AUTHOR=Tesfie Tigabu Kidie , Yehuala Tirualem Zeleke , Agimas Muluken Chanie , Yismaw Getaneh Awoke , Wubante Sisay Maru , Fente Bezawit Melak , Derseh Nebiyu Mekonnen 

TITLE=Predicting the individualized risk of human immunodeficiency virus infection among sexually active women in Ethiopia using a nomogram: prediction model development and validation

JOURNAL=Frontiers in Public Health

VOLUME=Volume 12 - 2024

YEAR=2024

URL=https://www.frontiersin.org/journals/public-health/articles/10.3389/fpubh.2024.1375270

DOI=10.3389/fpubh.2024.1375270

ISSN=2296-2565

ABSTRACT=Introduction: Women are more vulnerable population groups to HIV infection due to biological and socioeconomic reasons. Developing prediction model for these vulnerable population to estimate individualized risk for HIV infection is relevant for targeted preventive interventions. The objective of the study was to develop and validate a risk prediction model allowing easy estimations of HIV infection risk for sexually active women in Ethiopia.
Methods: Ethiopian Demographic and Health Survey 2016 data, which comprised 10253 representative sexually active women was used for model development. Variables were selected using Least Absolute Shrinkage and Selection Operator (LASSO).  Variables selected by LASSO were incorporated in the multivariable mixed-effect logistic regression model. Based on the multivariable model, easy to use nomogram was developed to facilitate its applicability. Performance of the nomogram was evaluated using discrimination and calibration abilities, brier score, sensitivity and specificity.  Internal validation was carried out using bootstrapping method. 
Results: The model selected 7 predictors of HIV infection encompassing age, education, marital status, sex of household head, age at first sex, lifetime multiple sexual partners and residence. The nomogram had discriminatory power of 89.7% (95% CI: 88.0, 91.5) and calibration p-value of 0.536. In addition, sensitivity and specificity of the nomogram was 74.1% (95% CI: 68.4, 79.2), 80.9% (95% CI: 80.2, 81.7), respectively. The internally validated model had discriminatory ability of 89.4% (95% CI: 87.7, 91.1) and calibration p-value of 0.195. Sensitivity and specificity after validation was 72.9% (95% CI: 67.2, 78.2) and 80.1% (95% CI: 79.3, 80.9), respectively. 
Conclusion: A new prediction model that quantifies individualized risk of HIV infection has been developed in the form of nomogram and internally validated. It has very good discriminatory power and well-calibration ability. This model can facilitate identification of sexually active women at high-risk of HIV infection for targeted preventive measures.