Surveying the literature from animal experiments: Systematic review and meta-analysis are important contributions

doi:10.1136/bmj.331.7508.110-b

BMJ 2005;331:110 (9 July), doi:10.1136/bmj.331.7508.110-b

Letter

Surveying the literature from animal experiments

Systematic review and meta-analysis are important contributions

The first 150 words of the full text of this article appear below.

EDITOR—We agree with Lemon and Dunnett that better methodsof surveying the literature on animal experiments are needed,¹but we do not share their confidence in the utility of non-systematicreviews.

In clinical trials, systematic review and meta-analysis havemade important contributions to our understanding of sourcesof bias, and the quality of clinical trials has improved asa result. We believe that the same approach can be used to increaseour understanding of sources of bias in animal experiments,again leading to improvements in study quality.

Not to publish negative results from clinical trials is widelyaccepted as unethical, because this may lead to the study beingunwittingly repeated by other investigators, exposing trialparticipants to risk of drug side effects with no prospect ofbenefit. We consider that non-publication of data from animalstudies is equally unethical.

It is argued that tight control of experimental conditions . . . [Full text of this article]

Malcolm R Macleod, specialist registrar, neurology

Western General Hospital, Edinburgh EH4 2XU malcolm@apoptosis.freeserve.co.uk

Shah Ebrahim, professor of epidemiology of ageing

Department of Social Medicine, University of Bristol, Bristol BS8 2PR

Ian Roberts, professor

Department of Epidemiology and Population Health, London School of Hygiene and Tropical Medicine, London WC1E 7HT