Data quality of device identification in FDA MAUDE reports for cardiovascular devices
DIGITAL HEALTH · 2026-02 · 12
DOI: 10.1177/20552076261504493 ↗Kurz gefasst
Fachartikel in der Zeitschrift DIGITAL HEALTH, online erschienen am 02.11.2026. Verfasst von Matthew W. Segar und 4 weiteren Autor:innen. Die Arbeit nennt 24 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Objective
To quantify how often cardiovascular adverse event reports in the FDA Manufacturer and User Facility Device Experience (MAUDE) database can be resolved to a specific device in the Global Unique Device Identification Database (GUDID) and to characterize identification failures.
Methods
Cross-sectional analysis of MAUDE reports received from January 2019 through February 2026 for 32 cardiovascular product codes in 7 categories. Reports were matched to GUDID by UDI-DI (Tier 1), model number (Tier 2), catalog number (Tier 3), or fuzzy brand name (Tier 4), and Tier 4 accuracy was assessed in a stratified random sample. Manufacturer names were deduplicated using DUNS numbers.
Results
Among 942,414 reports, 880,039 (93.4%) were matched: 57.4% at Tier 1, 10.0% at Tier 2, 1.1% at Tier 3, and 24.9% at Tier 4. Among 328 sampled Tier 4 matches, category-weighted brand-level accuracy was 84.2% (95% CI 79.4–88.9); most errors involved another component of the same system, and the specific model was rarely identified. UDI-DI field presence rose from 52.0% in 2019 to 82.6% in 2026. Death reports had the highest match rate (96.2%) but the lowest Tier 1 rate (48.4%). Medtronic appeared under 34 name variants.
Conclusion
Most cardiovascular MAUDE reports can be linked to a GUDID device entry, but 26.6% of matched reports rely on fuzzy brand name matching, which identifies the brand rather than the model and misattributes about one in six reports. Researchers should report device identification methods, verify matching for their product codes, and conduct sensitivity analyses across matching stringencies.
Zitieren
APA
Segar, M. W., Lambeth, K. D., Parsa, S., Sinha, K., & Razavi, M. (2026). Data quality of device identification in FDA MAUDE reports for cardiovascular devices. DIGITAL HEALTH, 12. https://doi.org/10.1177/20552076261504493
Vancouver
Segar MW, Lambeth KD, Parsa S, Sinha K, Razavi M. Data quality of device identification in FDA MAUDE reports for cardiovascular devices. DIGITAL HEALTH. 2026;12. doi:10.1177/20552076261504493
BibTeX
@article{Segar2026,
author = {Segar, Matthew W. and Lambeth, Kaleb D. and Parsa, Shyon and Sinha, Kavya and Razavi, Mehdi},
title = {Data quality of device identification in FDA MAUDE reports for cardiovascular devices},
journal = {DIGITAL HEALTH},
year = {2026},
volume = {12},
doi = {10.1177/20552076261504493}
}
EUVIMED