Childhood cancers: space-time distribution in Britain

Area of Science:

• Epidemiology
• Environmental Health
• Pediatric Oncology

Background:

• Childhood cancers, including leukaemias and solid tumours, represent a significant health concern.
• Understanding the aetiology of childhood cancers is crucial for prevention and early detection.
• Previous studies have explored geographical and temporal patterns in cancer incidence, but space-time interactions require further investigation.

Purpose of the Study:

• To investigate space-time clustering in childhood cancers (leukaemias and solid cancers) at birth, diagnosis, and death.
• To determine if observed patterns are statistically independent or interconnected phenomena.
• To identify distinct space-time patterns specific to childhood leukaemias versus solid cancers.

Main Methods:

• Utilized a national dataset of 22,360 children (aged 0-15) diagnosed with fatal cancers between 1953-1980 in England, Scotland, and Wales.
• Employed space-time cluster analysis, dividing data into hypothesis generation and testing sets.
• Analyzed clustering based on date and place of birth, diagnosis, and death.

Main Results:

• Significant space-time clustering was observed for childhood leukaemias and lymphomas based on birth date/place (within 1 km, 5 months) and diagnosis date/place (within 3-5 km, 9 months).
• No significant clustering was found among childhood solid cancers.
• Findings were robust, confirmed by independent analyses on separate data subsets.

Conclusions:

• Birth and diagnosis clustering in leukaemias are separate, statistically independent phenomena.
• Clustering patterns suggest multiple, localized exposures to hazards with short latent intervals, possibly infectious agents or environmental toxins.
• The distinct clustering patterns for leukaemias and solid cancers imply different etiological factors may be involved.