Lessons learnt from assessing and improving accuracy and positive predictive value of the national HIV testing...

Area of Science:

• Public Health
• Infectious Disease Epidemiology
• Diagnostic Accuracy Studies

Background:

• Nigeria's national HIV rapid testing algorithm, implemented in 2007, uses a sequence of three HIV rapid test kits.
• A 2018 survey revealed significant discordance between the first and second HIV tests, prompting an evaluation of the algorithm's performance.
• This study addresses the need to evaluate the Nigerian national HIV rapid testing algorithm.

Purpose of the Study:

• To evaluate the performance of Nigeria's national HIV rapid testing algorithm.
• To assess the algorithm's accuracy in field and laboratory settings.
• To provide context-specific considerations for evaluating HIV testing algorithms.

Main Methods:

• A two-phased evaluation using retrospective data (2017-2019) from 24 sites and prospective data from 2895 adults (2020).
• Prospective evaluation conducted in field settings and under controlled conditions at the National Reference Laboratory across Nigeria's six geopolitical zones.
• Quality assurance measures included stakeholder engagement, personnel training, and supportive supervision.

Main Results:

• The HIV testing algorithm demonstrated higher sensitivity and specificity when evaluated in the National Reference Laboratory compared to field assessments.
• Quality assurance approaches were crucial for achieving high-quality study outcomes.
• Significant discordance between initial HIV tests highlighted the need for algorithm evaluation.

Conclusions:

• The evaluation provided valuable context-specific insights for applying World Health Organization recommendations to Nigeria's national HIV rapid testing algorithm.
• Quality assurance is integral to the reliable evaluation of HIV testing algorithms.
• Comparison of testing algorithms against a gold standard is recommended for future evaluations.