Validation of US CDC National Death Index mortality data, focusing on differences in race and ethnicity.

OBJECTIVES: The US Center for Disease Control and Prevention's National Death Index (NDI) is a gold standard for mortality data, yet matching patients to the database depends on accurate and available key identifiers. Our objective was to evaluate NDI data for future healthcare research studies with mortality outcomes. METHODS: We used a Kaiser Permanente Mid-Atlantic States' Virtual Data Warehouse (KPMAS-VDW) sourced from the Social Security Administration and electronic health records on members enrolled between 1 January 2005 to 31 December 2017. We submitted data to NDI on 1 036 449 members. We compared results from the NDI best match algorithm to the KPMAS-VDW for vital status and death date. We compared probabilistic scores by sex and race and ethnicity. RESULTS: NDI returned 372 865 (36%) unique possible matches, 663 061 (64%) records not matched to the NDI database and 522 (<1%) rejected records. The NDI algorithm resulted in 38 862 records, presumed dead, with a lower percentage of women, and Asian/Pacific Islander and Hispanic people than presumed alive. There were 27 306 presumed dead members whose death dates matched exactly between the NDI results and VDW, but 1539 did not have an exact match. There were 10 017 additional deaths from NDI results that were not present in the VDW death data. CONCLUSIONS: NDI data can substantially improve the overall capture of deaths. However, further quality control measures were needed to ensure the accuracy of the NDI best match algorithm.

Development of a Common Data Model for a Multisite and Multiyear Study of Virtual Visit Implementation: A Case Study.

BACKGROUND/OBJECTIVE: In multisite studies, a common data model (CDM) standardizes dataset organization, variable definitions, and variable code structures and can support distributed data processing. We describe the development of a CDM for a study of virtual visit implementation in 3 Kaiser Permanente (KP) regions. METHODS: We conducted several scoping reviews to inform our study's CDM design: (1) virtual visit mode, implementation timing, and scope (targeted clinical conditions and departments); and (2) extant sources of electronic health record data to specify study measures. Our study covered the period from 2017 through June 2021. Integrity of the CDM was assessed by a chart review of random samples of virtual and in-person visits, overall and by specific conditions of interest (neck or back pain, urinary tract infection, major depression). RESULTS: The scoping reviews identified a need to address differences in virtual visit programs across the 3 KP regionsto harmonize measurement specifications for our research analyses. The final CDM contained patient-level, provider-level, and system-level measures on 7,476,604 person-years for KP members aged 19 years and above. Utilization included 2,966,112 virtual visits (synchronous chats, telephone visits, video visits) and 10,004,195 in-person visits. Chart review indicated the CDM correctly identified visit mode on>96% (n=444) of visits, and presenting diagnosis on >91% (n=482) of visits. CONCLUSIONS: Upfront design and implementation of CDMs may be resource intensive. Once implemented, CDMs, like the one we developed for our study, provide downstream programming and analytic efficiencies by harmonizing, in a consistent framework, otherwise idiosyncratic temporal and study site differences in source data.