Use of electronic health record data and machine learning to identify candidates for HIV preexposure prophylaxis: A modeling study
The Lancet HIV Oct 06, 2019
Marcus JL, et al. - Researchers sought to develop and validate an HIV prediction model that may aid in recognizing potential preexposure prophylaxis (PrEP) candidates in a large healthcare system. From a large integrated healthcare system, Kaiser Permanente Northern California, HIV-uninfected adult members who were not yet using PrEP and had at least 2 years of previous health plan enrolment with at least one outpatient visit from Jan 1, 2007, to Dec 31, 2017, were included in the study. Employing 81 electronic health record (EHR) variables, they applied the least absolute shrinkage and selection operator regression and retained 44 predictors in the full model. The study sample comprised 3,750,664 patients; 3,143,963 were included in the development dataset and 606,701 in the validation dataset. Among these, 784 patients developed HIV within 3 years of baseline. The model had an AUC of 0·86 (95% CI 0·85–0·87) for incident HIV cases in 2007–14. Model performance remained high in the validation dataset, (AUC 0·84, 0·80–0·89). Findings suggest that the prediction models using EHR data may assist in distinguishing patients at high risk of HIV acquisition who could benefit from PrEP.
Go to Original
Only Doctors with an M3 India account can read this article. Sign up for free or login with your existing account.
4 reasons why Doctors love M3 India
-
Exclusive Write-ups & Webinars by KOLs
-
Daily Quiz by specialty
-
Paid Market Research Surveys
-
Case discussions, News & Journals' summaries