American Journal of Epidemiology Advance Access originally published online on October 12, 2005
American Journal of Epidemiology 2005 162(11):1132-1133; doi:10.1093/aje/kwi333
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Letter to the Editor |
RE: "DETECTING SMALL-AREA SIMILARITIES IN THE EPIDEMIOLOGY OF CHILDHOOD ACUTE LYMPHOBLASTIC LEUKEMIA AND DIABETES MELLITUS, TYPE 1: A BAYESIAN APPROACH"
1 Cancer Epidemiology Unit, CeRMS and CPO, University of Turin, 10127 Turin, Italy
2 Medical Statistics and Epidemiology Unit, Department of Medical Sciences, Eastern Piedmont University, 28100 Novara, Italy
3 Department of Internal Medicine, University of Turin, 10127 Turin, Italy
4 Division of Pediatrics, Medical Sciences Department, Eastern Piedmont University, 28100 Novara, Italy
(e-mail: lorenzo.richiardi{at}unito.it)
Feltbower et al. (1
) failed to replicate previous findings from an ecologic study of a positive international correlation (r = 0.53) between the incidence rates of childhood acute lymphoblastic leukemia and childhood type 1 (insulin-dependent) diabetes (2). Carrying out an elegant analysis, they investigated whether the incidences of the two diseases correlate within small areas in a region in the north of the United Kingdom. The estimated correlation was 0.33, which decreased to a negligible correlation after adjustment for the deprivation index.
We took advantage of two Italian registries, the Childhood Cancer Registry of Piedmont (3) and the Type 1 Diabetes Registry of the Province of Turin (4), to assess whether the temporal trends in incidence rates of the two diseases are parallel in the part of the Piedmont region covered by both registries (Province of Turin, 270,313 children aged 0–14 years in 1995). The first registry was established at the end of the 1960s, and it includes complete and reliable data on acute lymphoblastic leukemia incidence for the period 1975–2001, whereas the second registry records information on cases of type 1 diabetes from 1984 through 2000. Separate reports of temporal trends in the incidences of acute lymphoblastic leukemia (3) and type 1 diabetes (4) have been published previously.
We carried out analyses of incidence of the two diseases by calendar period and birth cohort for the period 1984–2000 and ages 0–14 years. In order to analyze a sufficient number of points in time, analyses by the cohort of birth were limited to the 10-year age classes in which the incidences peak: 0–9 years for acute lymphoblastic leukemia (75 cases born between 1979 and 1986) and 5–14 years for diabetes (169 cases born between 1979 and 1986). We used the Pearson correlation coefficient (r) to estimate the correlation.
Overall crude yearly incidence rates per 100,000 inhabitants were 10.3 for diabetes, based on 525 observed cases, and 3.9 for acute lymphoblastic leukemia, based on 199 cases. The age-standardized incidence rates (European standard population) of acute lymphoblastic leukemia and type 1 diabetes versus calendar period are reported in the left side of figure 1. The correlation was weak, with a correlation coefficient of 0.24 (p = 0.38). As shown in the right side of figure 1, the incidence rates of type 1 diabetes and acute lymphoblastic leukemia did not positively correlate within the same birth cohorts (r = –0.34, p = 0.41).
|
The main strength of our study is that we were able to analyze the incidence of both diseases within the same birth cohorts by use of information from population-based registries; the main limitation is the limited number of cases involved in the analysis, resulting in unstable incidence estimates. Our results support those of Feltbower et al. (1
), indicating that, despite the epidemiologic similarities between acute lymphoblastic leukemia and type 1 diabetes, their incidences do not correlate when analyzed in well-defined populations. ACKNOWLEDGMENTS
This work has been carried out within the framework of projects partially supported by the Special Project "Oncology," Compagnia San Paolo/FIRMS, and the Italian Association for Research on Cancer (AIRC).
The authors thank Alberto Martinasso and Valentina Vergano for their comments on the manuscript.
Conflict of interest: none declared.
NOTES
Editor's note: In accordance with Journal policy, Feltbower et al. were asked whether they wished to respond to this letter, but they chose not to do so.
References
- Feltbower RG, Manda SOM, Gilthorpe MS, et al. Detecting small-area smilarities in the epidemiology of childhood acute lymphoblastic leukemia and diabetes mellitus, type 1: a Bayesian approach. Am J Epidemiol 2005;161:1168–80.
[Abstract/Free Full Text] - Feltbower RG, McKinney PA, Greaves MF, et al. International parallels in leukaemia and diabetes epidemiology. Arch Dis Child 2004;89:54–6.
[Abstract/Free Full Text] - Magnani C, Dalmasso P, Pastore G, et al. Increasing incidence of childhood leukemia in northwest Italy, 1975–98. Int J Cancer 2003;105:552–7.[CrossRef][ISI][Medline]
- Bruno G, Merletti F, Biggeri A, et al. Increasing trend of type I diabetes in children and young adults in the Province of Turin (Italy). Analysis of age, period, and birth cohort effects from 1984 to 1996. Diabetologia 2001;44:22–5.[CrossRef][ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||