Developing and validating a clinical risk score to predict losses in the PISCIS cohort of people with HIV

Area of Science:

• Public Health
• Epidemiology
• HIV Medicine

Background:

• Loss to follow-up (LTFU) from HIV care increases health risks for people living with HIV (PLWH).
• Predicting LTFU is crucial for maintaining patient health and optimizing HIV care management.
• Developing targeted interventions requires accurate identification of at-risk individuals.

Purpose of the Study:

• To develop and validate a predictive risk score for LTFU among PLWH in Catalonia and the Balearic Islands.
• To identify key factors associated with LTFU in this population.
• To provide a tool for early intervention and improved patient retention in HIV care.

Main Methods:

• A cohort of 6661 people living with HIV (PLWH) was analyzed.
• Logistic regression models identified independent predictors of LTFU (defined as no contact with HIV care for ≥12 months).
• A 10-fold cross-validation was used to assess model performance and calculate the Area Under the ROC Curve (AUC).

Main Results:

• Key predictors of LTFU included younger age (<34 years), not being born in Spain, being a man who injects drugs, detectable viral load, and diagnosis within the last 2.5 years.
• The risk score's validation yielded a mean AUC of 0.69.
• The score identified 28.8% of PLWH at medium risk and 3.4% at high risk of LTFU.

Conclusions:

• The developed risk score can effectively predict LTFU in people living with HIV.
• These findings support the implementation of targeted strategies to prevent LTFU and improve engagement in HIV care.
• Proactive identification of at-risk individuals is essential for better health outcomes in PLWH.