Social determinants of health and interruption in treatment outcomes within Haiti’s electronic medical record system for people living with HIV

Abstract

Haiti has made significant progress in reducing the national burden of HIV, however patient retention in care and treatment persists as a key challenge. Low-resource settings are increasingly using electronic medical records (EMRs) to strengthen their health information systems and improve patient outcomes. Social determinants of health (SDOH) like employment, education, food insecurity, and stigma are documented risk factors for poor HIV treatment outcomes but are often not captured or collected in structured formats in EMRs. We conducted an exploratory study assessing the types and completeness of SDOH data in Haiti’s EMR and investigated associations with interruption in treatment (IIT) at 6- and 12-months post-antiretroviral therapy (ART) initiation. We used patient-level data from 51 clinic sites. Eligible patients were adults with a documented ART start date between July 2016 and January 2022. SDOH variables were categorized by the following six domains: 1) economic stability, 2) education, 3) food, 4) social and community context, 5) physical and built environment, and 6) health and clinical care. A mixed effects logistic regression model was fit to assess the relationship between independent SDOH variables and IIT. We found a lack of variables related to physical and built environment, food, education, and economic stability in Haiti’s EMR. Data completion varied by variable type for several SDOH variables available for analysis. Approximately one in ten patients had IIT at 6 months and nearly one in five had IIT at 12 months. Documentation of the HIV status of other household members, a possible marker for social context, was associated with a lower likelihood of IIT. There were several SDOH reasons associated with higher risk of IIT, including travel and lack of food. This analysis highlights a missed opportunity to collect individual-level SDOH data within Haiti’s EMR and identify clients at greatest risk of IIT who could benefit from interventions to provide enhanced support for ART retention in care.