Methods: The linkage methodology consists of: 1) the development of self- correcting, patient-level custom linkage profiles across databases, 2) a deterministic (rule-based) record linkage using exact and fuzzy text matching techniques, 3) a probabilistic linkage using data mining algorithms, and 4) a clerical-review record linkage. Using a dataset extracted from the linked repository, we compared five maternal and infant health status indicators in three race/ethnicity groups—Hispanic, African American and Caucasian—among women whose pregnancies were funded by Medicaid. We examined rates of inadequate prenatal care, pre-pregnancy weight, neonatal intensive care, low birth weight, and infant mortality in Medicaid deliveries in Florida. Population estimates for the 2016 calendar year will be available in June 2018.
Results: Results suggest that among Medicaid covered women in 2015, Hispanics showed a lower percent of inadequate prenatal care (14.6%), compared to African American (18.0%) and Caucasian (15.8%) women. A higher percent of Hispanic women in Medicaid, however, were overweight pre-pregnancy (30.6 %) when compared with African American (26.6 %) and Caucasian (24.7 %) women. Nevertheless, a lower rate of Hispanic infants required neonatal intensive care (8.9%) compared to African American (15.2%) and Caucasian (9.7%) infants. A lower rate of Hispanic (7.9 %) and Caucasian (8.3 %) showed low birth weight when compared with African American infants (14.2%). Finally, among Medicaid covered women in 2014, the Hispanic infant mortality rate per 1,000 live births was substantively lower than that of African Americans and Caucasians (4.9 vs. 11.1 and 6.7, respectively).
Conclusions: In conclusion, we present a method for linking and cataloging data across multiple, disparate data sources to investigate service delivery, utilization patterns, and care quality. The linkage methodology allows advancing knowledge on reduction of disparities in long-term outcomes and promoting health equity among individuals exposed to high-risk settings.