Although seasonal trends in incidence and diagnosis of pediatric cancers have been widely investigated, the results have been inconclusive. A consistent seasonal trend may possibly provide etiological insights into pediatric cancers. This study aims to determine if there is a seasonal variation in cancer diagnoses in the pediatric population at the IWK Health Centre, a tertiary care center serving three Canadian provinces: Nova Scotia, New Brunswick, and Prince Edward Island. All pediatric cancer patients aged 0-20 y diagnosed from 1995 to 2015 at the center were included in this study. The annual data was divided into four seasonal periods (December to February, March to May, June to August, and September to November). The cancer diagnoses were categorized as leukemia, lymphoma, sarcoma, brain tumors, and miscellaneous. Seasonal variation was assessed by a harmonic function in a Poisson regression model. The amplitude of multiplicative change in the incidence rate caused by the seasonal variation is expressed as the incidence rate ratio (IRR). For all cancer diagnoses for the entire cohort of 1200 patients, the IRR was 1.03 [95% confidence interval (CI) 0.96-1.13]. None of the IRRs for the cancer groups indicated a statistically significant seasonality of cancer diagnosis: Leukemia 1.11 (95% CI 0.96-1.28); Lymphoma 1.17 (95% CI 0.93-1.47); Sarcoma 1.29 (95% CI 0.99-1.69); Brain tumors 1.16 (95% CI 0.97-1.38); Miscellaneous 1.09 (95% CI 0.93-1.27). The present study did not show a seasonal variation in the various cancer types in the pediatric population at the IWK.
Keywords: Childhood cancer; Pediatric oncology; Population based; Seasonality.