Diabetes mellitus and risk of injury requiring medical care: a nationwide population-based matched cohort study in Taiwan

Background Diabetes mellitus (DM) is associated with multiple microvascular and macrovascular complications; however, its association with clinically significant injury requiring medical care remains incompletely understood. This study aimed to investigate the association between DM and subsequent risk of injury requiring emergency department visits or hospitalization using a nationwide population-based cohort. Methods We conducted a nationwide matched cohort study using Taiwan’s National Health Insurance Research Database. Adults with newly diagnosed DM were identified and matched with individuals without DM at a 1:3 ratio using propensity score matching. The propensity score model incorporated demographic characteristics, socioeconomic factors, baseline comorbidities, healthcare utilization, and overall disease burden. Covariate balance after matching was assessed using standardized mean differences (SMDs), with SMD < 0.10 indicating adequate balance. The primary outcome was clinically significant injury requiring emergency department visits or hospitalization. Multivariable Cox proportional hazards models with robust variance estimation were used to evaluate the association between DM and injury risk. Results After propensity score matching, 31,424 individuals with DM and 94,272 matched individuals without DM were included. All baseline covariates achieved adequate balance after matching, with standardized mean differences (SMDs) <0.10. During follow-up, clinically significant injury requiring emergency department visits or hospitalization occurred more frequently among individuals with DM than among those without DM. In multivariable Cox proportional hazards models accounting for the matched cohort design, DM was independently associated with an increased risk of clinically significant injury (adjusted hazard ratio [aHR], 1.645; 95% confidence interval [CI], 1.202–2.015; p<0.001). The association remained robust in sensitivity analyses and was consistent across subgroup analyses stratified by age and sex, with no significant evidence of effect modification. Conclusion In this nationwide population-based matched cohort study, DM was associated with an increased risk of clinically significant injury requiring emergency or inpatient care. These findings highlight the importance of incorporating injury risk assessment into comprehensive diabetes management, particularly among individuals with increased vulnerability due to age, diabetes-related complications, or functional impairment. Further studies incorporating detailed clinical information, including glycemic control and diabetes severity, are warranted to clarify the mechanisms underlying this association.