Optimizing screening for acute human immunodeficiency virus infection with pooled nucleic acid amplification tests

Area of Science:

• Biostatistics
• Public Health
• Infectious Disease Epidemiology

Background:

• Identifying acute human immunodeficiency virus infection (AHI) is vital for public health.
• Nucleic acid amplification testing (NAAT) is effective but costly for widespread individual screening.
• Optimizing pooled NAAT (group testing) strategies is challenging but necessary for efficiency.

Purpose of the Study:

• To develop and compare simple biostatistical models for specimen pooling with NAAT for AHI detection.
• To evaluate the sensitivity, efficiency, and positive predictive value (PPV) of different pooling algorithms.
• To provide practical tools, like reference tables and a web calculator, to aid investigators in selecting pooling strategies.

Main Methods:

• Developed flexible biostatistical models integrating group testing theory, assay characteristics, and viral dynamics.
• Modeled and compared three pooling algorithms: two-stage minipools (D2), three-stage hierarchical pools (D3), and square arrays with master pools (A2m).
• Validated model results using stochastic simulations.

Main Results:

• All three pooling strategies (D2, D3, A2m) improved efficiency and PPV compared to individual NAAT for AHI case detection.
• D3 and A2m algorithms generally outperformed D2 in efficiency and PPV.
• The A2m algorithm typically demonstrated superior PPV compared to D3.

Conclusions:

• Simple pooling models can effectively guide the selection of appropriate algorithms for AHI detection.
• The developed models and tools facilitate the implementation of efficient pooled NAAT strategies in diverse settings.
• Strategic use of these models can enhance AHI surveillance and control efforts.