Article Text

Download PDFPDF

Cancer and geography
003 Demographic analyses of primary bone cancer in 0–49 year olds in Great Britain, 1980–2005: a small-area approach
  1. K Blakey1,
  2. R G Feltbower2,
  3. R C Parslow2,
  4. P W James1,
  5. B Gómez Pozo1,
  6. C Stiller3,
  7. T J Vincent3,
  8. P Norman4,
  9. P A McKinney2,
  10. M F Murphy3,
  11. A W Craft5,
  12. R J Q McNally1
  1. 1Institute of Health and Society, Newcastle University, Newcastle-upon- Tyne, UK
  2. 2Paediatric Epidemiology Group, University of Leeds, Leeds, UK
  3. 3Childhood Cancer Research Group, Department of Paediatrics, University of Oxford, Oxford, UK
  4. 4School of Geography, University of Leeds, Leeds, UK
  5. 5Northern Institute of Cancer Research, Newcastle University, Newcastle-upon- Tyne, UK


Objective To examine geographical patterning in the incidence of primary bone cancers diagnosed in 0–49 year olds in Great Britain (GB) during the period 1980–2005. The analyses focussed on the two most common types in this age range, osteosarcoma and Ewing sarcoma. We specifically aimed to analyse putative associations with area characteristics including deprivation and population density.

Design The study accessed multiple data sources including population census, digital boundary, postcode directories and time series of Townsend deprivation scores. Incidence data from the 10 regional cancer registries in GB were accessed and analysed by census small-area level (census ward level for England and Wales; postcode sector level for Scotland).

Setting GB.

Participants Data from patients (0 to 49 years) diagnosed with a primary bone cancer between 1st January 1980 and 31st December 2005 and registered with one of the 10 regional cancer registries in GB.

Main outcome measure Negative binomial regression was used to examine the relationship between incidence rates and population density, Townsend deprivation index (and its components). The models were fitted to small-area level data and aggregated to four age bands (0–14 years; 15–29 yrs; 30–49 years and 0–49 years) and by gender with the logarithm of the “at risk” population as an offset.

Results The study analysed 2566 cases of osteosarcoma and 1650 cases of Ewing sarcoma. After adjustment for age and gender there was a statistically significant negative association for the incidence of osteosarcoma with area-level Townsend deprivation score (RR for one unit increase in level of deprivation 0.975; 95% CI 0.963 to to 0.986). For Ewing sarcoma, after adjustment for age and gender, there was a statistically significant negative association for incidence with population density (RR for an increase of one person per hectare 0.981; 95% CI 0.972 to to 0.989) and also with non-car ownership (RR for an increase of one percent in non-car ownership 0.996; 95% CI 0.993 to to 1.000).

Conclusion Higher deprivation appears to have a protective effect on the incidence of osteosarcoma. Higher incidence of Ewing sarcoma was associated with living in less densely populated areas and greater levels of car ownership, both of which are characteristic of rural areas. This study contributes to the growing body of evidence linking risk of Ewing sarcoma to some aspect of agriculture and suggests further study of environmental exposures or land use may be informative.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.