population.

Details

Metrics

7 Record Views
Logo image