Multi-jurisdiction linkage of PRAMS and hospital discharge data: Methods, key challenges, and practical applications

Linking Pregnancy Risk Assessment Monitoring System (PRAMS) data with hospital discharge data (HDD) offers opportunities to explore self-reported factors on PRAMS with maternal health outcomes like severe maternal morbidity (SMM) and hypertensive disorders in pregnancy (HDP). This study outlines the methods and challenges with creating a multi-jurisdiction PRAMS-HDD record level linked dataset using existing data linkages for assessing factors associated with SMM and HDP.

Four jurisdictions participating in a PRAMS data linkage learning community linked PRAMS and HDD deliveries from 2017 to 2020 using various linkage strategies. Both SMM and HDP were identified in the HDD data using a standardized algorithm provided by the researcher. Record level PRAMS data with SMM and HDP indicators were then combined across states. Potential impact of linkage rates and quality, prevalence of the outcome, sampling strata, and response rates were assessed.

Among the four states, 17,878 PRAMS respondents were linked to a hospital delivery recorded in the HDD. Linkage rates varied from 63 % to 99 % (86 % overall). 227 SMM (1.3 %) and 2889 HDP (17.2 %) cases were identified and varied significantly by state. Linkage rates influenced reliability of outcome classification.

Multi-jurisdiction PRAMS linkage offers potential for evaluating maternal health outcomes. However, variations in data quality, linkage rates and sampling strategy impact result reliability. Future multi-jurisdiction PRAMS linkage studies should prioritize standardizing data structures, protocols, and linkage methodology to maximize results quality and generalizability.