OP1-2 - Characterizing Fit-for-Purpose Real-World Data: An Assessment of a Mother–Infant Linkage in the Japan Medical Data Center Claims Database

Introduction: The potential for administrative databases to inform medication safety during pregnancy has been increasingly recognized. Mother–infant linkages in databases enable evaluation of infant outcomes. However, database availability, which has increased in recent years, does not inherently dictate suitability to generate evidence to inform regulatory decision making (“fit for regulatory purpose”), emphasizing the importance of fit-for-purpose database evaluation.

Objective: To use the Duke-Margolis framework to assess whether a linked cohort of mothers and infants in the Japan Medical Data Center (JMDC) claims database is fit for purpose within the regulatory context of estimating infant outcomes associated with in utero exposure to marketed medications.

Methods: The Duke-Margolis framework considers whether a database is fit for regulatory purpose based on relevancy and quality. Relevancy relates to capacity to answer the research question, in terms of availability of critical data fields and a sufficiently sized, representative population. Quality relates to ability to validly answer the research question, in terms of data completeness, accuracy, and transparency. To assess these considerations, we estimated the number of pregnancies that could be linked to an infant among females ages 12–55 in the JMDC between January 2005 and March 2022 using two different linkage approaches. Descriptive characteristics were examined.

Results: In terms of relevancy, we determined that critical data fields (maternal medication exposures, infant major congenital malformations, covariates) were available. Family identification codes permitted patient-level mother–infant linkage. 385,295 total mother–infant pairs were identified, representing about 2% of live births in Japan during the study period. About 47,000 congenital malformations were observed among these pairs. Comparison to publicly available data from Japan suggested congenital malformations were over-represented (12.2% versus 5.7%) and preterm births were under-recorded (3.6% versus 5.6%) in this population. Maternal characteristics, however, appeared mostly consistent with the population of same-aged females in Japan. 

In terms of quality, our methods were expected to accurately identify the complete set of mothers and infants in the JMDC enrolled in a shared health insurance plan. Women with evidence of a live birth delivery had a linkage rate of about 50%, which aligns with expectations of infant insurance coverage under the mother’s, versus other parent’s, plan. Cross-tabulation of values indicated for the relationship of the “mother” and “infant” to the insurance holder allowed for confirmation of plausible biologic mother–infant pairs. However, the completeness and accuracy of gestational age information was limited given the lack of live birth delivery codes for the majority (60%) of the cohort coupled with suppression of infant birth dates and inaccessibility of International Classification of Diseases codes with fifth level digits (where gestational week information would have been available) in the database.

Conclusions: Results suggest the JMDC may be well-suited for descriptive studies of pregnant women in Japan (e.g., comorbidities, medication usage). More work is needed to identify a method to assign pregnancy onset and delivery dates so that in utero exposure windows can be defined more precisely as needed for many regulatory postapproval pregnancy safety studies.