Validation of an Optimized Algorithm to Identify Persons Living with Diagnosed HIV from the New York State Medicaid Data, 2006-2014

Abstract

Algorithms are regularly used to identify persons living with diagnosed HIV (PLWDH) in the Medicaid data. To the authors’ knowledge, there are no published reports of an HIV algorithm from claims codes that have been compared to an HIV surveillance system to assess its sensitivity, specificity, positive predictive value and negative predictive value in identifying PLWDH. The aims of this study were to: 1) develop an algorithm that could identify PLWDH in New York Medicaid data from 2006-2014; and 2) validate this algorithm using the New York HIV surveillance system. Classification and regression tree analysis identified 16 nodes that were combined to create a case-finding algorithm with five criteria. This algorithm identified 86,930 presumed PLWDH, of which 88.0% were verified by matching to the surveillance system. The algorithm yielded a sensitivity of 94.5%, a specificity of 94.5%, a positive predictive value of 88.0%, and a negative predictive value of 97.6%. This validated algorithm has the potential to improve the utility of the Medicaid data for assessing health outcomes and programmatic interventions.

American Journal of Epidemiology. 2019 Oct. doi: 10.1093/aje/kwz225.