A Risk-Score for Predicting the Presence of Treatable Sexually Transmitted Infections in Kenyan Women Planning Conception

Abstract

Treatable sexually transmitted infections (STI) disproportionately affect women’s reproductive health and contribute to poor neonatal outcomes. Because these infections are frequently asymptomatic, identifying women at higher risk of having STIs using risk scores may provide a cost-effective screening approach in regions where universal screening of pregnancy planners is not performed. The aim of this study was to determine the prevalence and correlates of Chlamydia trachomatis, Neisseria gonorrhoeae, and Trichomonas vaginalis among Kenyan pregnancy planners, and to develop a risk score to identify those with a higher likelihood of current chlamydial infection. Kenyan women with fertility intent enrolled into a cohort study between April 2017 and March 2020 contributed data for this cross-sectional analysis. Logistic regression was used to estimate odds ratios of the association between demographic, behavioral, and clinical risk factors and prevalent STIs, both as a group, and for C. trachomatis alone. Based on the regression coefficients, prediction models were developed to identify women with increased likelihood of current C. trachomatis infection. The most common STI was C. trachomatis (51/691, 7.4%); N. gonorrhoeae (5/691, 0.5%), and T. vaginalis infections (6/687, 0.9%) were rare. The prevalence of any one or more of these STI was 60/688 (8.7%). Risk factors for any STI included age less than 25 (OR 2.41; 95%Cl: 0.69-8.48), partner’s age less than 25 (OR 17.22; 95%Cl: 3.74-79.24), Alcohol Use Disorders Identification Test (AUDIT) score ≥8 (OR 5.54; 95%Cl: 1.61-19.04), and bacterial vaginosis (BV) (OR 2.49; 95%Cl: 1.46-4.24). A risk score for predicting C.trachomatis infection, ranging from 0-6, derived from the participant’s age, AUDIT score, and BV status yielded an area under receiver operating curve (AUROC) of 0.78 (95%Cl: 0.72-0.84). Using a score cutoff of 0 versus ≥1, 478/691 (69.2%) were classified as higher-risk for C. trachomatis (sensitivity=98.0%, 95%Cl: 89.6-100.0; specificity=33.1%, 95%Cl: 29.5-36.9). At a higher cutoff of ≤2 versus ≥3, the risk score identified 31.8% of women as higher risk (sensitivity=70.6%, 95%Cl: 56.2-71.3, specificity=71.3%, 95%Cl: 67.7-74.5). Among women classified as higher risk by the risk scoring tool, the numbers needed to screen using nucleic acid amplification based tests (NAAT) testing to detect one C. trachomatis infection were 10.0 for the 0 versus ≥1 cutoff and 7.7 for the higher cutoff of ≤2 versus ≥3. This risk-scoring tool may be useful for identifying higher risk women for C. trachomatis screening among Kenyan pregnancy planners. Employing the tool could provide a cost-conscious approach for initiating species-specific testing for C. trachomatis infection.