Endgames Statistical question

Risk

BMJ 2008; 337 doi: https://doi.org/10.1136/bmj.a2293 (Published 30 October 2008) Cite this as: BMJ 2008;337:a2293
  1. John Fletcher, clinical epidemiologist
  1. 1BMJ, London WC1H 9JR
  1. jfletcher{at}bmj.com

    A group of investigators decided to test the hypothesis that use of postmenopausal hormone replacement therapy increases the risk of breast cancer. They identified 500 women with newly diagnosed breast cancer, and 500 women of the same age, from the same neighbourhoods as the first group of women, who did not have breast cancer. Among the women with breast cancer, 250 reported having used postmenopausal hormones. Among those without breast cancer, 175 reported use of such hormones.

    Which of the following statements is or are true?

    a) The excess risk of breast cancer in women taking HRT is 15%—that is, (250-175)/500

    b) The risk of breast cancer in this group of women was 50%—that is, 500/(500+500)

    c) The relative risk of breast cancer in women taking HRT was 1.43—that is (250/500)/(175/500)

    d) The odds ratio for breast cancer in women taking HRT was 1.86—that is (250/250)/(175/325)

    Answer

    d—In a case control study incidence and absolute risks, including differences in absolute risk, cannot be calculated because cases and controls are selected by the researcher. Odds and odds ratios are used to express associations in case control studies. In carefully designed case control studies the odds ratio is the most accurate estimate of relative risk. For this to apply requires either rare outcomes or incidence based recruitment of cases and controls.

    Notes

    Cite this as: BMJ 2008;337:a2293

    View Abstract

    Sign in

    Log in through your institution

    Subscribe