Br J Cancer. 2026 Jul 23. doi: 10.1038/s41416-026-03560-5. Online ahead of print.
ABSTRACT
BACKGROUND: Limited knowledge exists on the association between hospital contacts before a childhood cancer diagnosis and the risk of early mortality.
METHODS: We conducted a register-based cohort study including 14,089 children with cancer diagnosed in 1989 – 2014 in Denmark and Sweden. Odds ratios (ORs) and 95% confidence intervals (CIs) of early mortality (deaths within 0-90 days) were estimated by nights in hospital up to two years prior cancer diagnosis and diagnoses within specific main diagnostic categories using logistic regression.
RESULTS: Children hospitalized ≥14 nights within 2 years-14 days and 3 months-14 days before cancer diagnosis faced elevated ORs of early mortality (4.6, 95% CI 3.2-6.4; 4.4, 2.9-6.7). The association was not explained by maternal education. Early mortality was associated with diagnoses within several diagnostic categories prior to cancer diagnosis, including infectious (OR 1.5, 95% CI 1.1-2.0) and endocrine diseases (OR 1.6, 95% CI 1.02-2.5), blood (OR 3.3, 95% CI 2.1-5.1) and digestive disorders (OR 1.9, 95% CI 1.4-2.6), and symptoms with no specific diagnosis (OR 1.4, 95% CI 1.1-8).
CONCLUSION: Children with increased time in hospital before cancer diagnosis had a higher risk for early mortality, which indicates underlying health vulnerabilities or pre-existing fragility in this population.
PMID:42493601 | DOI:10.1038/s41416-026-03560-5
