Identifying Medicare beneficiaries with dementia

Abstract

Background/Objectives No data exist regarding the validity of International Classification of Disease (ICD)‐10 dementia diagnoses against a clinician‐adjudicated reference standard within Medicare claims data. We examined the accuracy of claims‐based diagnoses with respect to expert clinician adjudication using a novel database with individual‐level linkages between electronic health record (EHR) and claims. Design In this retrospective observational study, two neurologists and two psychiatrists performed a standardized review of patients' medical records from January 2016 to December 2018 and adjudicated dementia status. We measured the accuracy of three claims‐based definitions of dementia against the reference standard. Setting Mass‐General‐Brigham Healthcare (MGB), Massachusetts, USA. Participants From an eligible population of 40,690 fee‐for‐service (FFS) Medicare beneficiaries, aged 65 years and older, within the MGB Accountable Care Organization (ACO), we generated a random sample of 1002 patients, stratified by the pretest likelihood of dementia using administrative surrogates. Intervention None. Measurements We evaluated the accuracy (area under receiver operating curve [AUROC]) and calibration (calibration‐in‐the‐large [CITL] and calibration slope) of three ICD‐10 claims‐based definitions of dementia against clinician‐adjudicated standards. We applied inverse probability weighting to reconstruct the eligible population and reported the mean and 95% confidence interval (95% CI) for all performance characteristics, using 10‐fold cross‐validation (CV). Results Beneficiaries had an average age of 75.3 years and were predominately female (59%) and non‐Hispanic whites (93%). The adjudicated prevalence of dementia in the eligible population was 7%. The best‐performing definition demonstrated excellent accuracy (CV‐AUC 0.94; 95% CI 0.92–0.96) and was well‐calibrated to the reference standard of clinician‐adjudicated dementia (CV‐CITL <0.001, CV‐slope 0.97). Conclusion This study is the first to validate ICD‐10 diagnostic codes against a robust and replicable approach to dementia ascertainment, using a real‐world clinical reference standard. The best performing definition includes diagnostic codes with strong face validity and outperforms an updated version of a previously validated ICD‐9 definition of dementia.

Development and Validation of a Deep Learning Model for Earlier Detection of Cognitive Decline From Clinical Notes in Electronic Health Records

“…Most studies that detected cognitive decline used structured EHR data and focused on later stages of cognitive decline. 10 , 11 , 12 , 22 Two studies used free-text notes for detecting MCI and/or dementia. 11 , 12 The first used the Mayo Clinic Study of Aging population to assess keyword search for identifying patients with cognitive impairment and dementia among intensive care unit patients who received formal cognitive evaluation.…”

“…Prior studies primarily focused on the stages of cognitive decline from MCI to dementia. 10 , 11 , 12 Limited research has focused on detection of early cognitive decline preceding MCI or use of unstructured EHR data (clinical notes).…”

Development and Evaluation of a Natural Language Processing Annotation Tool to Facilitate Phenotyping of Cognitive Status in Electronic Health Records: Diagnostic Study

“…Several algorithms have been developed for phenotyping cognitive status; some studies used structured data, such as diagnosis codes, missed appointments, or health care utilization patterns [ 15 , 23 ], whereas others have applied natural language processing (NLP) to unstructured notes [ 18 , 19 , 24 ]. None of these prior efforts combined both structured and unstructured input modalities, and manual annotation by clinical experts is limited by the lack of available tools to facilitate efficient chart review [ 25 ].…”

“…It highlights key information and presents a summarized view to the annotator. We evaluated NAT in two EHR data sets: (1) Medicare beneficiaries from the Mass General Brigham (MGB) Accountable Care Organization (ACO) who were labeled in another study using manual chart reviews [ 15 ] and (2) MGB patients with laboratory confirmed SARS-CoV-2 (a case-control study to investigate the effects of COVID-19 on people with and without HIV was used as an exemplar of a research cohort that requires labeling of cognitive status). We evaluated interrater agreement in the first data set and compared it to interrater agreement in Epic—the EHR system used at MGB since 2015.…”

The persistent disparity in brain health among aging people with HIV

“…This bias would likely underestimate the magnitude of the disparity in dementia rates between PWH and PWoH. Furthermore, dementia adjudication using ICD codes increases diagnostic error [32,33], and it is uncertain whether there may exist biases or differential attribution of dementia by clinicians to PWH vs. PWoH. Studies dedicated to evaluate cognition in PWH may help clarify these uncertainties.…”