JMIR Publications

ABSTRACT

Background:

Childhood cancer causes significant long-term morbidity and mortality. Therefore, tracking the complications and assessing the long-term health risks in pediatric and adolescent cancer survivors are necessary.

Objective:

This study aims to explore the temporal diagnostic patterns of all childhood cancer survivors based on nationwide cohort data and to evaluate late-stage mortality by applying a trajectory group-based method.

Methods:

Data was collected from the nationwide claims database of the entire Korean population. Patients diagnosed and treated for cancer before the age of 20 between 2003 and 2017 were included. With 8,119 patients who survived > 10 years, three trajectory groups were classified according to yearly changes in the number of diagnoses (the lowest in group 1 and the highest in group 3). We analyzed whether there were significant differences among trajectory groups. Additionally, we calculated the hazard ratio (HR) for all deaths by trajectory group through a cox regression model, and plotted the pattern using the Kaplan-Meier survival curve.

Results:

There were significant differences in the patterns of most of the comorbidity and survival rates between three trajectory groups. In particular, the mental and behavioral disorders, neoplasms and diseases of the blood organs were higher in group 3 than the other two groups. Overall survival was the highest and lowest in groups 1 and 3, respectively (P < .001). In multivariate analysis for late mortality, group 3 had the highest hazard ratios of 4.37 (95% CI, 2.57–7.42; P<.001).

Conclusions:

Our study presented that the pattern of comorbidities significantly differed according to the trajectory group for late mortality; this could be the reference for physicians to recognize patients at risk for late mortality in childhood cancer survivors.