Development of a risk estimation model for condomless sex among college students in Zhuhai, China: a cross-sectional study

Abstract

Background: Condom use at last intercourse is an effective indicator for human immunodeficiency virus (HIV) prevention. To identify at-risk individuals and improve prevention strategies, this study explored factors associated with condomless sex at last intercourse in the last year and developed a risk estimation model to calculate the individual possibility of condomless sex among college students in Zhuhai, China.

Methods: A cross-sectional study was conducted among 1430 college students who had sex in the last year from six universities in Zhuhai. The least absolute shrinkage and selection operator (LASSO) and logistic regression were performed to explore the predictors of condomless sex. The nomogram was constructed to calculate the individual possibility of condomless sex. Discrimination and calibration of the nomogram were evaluated using the area under the receiver-operator characteristic curve (AUROC) and the calibration curve.

Results: The proportion of students who had condomless sex at last intercourse was 18.2% (260/1430). Students who had experienced more types of intimate partner violence (aOR, 1.58; 95% CI, 1.31 ~ 1.92) and had anal sex (aOR, 1.75; 95% CI, 1.06 ~ 2.84) were more likely to have condomless sex. Students who had heterosexual intercourse (aOR, 0.37; 95% CI, 0.21 ~ 0.70), used condoms at first sex (aOR, 0.20; 95% CI, 0.14 ~ 0.27), had high attitudes towards condom use (aOR, 0.87; 95% CI, 0.80 ~ 0.95) and self-efficacy for condom use (aOR, 0.84; 95% CI, 0.78 ~ 0.90) were less likely to have condomless sex. The nomogram had high accuracy with an AUROC of 0.83 and good discrimination.

Conclusions: Intimate partner violence, anal sex, condom use at first sex, attitude towards condom use, and self-efficacy for condom use were associated with condomless sex among college students. The nomogram was an effective and convenient tool for calculating the individualized possibility of condomless sex among college students. It could help to identify individuals at risk and help universities and colleges to formulate appropriate individualized interventions and sexual health education programs.

Keywords: College students; Condom use; Condomless sex; LASSO; Nomogram.

Fig. 1

Nomogram for Predicting the Possibility of Condomless Sex and Its Predictive Performance. Nomogram for Predicting the Possibility of Condomless Sex in College Students Who Had Sex in the Last Year in Zhuhai, China. In the nomogram above, the options for each variable correspond to a particular point on the top row. The points for each variable are added together to give a total score, which refers to the probability on the bottom row. *IPV. Imitate partner violence

Fig. 2

The area under the receiver-operator characteristic curves (AUROC) and the calibration curve of the nomogram for predicting condomless sex. A The AUROC of Derivation cohort. B The calibration curve of Derivation cohort. C The AUROC of Validation cohort. D The calibration curve of Validation cohort