Endgames Statistical Question

Control groups

BMJ 2010; 340 doi: https://doi.org/10.1136/bmj.c391 (Published 28 January 2010) Cite this as: BMJ 2010;340:c391
  1. Philip Sedgwick, senior lecturer in medical statistics
  1. 1Centre for Medical and Healthcare Education, St George’s, University of London, Tooting, London SW17 0RE
  1. p.sedgwick{at}sgul.ac.uk

    Researchers undertook a randomised controlled trial to assess the effect of zinc supplementation on the severity and duration of diarrhoea caused by cholera in children.1 Children were recruited if they had watery diarrhoea and dark field examination of stool was positive for Vibrio cholera (confirmed by stool culture). Children were randomly allocated to zinc supplementation or placebo. Zinc supplementation was found to reduce the duration and severity of diarrhoea in children with cholera.

    Which of the following statements, if any, are true?

    • a) The control group comprised children without diarrhoea

    • b) Placebo was the control treatment

    • c) Placebo was a negative control treatment

    • d) Placebo was a positive control treatment


    b and c are both true; a and d are false.

    Clinical trials should always be comparative, so zinc supplementation was compared with a control treatment. The trial was placebo controlled—zinc supplementation was compared with placebo (answer b). Placebos were described in last week’s question.2 The response of the placebo treatment group is quantified as the placebo effect. By comparing zinc supplementation with placebo, it was possible to estimate the therapeutic effects of zinc supplementation and whether supplementation enhances recovery. If zinc supplementation had been compared with a standard treatment, it would have been possible to establish whether supplementation was superior, or at least equivalent, to such treatment.

    The trial used current rather than historical controls—children were allocated to placebo or zinc supplementation concurrently. Historical controls would have provided information about the epidemiology of recovery from diarrhoea caused by cholera. However, natural epidemiology may vary with time. Therefore, any differences between zinc supplementation and historical controls may not have resulted from the effect of supplementation alone.

    A placebo is referred to as a negative control because it is an inert treatment with no known active therapeutic effects (answer c). If the control treatment had been a standard therapeutic regimen then it would have been referred to as a positive control; therefore answer d is false.

    Children were included in the trial if they had diarrhoea caused by cholera; therefore answer a is false. The children were randomly allocated to receive zinc supplementation or placebo, with placebo acting as the control treatment. The group receiving control treatment in a trial should not be confused with control subjects in a case-control study. A case-control study has two groups of people who are selected on the basis of their disease status: one group with the disease (the cases) and one without (the controls). Case-control studies are retrospective—individuals are asked about past exposure to proposed risk factors. The aim is to provide insight into which factors may increase or decrease the risk of disease. Case-control studies have been described in previous questions.3 4


    Cite this as: BMJ 2010;340:c391


    • Competing interests: None declared.


    View Abstract

    Log in

    Log in through your institution


    * For online subscription