LAI-PrEP Implementation Science — Nyx Institute for Computational Medicine

The Bridge Period
Clinical Decision Tool

Two companion papers — one conceptual framework, one computational validation — built the scientific foundation for a clinical decision support tool that navigates the 2–8 week structural vulnerability between LAI-PrEP prescription and first injection, where 47.1% of patients are currently lost.

Scroll to explore
The Problem

The Bridge Period Paradox

Long-acting injectable PrEP (LAI-PrEP) offers over 96% efficacy in HIV prevention — superior to every oral formulation studied. Yet a structural failure in healthcare delivery erases that advantage before it begins.

47.1%

of prescribed patients never receive their first injection

Not because the medication fails. Because the healthcare system does. The 2–8 week "bridge period" between prescription and first injection — required for mandatory HIV testing, insurance authorization, and appointment coordination — creates a structural vulnerability unique to LAI-PrEP. The pharmacological constraint cannot be eliminated for new starts; it can only be systematically managed. That is what this tool does.

Paradoxically, patients who survive the bridge period show 81–83% persistence — far exceeding oral PrEP's 52%. The bottleneck is not retention. It is initiation.

Scientific Foundation

Two Papers. One Framework.

The tool is the product of a deliberate research architecture: one paper to define the problem and build the evidence library, one paper to prove the solution works at global scale.

Paper I — Conceptual Foundation
Accepted — In Press — Viruses 2026

Bridging the Gap: The PrEP Cascade Paradigm Shift for Long-Acting Injectable HIV Prevention

Defines the bridge period as a distinct, measurable cascade step. Synthesizes data from HPTN 083, HPTN 084, PURPOSE-1/2 (>15,000 participants) with real-world implementation cohorts. Characterizes 13 structural barriers across 7 population categories. Builds the 21-intervention evidence library with mechanism classification, effect sizes, and evidence tier ratings across 6 supplementary files. This is the intellectual architecture the tool required to function.

Demidont, A.C. Bridging the Gap: The PrEP Cascade Paradigm Shift for Long-Acting Injectable HIV Prevention. Viruses 2026. Manuscript ID: viruses-4192169. Accepted 5 March 2026.
Paper II — Computational Validation
Published — Viruses 2026, 18, 237

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

Validates the clinical decision support tool across 21.2 million synthetic patients — the UNAIDS 2025 global PrEP target — achieving policy-grade statistical precision (±0.018 pp). Progressive validation at 1K, 1M, 10M, and 21.2M scales. 18/18 edge cases passing. Demonstrates the tool's mathematical architecture — multiplicative barrier compounding, mechanism diversity scoring — functions correctly at any scale from a single clinic to global deployment.

Demidont, A.C. Computational Validation of a Clinical Decision Support Algorithm for LAI-PrEP Bridge Period Navigation at UNAIDS PrEP Target Scale. Viruses 2026, 18, 237. https://doi.org/10.3390/v18020237

LAI-PrEP Bridge Period Clinical Decision Support Tool

A Python-based, configuration-driven clinical decision support tool that synthesizes the evidence from both companion papers into patient-level guidance. The tool stratifies individual bridge period risk, identifies the specific structural barriers affecting each patient, and recommends evidence-based interventions prioritized by expected impact — with mechanism diversity scoring to prevent redundant recommendations.

21.2M
synthetic patients validated at UNAIDS scale
±0.018
pp precision — policy-grade statistical validity
21
evidence-based interventions with mechanism scoring
18/18
edge case tests passing — 100% test coverage
7
population categories with evidence-based baselines
13
structural barriers quantified with attrition impacts
Intervention Framework

Eliminate. Compress. Navigate.

The tool operationalizes a three-mechanism framework for bridge period management, drawing on the 21-intervention evidence library synthesized in Paper I and validated computationally in Paper II.

Eliminate

Oral-to-injectable same-day switching for the 3.8 million patients already on oral PrEP. Existing negative test results and established clinical engagement remove the bridge period entirely. The highest single-impact intervention identified — +35% absolute improvement. The tool prioritizes this pathway whenever patient history permits.

Compress

Dual HIV testing (antigen/antibody + RNA) shortens the mandatory window period from 45 days to 10–14 days. Point-of-care RNA platforms enable same-visit testing. The tool selects accelerated diagnostic pathways for patients where same-day switching is not feasible, compressing the structural vulnerability window to its minimum safe duration.

🧭

Navigate

Dedicated patient navigation, transportation support, telehealth integration, and harm reduction co-location for patients facing multiple co-occurring barriers. Patients with 3+ structural barriers face less than 15% baseline probability. The tool's mechanism diversity scoring ensures navigation interventions address distinct barrier types rather than duplicating effort.

Health Equity

The Equity Inversion

The tool's most important finding: populations with the lowest baseline bridge period success rates show the greatest relative improvement with targeted interventions. Systematic implementation narrows health equity gaps rather than widening them.

PWID
10.4%
↓ baseline  →  with interventions ↑
37.8%
+265% relative improvement
Adolescents (16–24)
16.3%
↓ baseline  →  with interventions ↑
40.3%
+147% relative improvement
Sub-Saharan Africa
21.7%
↓ baseline  →  with interventions ↑
39.2%
62% of global PrEP need
MSM (baseline)
33.1%
↓ baseline  →  with interventions ↑
48.5%
Highest current LAI-PrEP uptake

At UNAIDS 2025 target scale: 4.1 million additional successful transitions, 100,000 HIV infections prevented annually, $40 billion in averted lifetime treatment costs. 11:1 five-year return on investment.

Access & License

Pharma-Restricted Open Healthcare License v1.0

The tool is freely available to healthcare providers, researchers, and public health agencies. Pharmaceutical entities require written permission for commercial use.

Free Access

Healthcare providers & clinics
Researchers & academics
Non-profit organizations
Government & public health agencies (CDC, NIH, WHO, UNAIDS)
Safety-net & under-resourced providers

Pharma / Commercial

Patient care use: notify within 30 days
Commercial use: written permission required
Contact: acd@nyxcomputationalmedicine.org

Conflict of interest: Dr. Demidont was formerly employed by Gilead Sciences (ending October 2024); all holdings divested December 2024. This research received no industry funding. Disclosed per ICMJE guidelines in all publications.