Simple screening tool to help identify high-risk children for targeted HIV testing in malawian inpatient wards

Background: To meet global AIDS goals, pediatric HIV diagnosis must be strengthened. Provider-initiated testing and counseling, which is recommended by the WHO, faces persistent implementation challenges in low-resource settings. Alternative approaches are needed. Setting: Malawi has achieved high coverage of HIV diagnosis and treatment, but there are gaps among pediatric populations. This study assessed the sensitivity and specificity of a brief screening tool to identify at-risk pediatric patients for targeted HIV testing in Malawi. Methods: A tool containing 6 yes/no items was used for children (aged 1-15 years) in the inpatient pediatric wards at 12 hospitals in Malawi (July 2016-July 2017). Questions were based on an established tool, translated to Chichewa, and implemented by HIV diagnostic assistants. All participating children were provided HIV testing and counseling per Ministry of Health guidelines. Analysis estimated the tool's characteristics including sensitivity, specificity, negative, and positive predictive values. Results: HIV prevalence among the 8602 participants was 1.1% (n = 90). Children with a screening tool score of 1 had double the odds of being HIV positive than those with a score of 0. Frequent sickness was the most sensitive predictor of HIV status (55.1%), and having a deceased parent was the most specific (96.7%). False classification of HIV-negative status was rare (n = 14) but occurred more often among boys and younger children. Conclusions: A brief screening tool for pediatric inpatients helped target HIV testing in those most at risk in a low-pediatric-prevalence, resource-constrained setting. Future research should include a direct, rigorous comparison with PITC including comparative effectiveness, efficiency, and cost effectiveness.