Abstract
INTRODUCTION: The inclusion of spinal cord injury (SCI) in the Global Burden of Diseases (GBD) study in 2016 as a distinct entity represented a significant milestone in recognizing and quantifying its global health impact. The study's purpose was to quantify the discrepancies between SCI incidence rates estimated by the GBD and those reported in published literature to assess the external validity of GBD outputs at the country level and propose methodological refinements to improve their utility for policymakers, clinicians, and global health stakeholders. METHODS: We compared GBD 2016, 2019, and 2021 country-level SCI incidence estimates with data from 10 systematic reviews and a scoping search in Embase. Studies were eligible for inclusion if they explicitly reported both case counts and incidence rates of SCI at the country level. Two authors checked the data for accuracy and reliability purposes. We categorized studies into two periods - pre-2000 and 2000 to 2024 - to align with key GBD benchmark years: 1990 for pre-2000 data, and 2016, 2019, and 2021 for post-2000 data. A meta-analysis calculated SCI incidence rates for each country across two timeframes (pre-2000 and 2000-2024). We then computed a discrepancy ratio (DR) by dividing GBD estimates by literature-based rates, categorizing discrepancies as low (DR < 2), moderate (2 to <5), substantial (5 to <10), or significant (DR > 10). Further analysis explored countries with national registries, the impact of prehospital mortality, and GBD's SCI definition and reporting methodology. RESULTS: In our analysis of 30 countries, we identified a high proportion of significant (50%) and substantial (23.4%) discrepancies relative to GBD 2016 estimates. When compared against GBD 2021 data, significant discrepancies decreased to 6.6%, though substantial discrepancies rose to 50%; indicating an overall improvement but persistent overestimation. Japan, South Africa, Taiwan, and Tanzania showed low discrepancies, while Denmark, France, Italy, and Spain continued to have significant discrepancies in GBD 2021. We identified three key areas for GBD improvement: unclear inclusion of non-traumatic SCI, lack of incidence stratification by severity, and omission of sports-related SCI as a distinct etiology. CONCLUSION: Early GBD SCI estimates were often overestimates, but continuous methodological refinements are leading to better alignment with real-world data. Future GBD reports should further enhance interpretability by integrating injury severity data and a more granular breakdown of SCI etiologies.