HIV-ASSIST

HIV-ASSIST provides decision support for personalized antiretroviral (ARV) regimen selection by ranking candidate regimens using patient-specific and virus-specific factors to prioritize viral suppression, tolerability, and adherence.

Key Features:

• Personalized ARV Regimen Selection: Ranks antiretroviral regimens using a composite objective that prioritizes viral suppression while maximizing tolerability and adherence, incorporating comorbidities, treatment history, and virus-specific attributes such as HIV mutations.
• Multiple-Criteria Decision Analysis Framework: Implements a multiple-criteria decision analysis framework to construct mathematical algorithms that synthesize modifying factors including pill burden, side effects, barriers to resistance, and impact on comorbidities.
• Educational Resource: Provides educational content for clinicians about ART options and principles for tailoring regimens to individual patient needs.

Scientific Applications:

• Clinical Decision Support: Generates ranked lists of ARV options tailored to individual patient scenarios to inform clinician regimen selection.
• Research Validation: Validated against prescribing choices of experienced HIV providers at four academic centers, showing 99% concordance for ARV-naive patients and 84%–88% concordance for ARV-experienced patients.

Methodology:

Uses multiple-criteria decision analysis to construct mathematical algorithms that integrate patient-specific and virus-specific data (including HIV mutations) into regimen rankings; validation compared tool recommendations with provider selections across hypothetical patient-case scenarios.