Development of a Human Immunodeficiency Virus Risk Prediction Model Using Electronic Health Record Data From an Academic Health System in the Southern United States. (19th September 2022)

• Record Type: Journal Article
• Authors: Burns, Charles M; Pung, Leland; Witt, Daniel; Gao, Michael; Sendak, Mark; Balu, Suresh; Krakower, Douglas; Marcus, Julia L; Okeke, Nwora Lance; Clement, Meredith E
• Is Part Of: Clinical infectious diseases. Volume 76:Number 2(2023)
• Page Start: 299
• Page End: 306
• Publication Date: 2022-09-19
• Subjects: PrEP -- southern United States -- prediction model -- women -- machine learning; Communicable diseases -- Periodicals; 616.905
• DOI: 10.1093/cid/ciac775
• Languages: English
• ISSNs: 1058-4838

Abstract

Abstract: Background: Human immunodeficiency virus (HIV) pre-exposure prophylaxis (PrEP) is underutilized in the southern United States. Rapid identification of individuals vulnerable to diagnosis of HIV using electronic health record (EHR)-based tools may augment PrEP uptake in the region. Methods: Using machine learning, we developed EHR-based models to predict incident HIV diagnosis as a surrogate for PrEP candidacy. We included patients from a southern medical system with encounters between October 2014 and August 2016, training the model to predict incident HIV diagnosis between September 2016 and August 2018. We obtained 74 EHR variables as potential predictors. We compared Extreme Gradient Boosting (XGBoost) versus least absolute shrinkage selection operator (LASSO) logistic regression models, and assessed performance, overall and among women, using area under the receiver operating characteristic curve (AUROC) and area under precision recall curve (AUPRC). Results: Of 998 787 eligible patients, 162 had an incident HIV diagnosis, of whom 49 were women. The XGBoost model outperformed the LASSO model for the total cohort, achieving an AUROC of 0.89 and AUPRC of 0.01. The female-only cohort XGBoost model resulted in an AUROC of 0.78 and AUPRC of 0.00025. The most predictive variables for the overall cohort were race, sex, and male partner. The strongest positive predictors for the female-only cohort were history of pelvic inflammatory disease, drug use, and tobacco use. Conclusions: Our machine-learning models were able to effectively predict incident HIV diagnoses including among women. This study establishes feasibility of using these models to identify persons most suitable for PrEP in the South. Abstract : We created machine-learning models to predict incident human immunodeficiency virus (HIV) diagnosis as a marker for pre-exposure prophylaxis (PrEP) candidacy among patients from a southern medical system. The models effectively predicted incident HIV diagnoses and were more effective among women compared to prior models.