NEW HAVEN, CT — More than half of older adults with Alzheimer’s Disease and Related Dementias (ADRD) are undiagnosed, limiting timely access to person-centered care, according to a new study.
“Therefore, clinicians, researchers, and population health managers need scalable, reproducible approaches to monitor both prevalence and diagnostic gaps,” according to a study published in the Journal of the American Geriatric Society.1
The Yale School of Medicine-led authors said they evaluated whether “a decision-analytic modeling framework can translate a limited number of clinician-adjudicated cases of ADRD into a probabilistic computational phenotype for accurate, population-level assignments of probable ADRD in the emergency department (ED) setting using routinely collected electronic health record (EHR) data.” The James J. Peters VAMC in Bronx, NY, participated in the study.
The retrospective cohort study involved 5,000 adults aged 65 or older from nine EDs within a large integrated health system (2014-2022). “We randomly selected 500 individuals for clinician adjudication of dementia status (reference cohort), reserving the remaining 4,500 as a phenotyping cohort,” according to the study team. “We developed the phenotype as a logistic regression model trained on adjudicated cases, embedding pattern-mixture multiple imputation to address information bias. We applied decision-curve analysis to evaluate clinical utility across probabilistic thresholds. We applied the phenotype to assign dementia status to 4,500 unadjudicated patients and compared clinical characteristics to adjudicated cases.”
Participants’ mean age was 77.4 years old; 55.4% were women; and 102 individuals (20.4%) had clinician-adjudicated ADRD. The results indicated that the model demonstrated good discrimination (AUROC 0.87; 95% CI 0.82-0.91). The researchers advised that decision-curve analysis revealed net clinical benefit across examined thresholds (predicted probabilities 12%-32%), identifying an additional 16-18 probable ADRD cases per 100 older adults.
Among those without ADRD-related diagnosis codes, net benefit ranged from 8 to 13 additional correct identifications per 100, they said, with phenotype-assigned cases closely resembling clinician-adjudicated cases (standardized mean differences ≤ 0.20).
“A probabilistic computational phenotype derived from routinely collected EHR data accurately reproduced clinician-adjudicated ADRD status and demonstrated net clinical benefit, including among ED patients whose ADRD was not captured by diagnosis codes,” the authors concluded. “Adoption of this replicable framework may enable healthcare organizations to strengthen ADRD surveillance and reduce underdiagnosis.”
- Festa N, Alexovitz K, Sifnugel N, Cohen I, Faustino IV, Khasnavis S, Young J, Iscoe M, Mecca AP, Han L, Hwang U. Assigning Probable Dementia Status Using Routinely Collected Electronic Health Record Data. J Am Geriatr Soc. 2026 Aug;74(8):2326-2337. doi: 10.1111/jgs.70562. Epub 2026 Jul 6. PMID: 42410282; PMCID: PMC13496366.
