IJE Advance Access originally published online on July 31, 2007
International Journal of Epidemiology 2007 36(4):929; doi:10.1093/ije/dym061
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Letters to the Editor |
Childhood leukaemia and socioeconomic status
1Cancer Control Research Program, British Columbia Cancer Agency, Vancouver, BC Canada.
2Department of Health Care and Epidemiology, University of British Columbia, Vancouver, BC Canada.
3Electric Power Research Institute Inc., Palo Alto CA.
4Health Analysis and Measurement Group, Statistics Canada, Ottawa Canada.
*Corresponding author. E-mail: mborugian{at}bccrc.ca
In a recent issue of the journal, we were pleased to see continued interest by Smith and colleagues1 in the relationship between socioeconomic status (SES) and the development of childhood leukaemia. We agree with the authors that the completeness and representativeness of their data as well as the ability to measure deprivation in several ways are strengths of their study.
We would like to point out to the authors that although on the surface the methodology of the UK investigation was different from our recent registry-based Canadian study,2 in fact, the studies were similar. The major difference was that our study did not include enhanced surveillance to identify the childhood leukaemia cases. Due to the centralized cancer registration and treatment system in Canada, however, it is likely that our registries do not suffer from the underreporting problems which seem to have been present in the UK cancer registration.3,4 Therefore, it is likely that both studies captured most, if not all, childhood leukaemia cases. The other differences were minor. We used the entire population as our comparison group rather than selecting a subset of the population as a control group, and we included a longer time period (1985–2001 for most provinces) in the analysis. Finally, our study was unable to look at SES at different time points.
In fact, the results of two studies gave very similar results for risk associated with SES at diagnosis/reference date. For example, based on 4024 cases of acute lymphoblastic leukaemia, comparing the poorest of five categories with the richest, we found a rate ratio (RR) of 0.86 [95% confidence interval (CI) 0.78–0.95] whereas based on 1578 cases the UK study found an odds ratio (OR) of 0.90 (95% CI 0.75–1.07). Due to the larger sample size of our study, we concluded that SES was significantly associated with the risk of childhood leukaemia. Had the UK study examined cases over a longer time period, it is quite possible that they would have also found a significant association. It is clear that continued discussion of this relationship is worthwhile, since SES may very well be an important factor in childhood leukaemia risk.
References
1 Smith A, Roman E, Simpson J, et al. Childhood leukaemia and socioeconomic status: fact or artefact? A report from the United Kingdom childhood cancer study (UKCCS). Int J Epidemiol (2006) 35:1504–13.
2 Borugian MJ, Spinelli JJ, Mezei G, et al. Childhood Leukemia and Socioeconomic Status in Canada. Epidemiology (2005) 16:526–31.[CrossRef][ISI][Medline]
3 Parkin D, Muir C, Whelan SL, et al. Cancer incidence in five continents. Volume VII. IARC Sci Publ (1997) 143:138–89.
4 Parkin D, Muir C, Whelan SL, et al. Cancer incidence in five continents. Volume VII. IARC Sci Publ (2002) 155:128–53.
CiteULike 
Connotea 
Del.icio.us What's this?
Related articles in Int. J. Epidemiol.:
- Childhood leukaemia and socioeconomic status
- Alex Smith, Eve Roman, and Jill Simpson
Int. J. Epidemiol. 2007 36: 1156.[Extract] [FREE Full Text]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||