Letters

Orlistat associated with hypertension

BMJ 2001; 322 doi: https://doi.org/10.1136/bmj.322.7278.110 (Published 13 January 2001) Cite this as: BMJ 2001;322:110

Digit preference lays conclusions about orlistat open to doubt

  1. G D Johnston, professor of clinical pharmacology
  1. Department of Therapeutics and Pharmacology, Queen's University, Belfast, Belfast BT9 7BL
  2. Pharmaceuticals Division Clinical Science, Safety Risk Management, F Hoffman-La Roche, CH-4070 Basle, Switzerland
  3. Regional Centre for Pharmacovigilance, Karolinska University Hospital, S-171 76, Stockholm, Sweden
  4. Pharmacovigilance Unit, Medical Products Agency, S-751 03, Uppsala, Sweden

    EDITOR—The drug point by Persson et al provides an excellent example of the common and widely criticised practice of digit preference when recording blood pressure.1 The British Hypertension Society guidelines recommend measuring blood pressure to the nearest 2 mm Hg.2 Persson et al did not adopt this method of measurement because the chances of recording 12 zeros are several million to one.

    Bias of this kind could have a profound effect on the study's conclusions. For example, Persson et al concluded that 170/100 mm Hg (when taking orlistat) was greater than 160/90 mm Hg (when not taking the drug). If a blood pressure of 166/96 mm Hg was rounded up to 170/100 mm Hg and 164/94 mm Hg was rounded down to 160/90 mm Hg, then the true difference would be 2/2 mm Hg rather than the 10/10 mm Hg as recorded by the observer. Given the open nature of the investigation, the considerable day to day variation that can occur in measuring blood pressure, and the strong digit preference observed in this study, the conclusions must …

    View Full Text

    Sign in

    Log in through your institution

    Free trial

    Register for a free trial to thebmj.com to receive unlimited access to all content on thebmj.com for 14 days.
    Sign up for a free trial

    Subscribe