Population health and sociodemographic variables as predictors of access to cardiac medicine and surgery in Haiti

Abstract

Abstract Background In Haiti, cardiovascular disease is a leading cause of morbidity and mortality, with congenital and rheumatic heart disease comprising a large portion of disease burden. However, domestic disparities in cardiac care access and their impact on clinical outcomes remain poorly understood. We analyzed population-level sociodemographic variables to predict cardiac care outcomes across the 10 Haitian administrative departments. Methods This cross-sectional study combined data from a 2016–17 Haitian national survey with aggregate outcomes from the Haiti Cardiac Alliance (HCA) database (n = 1817 patients). Using univariate and multivariable regression analyses, the proportion of HCA patients belonging to each of three clinical categories (active treatment, lost to follow-up, deceased preoperatively) was modeled in relation to six population-level variables selected from national survey data at the level of the administrative department. Results In univariate analysis, higher department rates of childhood growth retardation were associated with a lower proportion of patients in active care (OR = 0.979 [0.969, 0.989], p = 0.002) and a higher proportion of patients lost to follow-up (OR = 1.016 [1.006, 1.026], p = 0.009). In multivariable analysis, the proportion of department patients in active care was inversely associated with qualified prenatal care (OR = 0.980 [0.971, 0.989], p = 0.005), and child growth retardation (OR = 0.977 [0.972, 0.983]), p = 0.00019). Similar multivariable results were obtained for department rates of loss to follow-up (child growth retardation: OR = 1.018 [1.011, 1.025], p = 0.002; time to nearest healthcare facility in an emergency: OR = 1.004 [1.000, 1.008, p = 0.065) and for preoperative mortality (prenatal care: OR = 0.989 [0.981, 0.997], p = 0.037; economic index: OR = 0.996 [0.995, 0.998], p = 0.007; time to nearest healthcare facility in an emergency: OR = 0.992 [0.988, 0.996], p = 0.0046). Conclusions Population-level survey data on multiple variables predicted domestic disparities in HCA clinical outcomes by region. These findings may help to identify underserved areas in Haiti, where increased cardiac care resources are required to improve health equity. This approach to analyzing clinical outcomes through the lens of population-level survey data may inform future health policies and interventions designed to increase cardiac care access in Haiti and other low-income countries.