The effect of health insurance on childhood cancer survival in the United States.

BACKGROUND

The effect of health insurance on childhood cancer survival has not been well studied. Using Surveillance, Epidemiology, and End Results (SEER) data, this study was designed to assess the association between health insurance status and childhood cancer survival.

METHODS

Data on cancers diagnosed among children less than 15 years old from 2007 to 2009 were obtained from the SEER 18 registries. The effect of health insurance at diagnosis on 5-year childhood cancer mortality was estimated with marginal survival probabilities, restricted mean survival times, and Cox proportional hazards (PH) regression analyses, which were adjusted for age, sex, race/ethnicity, and county-level poverty.

RESULTS

Among 8219 childhood cancer cases, the mean survival time was 1.32 months shorter (95% confidence interval [CI], -4.31 to 1.66) after 5 years for uninsured children (n = 131) versus those with private insurance (n = 4297), whereas the mean survival time was 0.62 months shorter (95% CI, -1.46 to 0.22) for children with Medicaid at diagnosis (n = 2838). In Cox PH models, children who were uninsured had a 1.26-fold higher risk of cancer death (95% CI, 0.84-1.90) than those who were privately insured at diagnosis. The risk for those with Medicaid was similar to the risk for those with private insurance at diagnosis (hazard ratio, 1.06; 95% CI, 0.93-1.21).

CONCLUSIONS

Overall, the results suggest that cancer survival is largely similar for children with Medicaid and those with private insurance at diagnosis. Slightly inferior survival was observed for those who were uninsured in comparison with those with private insurance at diagnosis. The latter result is based on a small number of uninsured children and should be interpreted cautiously. Further study is needed to confirm and clarify the reasons for these patterns. Cancer 2017;123:4878-85. © 2017 American Cancer Society.