ABSTRACT
Introduction: There are several risk factors being used to identify undiagnosed HIV-infected adults. As the number of undiagnosed people gets less and less, it is important to know if existing risk factors and risk assessment tools are valid for use. Methods: Data from the Tanzania and Zambia Population-Based HIV Impact Assessment (PHIA) household surveys which were conducted during 2016 was used. We first included 12 risk factors (being divorced, separated or widowed; having an HIV+ spouse; having one of the following within 12-months of the survey: paid work, slept away from home for ≥1-month, having multiple sexual partners, clients of sex workers, sexually transmitted infection, being tuberculosis suspect, being very sick for ≥3-months; ever sold sex; diagnosed with cervical cancer; and had TB disease into a risk assessment tool and assessed its validity by comparing it against HIV test result. Sensitivity, specificity and predictive value of the tool were assessed. Receiver Operating Characteristic (ROC) curve comparison statistics was also used to determine which risk assessment tool was better. Results: HIV prevalence was 2.3% (2.0%-2.6%) (n=14,820). For the tool containing all risk factors, HIV prevalence was 1.0% when none of the risk factors were present (Score 0) compared to 3.2% when at least one factor (Score ≥1) was present and 8.0% when ≥4 risk factors were present. Sensitivity, specificity, PPV, and NPV were 82.3% (78.6%-85.9%), 41.9%(41.1%-42.7%), 3.2%(2.8%-3.6%), and 99.0%(98.8%-99.3%), respectively. The use of a tool containing conventional risk factors (all except those related with working and sleeping away) was found to have higher AUC (0.65 vs 0.61) compared to the use of all risk factors (p value <0.001). Conclusions: The use of a screening tool containing conventional risk factors improved HIV testing yield compared to doing universal testing. Prioritizing people who fulfill multiple risk factors should be explored further to improve HIV testing yield.