HIV Patient Retention Predictors from Haiti's National EMR, 2018-2024

Jiaqin Wu's study analyzes data from Haiti’s national electronic medical record database, containing 115,822 people living with HIV across 167 health facilities from 2018 to 2024. The research developed machine learning models to forecast individual risk of loss to follow-up using demographic, clinical, and institutional predictors. The dataset was last updated on 2026-04-28.

Use Cases

• Training machine learning models to predict patient attrition based on demographic and clinical history.
• Identifying key institutional and care pattern predictors of loss to follow-up for targeted interventions.
• Analyzing the temporal performance of predictive models across different forecasting quarters.
• Interpreting model decisions using SHAP values to understand feature influence on risk scores.

Strengths

• Contains 115,822 patient records, providing a substantial cohort for analysis.
• Covers a 6-year time range (2018-2024) and 167 health facilities, offering broad temporal and geographic scope.
• Includes multiple predictor categories mentioned in the description: demographic, clinical, and institutional.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count for the underlying dataset is unknown, which may limit suitability assessment.
• The data file is a 15.2 KB DOCX document, suggesting the dataset itself may not be directly included or is extremely limited in scope.

Provenance

Source

Haiti’s national electronic medical record database.

Collection Method

Routine clinical data collection from health facilities.

Time Range

2018 to 2024

Freshness

Last updated 2026-04-28 05:01:17; freshness should be verified.

Geography

Haiti