In:
American Journal of Epidemiology, Oxford University Press (OUP), ( 2023-08-30)
Abstract:
We developed and validated a claims-based algorithm that classifies patients into obesity categories. Using Medicare (2007-2017) and Medicaid (2000-2014) claims data linked to two electronic health records (EHR) systems in Boston, Massachusetts, United States, we identified a cohort of patients with an EHR-based BMI measurement. We used regularized regression to select from 137 variables and built generalized linear models to classify patients with BMI≥25, BMI≥30 and BMI≥40. We developed the prediction model using EHR system-1 (training set) and validated it in EHR system-2 (validation set). The cohort contained 123,432 patients in the Medicare and 40,736 patients in the Medicaid population, respetively. The model comprised 97 variables in Medicare and 95 in Medicaid, including BMI-related diagnosis codes, cardiovascular and antidiabetic drugs, and obesity-related comorbidities. The area under the receiver-operating-characteristic curve (AUC) in the validation set was 0.72, 0.76, 0.86 (Medicare) and 0.66, 0.66, 0.70 (Medicaid) for BMI ≥25, BMI ≥30, BMI ≥40, respectively. The positive predictive value was 81.5%, 80.6%, 64.7% (Medicare) and 81.6%, 77.5%, 62.5% (Medicaid), for BMI≥25, BMI≥30, BMI≥40, respectively. The proposed model can identify obesity categories in claims databases when BMI measurements are missing and can be used for confounding adjustment, defining subgroups, or probabilistic bias analysis.
Type of Medium:
Online Resource
ISSN:
0002-9262
,
1476-6256
Language:
English
Publisher:
Oxford University Press (OUP)
Publication Date:
2023
detail.hit.zdb_id:
2030043-8
Bookmarklink