Skip to main content
Journal cover image

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

Publication ,  Journal Article
Burns, CM; Pung, L; Witt, D; Gao, M; Sendak, M; Balu, S; Krakower, D; Marcus, JL; Okeke, NL; Clement, ME
Published in: Clin Infect Dis
January 13, 2023

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.

Duke Scholars

Altmetric Attention Stats
Dimensions Citation Stats

Published In

Clin Infect Dis

DOI

EISSN

1537-6591

Publication Date

January 13, 2023

Volume

76

Issue

2

Start / End Page

299 / 306

Location

United States

Related Subject Headings

  • United States
  • Pre-Exposure Prophylaxis
  • Microbiology
  • Male
  • Machine Learning
  • Humans
  • HIV Infections
  • HIV
  • Female
  • Electronic Health Records
 

Citation

APA
Chicago
ICMJE
MLA
NLM
Burns, C. M., Pung, L., Witt, D., Gao, M., Sendak, M., Balu, S., … Clement, M. E. (2023). Development of a Human Immunodeficiency Virus Risk Prediction Model Using Electronic Health Record Data From an Academic Health System in the Southern United States. Clin Infect Dis, 76(2), 299–306. https://doi.org/10.1093/cid/ciac775
Burns, Charles M., Leland Pung, Daniel Witt, Michael Gao, Mark Sendak, Suresh Balu, Douglas Krakower, Julia L. Marcus, Nwora Lance Okeke, and Meredith E. Clement. “Development of a Human Immunodeficiency Virus Risk Prediction Model Using Electronic Health Record Data From an Academic Health System in the Southern United States.Clin Infect Dis 76, no. 2 (January 13, 2023): 299–306. https://doi.org/10.1093/cid/ciac775.
Burns, Charles M., et al. “Development of a Human Immunodeficiency Virus Risk Prediction Model Using Electronic Health Record Data From an Academic Health System in the Southern United States.Clin Infect Dis, vol. 76, no. 2, Jan. 2023, pp. 299–306. Pubmed, doi:10.1093/cid/ciac775.
Burns CM, Pung L, Witt D, Gao M, Sendak M, Balu S, Krakower D, Marcus JL, Okeke NL, Clement ME. Development of a Human Immunodeficiency Virus Risk Prediction Model Using Electronic Health Record Data From an Academic Health System in the Southern United States. Clin Infect Dis. 2023 Jan 13;76(2):299–306.
Journal cover image

Published In

Clin Infect Dis

DOI

EISSN

1537-6591

Publication Date

January 13, 2023

Volume

76

Issue

2

Start / End Page

299 / 306

Location

United States

Related Subject Headings

  • United States
  • Pre-Exposure Prophylaxis
  • Microbiology
  • Male
  • Machine Learning
  • Humans
  • HIV Infections
  • HIV
  • Female
  • Electronic Health Records