Computational Validation of a Clinical Decision Support Algorithm for LAI-PrEP Bridge Period Navigation at UNAIDS...

Area of Science:

• Public Health
• Infectious Disease Prevention
• Health Informatics

Background:

• Long-acting injectable HIV pre-exposure prophylaxis (LAI-PrEP) offers superior efficacy but faces significant implementation barriers, with 47% of patients missing their initial injection.
• Disparities in LAI-PrEP uptake exist among different populations (e.g., People Who Inject Drugs vs. Men Who Have Sex With Men) and regions.

Purpose of the Study:

• To develop and validate a clinical decision support tool to enhance LAI-PrEP initiation during the critical "bridge period."
• To stratify patient risk, identify barriers, and recommend evidence-based interventions for successful LAI-PrEP transitions.

Main Methods:

• Developed a clinical decision support tool integrating data from major LAI-PrEP trials and implementation studies.
• Employed an external configuration architecture and a library of 21 diverse interventions.
• Conducted progressive validation on scales ranging from 1,000 to 21.2 million patients, achieving 100% test pass rate for edge cases.

Main Results:

• The tool demonstrated increasing precision with scale, reaching ±0.018 pp at 21.2 million patients.
• Evidence-based interventions improved LAI-PrEP transition success from a baseline of 23.96% to 43.50% at the UNAIDS 2025 target scale.
• Interventions disproportionately benefited vulnerable groups, showing significant relative improvements for People Who Inject Drugs (+265%) and adolescents (+147%).

Conclusions:

• The validated tool shows policy-grade statistical precision in predicting and improving LAI-PrEP uptake.
• Systematic implementation support can effectively narrow health equity gaps.
• Successful LAI-PrEP transitions are projected to prevent 80,000-100,000 HIV infections annually, with an estimated USD 40 billion in averted treatment costs.