Commentary: Birth order and risk of childhood acute lymphoblastic leukaemia (ALL)

Open Access
Publisher Website
Google Scholar
Abstract
In this issue of the International Journal of Epidemiology, Dockerty and colleagues1 report a case-control study in which a number of parental, familial and social factors were evaluated in relation to risk of different childhood cancers. The rationale of the study is a rather curious mix of four different hypotheses, only loosely interconnected by the parameters studied. But it does have two outstanding features—its very large size and a design that minimizes selection bias. These provide the potential for robust statistical inquisition of associations, or the lack thereof. Cases (age 0–14 years) total almost 11 000 including 3878 leukaemias, all accrued from the British National Registry of Childhood Tumours for the period 1968–1986. Controls were selected from the Office of Population Censuses and Surveys (OPCS, now Office for National Statistics) birth data and matched for gender, date of birth (6-month periods) and birth registration sub-district.