A data-driven approach to implementing the HPTN 094 complex intervention INTEGRA in local communities.

Area of Science:

• Implementation Science
• Public Health
• Health Services Research

Background:

• People who inject drugs (PWID) face significant HIV burden due to syndemic factors like stigma, poverty, and limited access to medication for opioid use disorder (MOUD) and HIV treatment/prevention.
• Existing health services are often not co-located, hindering comprehensive care for PWID.
• Shifting drug use patterns and geographic variations in the US necessitate adaptive service delivery models.

Purpose of the Study:

• To evaluate a data-driven approach for optimizing the delivery of integrated MOUD and HIV services to PWID.
• To identify and characterize viable implementation neighborhoods for a mobile integrated care unit.

Main Methods:

• A hybrid, mixed-method, data-driven approach using the PRISM framework was employed.
• Triangulated pre-implementation (epidemiological surveillance) and implementation phase data (ecological assessments, meetings).
• Characterized environmental factors influencing neighborhood viability across five US cities.

Main Results:

• Neighborhood-level drug use, geographic dispersion, and socio-political factors influenced the effectiveness of epidemiological data in identifying suitable sites.
• Proximity to public transportation and PWID gathering spaces was crucial in dispersed populations.
• The utility of additional data sources (overdose, housing) for site selection was variable.

Conclusions:

• Data-driven strategies offer a pragmatic approach for real-time implementation of integrated care models for PWID.
• This methodology aids in scaling up complex interventions by considering local community environmental factors.
• Implementation science methods are vital for addressing the diverse needs of PWID across varied US landscapes.