Restriction of Abortion Data in Inpatient and Emergency Department Databases: Impacts on Early Pregnancy Research

The Healthcare Cost and Utilization Project (HCUP) State Inpatient Databases and State Emergency Department Databases are widely used to study differences in health outcomes across states with differing legal environments. In researching early pregnancy complications, maternal health outcomes, and abortion policy effects in particular, some partner states' restrictions on abortion-related records may prevent key analyses. We aimed to describe the scope of these restrictions and their research implications.

We reviewed publicly available 2022 Healthcare Cost and Utilization Project (HCUP) documentation and state partner notes for all State Inpatient Databases (SID) and State Emergency Department Databases (SEDD). We identified states with abortion-related limitations and categorized the affected data element and restriction type. To describe the practical impact of data restriction, we assessed purchased New York data for missing variables and the visits affected.

Three HCUP partner states restricted abortion-related data elements: New York, Washington, and Wisconsin. In New York SID and SEDD, selected fields are blanked when records indicate abortion; upon review of purchased data, this suppression extends to visits for ectopic, spontaneous abortion, and other pregnancy complications. Washington SID and SEDD data acknowledges limitations to abortion-related data, but do not specify which elements are restricted. In Wisconsin SEDD data, diagnosis and procedure coding is suppressed to eliminate the distinction between spontaneous and induced abortions.

Abortion-related data restrictions in secondary databases can prevent or bias analyses of early pregnancy complications and the health impacts of abortion legislation. For researchers, limited transparency about which elements are suppressed, particularly in states that do not specify restricted fields in advance, complicates data purchase decisions and reproducibility. Variation in restriction policies across states also undermines cross-state comparisons and trend analyses central to evaluating policy environments. Future work should develop transparent disclosure standards that better balance patient privacy (which remains crucial) with the societal need for rigorous, data-driven health policy evaluation.