Am J Epidemiol 2004; 159:717.
Copyright © 2004 by the Johns
Hopkins Bloomberg School of Public Health
LETTERS TO THE EDITOR |
THE AUTHORS REPLY
1 Epidemiology and Genetics Unit, University of Leeds, Leeds LS2 9LN, United Kingdom
2 Paediatric Epidemiology Group, University of Leeds, Leeds LS2 9LN, United Kingdom
We thank Professor Kinlen (1) and Dr. Tucker (2) for their observations on our recently published United Kingdom Childhood Cancer Study (UKCCS) report on childhood cancer and population mixing (3). The investigation was aimed at evaluating the relation between population mixing and childhood leukemia risk in small geographic areas across the United Kingdom. Subjects involved in the UKCCS were not selected on the basis of the leukemia incidence or population density of their region of residence. Furthermore, the census-based method used is reproducible and robust, and the analyses are not affected by participation or recall bias.
We are aware that Professor Kinlen’s hypothesis (4) is not applicable to the majority of leukemias diagnosed in children and that it is difficult to test in a national setting (5). In this context, we agree that it is important to focus on extremes of population mixing in sparsely populated areas (1), both within the United Kingdom and elsewhere in the world. With respect to the former, such an investigation will form the basis of an upcoming UKCCS analysis comparing small-area census data of cases with those of controls at birth as well as at diagnosis. The recent acquisition of the birth certificates of all subjects registered in the UKCCS means that we can investigate area characteristics at both of these time points, as well as examine mobility and other changes occurring in between.
Dr. Tucker (2) rightly notes that we did not provide data on the age and diagnostic group most relevant to the delayed infection hypothesis (4). When the analysis was restricted to common acute lymphoblastic leukemia (ALL) diagnosed between the ages of 2 and 5 years, the results were similar to those presented for total ALL (3). For example, for ALL in the lowest category of diversity of migrants, the odds ratio was 1.29 (95 percent confidence interval: 0.79, 2.12) as compared with 1.37 (95 percent confidence interval: 1.00, 1.86) for the totality. The results did not differ when adjustment was made for deprivation and rural status, and there was no evidence of increased risk for areas with a high volume of migrants.
The biologic diversity of childhood leukemias makes it unlikely that there is a solitary cause. It seems clear, however, that research on possible immunologic and infectious etiologies is worth pursuing.
REFERENCES
- Kinlen LJ. Re: "Childhood cancer and population mixing." (Letter). Am J Epidemiol 2004;159:716.
[Free Full Text] - Tucker M. Re: "Childhood cancer and population mixing." (Letter). Am J Epidemiol 2004;159:716–17.
[Free Full Text] - Law GR, Parslow R, Roman E. Childhood cancer and population mixing. United Kingdom Childhood Cancer Study Investigators. Am J Epidemiol 2003;158:328–36.
[Abstract/Free Full Text] - Greaves MF. Aetiology of acute leukaemia. Lancet 1997;349:344–9.[CrossRef][ISI][Medline]
- Kinlen LJ. Evidence for an infective cause of childhood leukaemia: comparison of a Scottish New Town with nuclear reprocessing sites in Britain. Lancet 1988;2:1323–7.[CrossRef][ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||