Incidence and Long‐Term Outcomes of Pediatric Tracheostomy in Japan: A Cohort Study Using a Japanese Claims Database

This is the authors' abstract. We don't add key points for this paper.

BACKGROUND: Epidemiological data on pediatric tracheostomy are limited, particularly in Asia. We estimated the incidence and evaluated long-term post-tracheostomy outcomes and their predictors using a large Japanese claims database. METHODS: We used the Japanese administrative claims database (JMDC) from 2005 to 2025. All individuals aged < 20 years observed during continuous enrollment contributed person-time to the incidence analysis. Individuals who underwent incident tracheostomy constituted the post-tracheostomy cohort. Cumulative incidences of decannulation and death were estimated using competing-risk methods, and predictors of each outcome were evaluated with cause-specific Cox proportional hazards models. RESULTS: Among 6,046,879 individuals aged < 20 years, 1032 underwent incident tracheostomy and constituted the post-tracheostomy cohort (median age, 1.2 years [IQR, 0.4-12.6]). The incidence was 4.0 (95% CI, 3.7-4.2) per 100,000 person-years; it was highest in infants (< 1 year: 42.4) and increased modestly in late adolescence (15-19 years: 2.8). The cumulative incidence of decannulation was 22.9% (95% CI, 20.3-25.5) at 1 year and 40.6% (95% CI, 37.1-44.1) at 5 years; the cumulative incidence of death was 12.9% (95% CI, 10.7-15.0) at 1 year and 22.2% (95% CI, 19.4-25.1) at 5 years. Older age (6-19 years) and trauma were associated with earlier decannulation, whereas malignancy, cardiovascular disease, congenital/genetic comorbidity, and post-cardiac arrest diagnosis were associated with earlier death. CONCLUSION: The incidence of pediatric tracheostomy peaked in infancy and increased modestly in late adolescence. During long-term follow-up, substantial proportions of patients experienced decannulation or death, with distinct predictors for each outcome.