Validated algorithms can help identify pulmonary arterial hypertension in administrative data

In this study, researchers developed and validated algorithms to help identify pulmonary arterial hypertension (PAH) in administrative data in Veterans Health Administration (VA) hospitals and a PAH referral center at Boston Medical Center (BMC).

Using International Classification of Diseases diagnosis codes, all patients in each setting diagnosed with pulmonary hypertension (PH) were identified and analyzed for diagnosis codes for PAH-associated conditions, procedure codes for right heart catheterizations (RHC), and pharmacy claims for PAH-specific therapy.

Algorithms were validated using gold standard review of primary clinical data and calculated sensitivity, specificity, positive predictive values (PPV), and negative predictive values.

A total of 12,012 patients with PH were identified in the VA setting and 503 patients in the BMC setting. When only PH diagnosis codes were used to identify PAH, the PPV was only 16% in the VA setting and 36% in the BMC setting. Only modest improvements were noted with the addition of PAH-associated conditions.

An algorithm combining ICD diagnosis codes, RHC codes, and PAH-specific therapy resulted in 97.1% ad 95% specificity and 70% and 86% PPV in the VA setting and the BMC settings, respectively.

Reference
Gillmeyer KR, Nunez ER, Rinne ST, et al. Development and validation of algorithms to identify pulmonary arterial hypertension in administrative data. Chest. 2020;S0012-3692(20)35455-6. DOI: 10.1016/j.chest.2020.12.010. Epub ahead of print. PMID: 33345949.