Seasonal variation in birth and diagnosis of cancer in children and young people in Northern England, 1968–2005
- 1Institute of Health and Society, Newcastle University, Sir James Spence Institute, Royal Victoria Infirmary, Newcastle upon Tyne, UK
- 2Northern Institute of Cancer Research, Newcastle University, Sir James Spence Institute, Royal Victoria Infirmary, Newcastle upon Tyne, UK
Objectives
To investigate seasonal variation in the incidence of cancer in children and young people, using population-based data.
Design
Data on all cases were extracted from the Northern Region Young Persons’ Malignant Disease Registry (NRYPMDR). The NRYPMDR is a specialist registry that records cancer cases in children and young adults covering the counties of Northumberland, Tyne and Wear, Durham, Teesside, and Cumbria (excluding Barrow-in-Furness).
Participants
All cancer cases aged 0–24 diagnosed during the period 1968–2005 and registered by the NRYPMDR.
Methods
The following diagnostic groups were analysed: leukaemia (acute lymphoblastic, acute non-lymphocytic), lymphoma (Hodgkin, non-Hodgkin), central nervous system tumours (astrocytoma, primitive neuroectodermal tumours (PNETS)), sympathetic nervous system tumours, retinoblastoma, renal tumours, hepatoblastoma, bone tumours (osteosarcomas, Ewing sarcoma), soft tissue sarcoma, germ cell tumours (gonadal, non-gonadal) and carcinomas (thyroid, skin, malignant melanoma, breast, cervical). The chi-squared heterogeneity test was used to test for departure from the uniform distribution. Poisson regression analysis was used to fit sinusoidal (harmonic) models to the data, using month of birth and month of diagnosis, respectively, as covariates in separate models. Analyses were carried out separately by gender and age group (0–14, 15–24 years).
Results
There were a total of 5909 childhood cancer cases; 2959 aged 0–14 years (1659 males, 1300 females) and 2950 aged 15–24 years (1590 males, 1358 females). For 0–14-year-old boys, there was statistically significant sinusoidal variation in month of birth for acute non-lymphocytic leukaemia (p = 0.04; peak in September) and astrocytoma (p = 0.03; peak in October). Based on month of diagnosis, there was statistically significant sinusoidal variation in girls for all lymphomas (p = 0.05; peak in March) and Hodgkin lymphoma (p = 0.005; peak in January), and in boys for osteosarcoma (p = 0.05; peak in October). For 15–24-year-olds, there was significant heterogeneity for germ cell tumours (p = 0.04), cervical (p = 0.03) and female breast carcinoma (p = 0.05), based on month of birth, and PNETs (p = 0.05) and skin carcinoma (p = 0.05), based on month of diagnosis. Significant sinusoidal variation in month of birth for malignant melanoma in females (p = 0.03; peak in March) and cervical carcinoma (p = 0.03; peak in October) was observed.
Conclusions
These findings suggest that seasonal environmental factors around the time of birth or time of diagnosis may be involved in the aetiology of specific diagnostic groups. Further research is needed to study possible aetiological mechanisms and factors. Putative agents include sunlight, pesticides, diet and infections.
