Article Text

Epidemiology of Childhood Cancer.
  1. Epidemiologist

    Statistics from

    Epidemiology of Childhood Cancer. By Little J. (Pp 386, paperback; £36.00.) Oxford University Press, 1999. ISBN 9 283 22149 4 .

    Paediatric oncology textbooks often open with a chapter on epidemiology, usually rather brief and tending to reflect the particular interests of the authors or editors. Numerous review articles have dealt with a few putative risk factors, but until now there has been no comprehensive review of the evidence on childhood cancer aetiology.

    Epidemiology of Childhood Cancer represents a Herculean feat of almost single handed compilation and synthesis (only one chapter has any co-authors). Therein lies its main strengths and weaknesses. The introductory chapter contains a lucid discussion of study design, including methods of control selection and the principal sources of bias. This underpins the review of analytic studies that occupies most of the book and could be read with profit by anyone contemplating setting up an epidemiological study. Chapter 2, on descriptive epidemiology, sets a pattern by considering all relevant studies published by early 1997 relating to each major type of cancer in turn. The next eight chapters form the core of the book, each covering a group of possible risk factors. The best of these provide detailed reviews of the evidence onenvironmental factors such as ionising radiation, electromagnetic fields (one chapter each), and exposure to chemicals and dusts. Where a factor has been the subject of several studies, their characteristics and results are tabulated. The tables do not usually give confidence intervals for relative risks, and some are very long and hard to follow; the one relating to birth weight occupies five pages and has 41 footnotes.

    The book could on occasion be more critical. For example, the author remarks that smoking in pregnancy is the most important single determinant of low birth weight but does not comment on how few analyses of birth weight and childhood cancer have allowed for maternal smoking or social class. The estimates of risks for siblings of children with cancer appear not to take into account the need to allow for method of ascertainment, and are of limited value for genetic counselling. A useful concluding chapter summarises factors investigated in relation to each diagnostic group, in each case ranging from those generally accepted to be associated with the specific cancer to those not generally associated. The bibliography of over 1000 references includes virtually every epidemiological study since 1990, but is less complete for earlier publications and relevant clinical and laboratory work. The index is not very helpful. Many entries for types of cancer are unnecessary, given the structure of each chapter, while factors such as paternal age (in chapter 3, six chapters away from maternal age) are unlisted.

    Inevitably, the book is already out of date. The past two years have seen the publication of series of papers from case control studies in North America, Germany, and New Zealand, major studies of childhood cancer in relation to vitamin K, and parental occupational exposure to ionising radiation, volume 2 of International Incidence of Childhood Cancer, and yet more studies of parental smoking. What is now needed is a continuously updated, truly systematic review of childhood cancer epidemiology, with studies evaluated according to explicit, uniform criteria. Notwithstanding the criticisms above, Epidemiology of Childhood Cancer is unrivalled in scope and level of detail.

    View Abstract

    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.