The T-MSIS Analytic Files (TAF) Analysis Reporting Checklist: A Guide for Research Using Medicaid Claims Data.

Importance: Medicaid is the single largest source of health care coverage in the US, but health policy research on the Medicaid program has historically lagged research on Medicare due to limited availability of high-quality administrative claims data across states. In 2019, the US Centers for Medicare & Medicaid Services released the T-MSIS Analytic Files (TAF), a new-generation federal Medicaid claims dataset that has catalyzed policy-relevant research on the Medicaid program. TAF data are highly complex, however, with meaningful differences in quality across states, years, and data elements. There is an urgent need for standardized reporting guidelines to ensure TAF-based research is high quality and reproducible. Objective: To develop a checklist to guide reporting of research using the TAF data. Evidence Review: The development of the TAF Analysis Reporting Checklist was led by a subcommittee of the Medicaid Data Learning Network (MDLN), a national consortium of research teams focused on developing and disseminating best practices for conducting health services research with the TAF data. The subcommittee first created a draft checklist drawing from published technical guidance on proper use of the TAF data, as well as published analyses of TAF data quality. This draft was iteratively refined based on feedback from (1) MDLN members; (2) the MDLN Advisory Group, composed of leaders in academia, government, and industry with Medicaid claims experience; (3) editors of health policy journals; and (4) the broader Medicaid research community. Findings: The final checklist includes 4 categories of items that are recommended for reporting in studies using the TAF data. This includes (1) details on the specific data used (files, years, release versions, and size of the data extract), (2) how the analytic sample was defined (eligibility criteria, enrollment span, and scope of benefits), (3) what states and/or territories were excluded from the analysis based on data quality concerns (and the exclusion criteria used to do so), and (4) additional information on special considerations, including use of spending data and changes in data quality over time. Conclusions and Relevance: The TAF Analysis Reporting Checklist represents a consensus effort to identify items researchers should report to promote transparency and reproducibility in TAF-based studies. This reporting is a key step in safeguarding the quality of research used to inform Medicaid policy.