• Marcus Müllner, Hugh Matthews, and Douglas G Altman.
• Universitätsklinik für Notfallmedizin, Allgemeines Krankenhaus Wien, Währinger Gürtel 18-20/6D, A-1090 Vienna, Austria. marcus.muellner@univie.ac.at.
• Ann. Intern. Med. 2002 Jan 15; 136 (2): 122-6.

BackgroundThe use of complex statistical models to adjust for confounding is common in medical research.ObjectiveTo determine the frequency and adequacy of adjustment for confounding in medical articles.DesignCross-sectional survey.Setting34 scientific medical journals with a high impact factor.MeasurementsFrequency of reporting on methods used to adjust for confounding in 537 original research articles published in January 1998.ResultsOf the 537 articles, 169 specified that adjustment for confounding was used. In 1 paper in 10, it was unclear which statistical method was used or for which variables adjustment was made. In 45% of papers, it was not clear how multicategory or continuous variables were treated in the analysis. Inadequate reporting was less frequent if an author was affiliated with a department of statistics, epidemiology, or public health and if articles were published in journals with a high impact factor.ConclusionsDetails of methods used to adjust for confounding are frequently not reported in original research articles.

12 keywords...

hide…