Methods to calculate relative risks, risk differences, and numbers needed to treat from logistic regression

  • Ralf Bender
    Corresponding author. Tel.: +49-221-35685-451; fax: +49-221-35685-891.
    Department of Medical Biometry, Institute for Quality and Efficiency in Health Care (IQWiG), Dillenburger Str. 27, D-51105 Cologne, Germany
    Search for articles by this author
  • Oliver Kuss
    Institute for Medical Epidemiology, Biostatistics, and Informatics (IMEBI), Martin-Luther-University Halle-Wittenberg, Halle (Saale), Germany
    Search for articles by this author
Published:September 18, 2009DOI:
      Our commentary underlines important issues that should be considered when clinically relevant effect measures are estimated from logistic regression by means of the methods recently proposed by Peter Austin.
      • The research question has to be taken into account.
      • The study design has to be taken into account.
      • The performance of point and interval estimates has to be investigated.
      • A part of the work is already published.
      To read this article in full you will need to make a payment

      Purchase one-time access:

      Academic & Personal: 24 hour online accessCorporate R&D Professionals: 24 hour online access
      One-time access price info
      • For academic or personal research use, select 'Academic and Personal'
      • For corporate R&D use, select 'Corporate R&D Professionals'


      Subscribe to Journal of Clinical Epidemiology
      Already a print subscriber? Claim online access
      Already an online subscriber? Sign in
      Institutional Access: Sign in to ScienceDirect


        • Austin P.C.
        Absolute risk reductions, relative risks, relative risk reductions, and numbers needed to treat can be obtained from a logistic regression model.
        J Clin Epidemiol. 2010; 63: 47-56
        • Localio A.R.
        • Margolis D.J.
        • Berlin J.A.
        Relative risks and confidence intervals were easily computed indirectly from multivariable logistic regression.
        J Clin Epidemiol. 2007; 60: 874-882
        • Bender R.
        • Blettner M.
        Calculating the “number needed to be exposed” with adjustment for confounding variables in epidemiological studies.
        J Clin Epidemiol. 2002; 55: 525-530
        • Bender R.
        • Kuss O.
        Confidence intervals for adjusted NNEs: a simulation study.
        J Clin Epidemiol. 2003; 56 ([Letter]): 205-206
        • Bender R.
        • Kuss O.
        • Hildebrandt M.
        • Gehrmann U.
        Estimating adjusted NNT measures in logistic regression analysis.
        Stat Med. 2007; 26: 5586-5595
        • Bender R.
        • Hildebrandt M.
        • Kuss O.
        Die Berechnung adjustierter NNEs in Kohortenstudien.
        in: Kaiser K. Bödeker R.-H. Statistik und Datenanalyse mit SAS®−Proceedings der 10. Konferenz der SAS®-Anwender in Forschung und Entwicklung (KSFE). Shaker, Aachen2006: 19-33
      1. Laubender RP, Bender R. Estimating adjusted risk difference (RD) and number needed to treat (NNT) measures in the Cox regression model. Stat Med; under review.

        • Hauck W.W.
        • Neuhaus J.M.
        • Kalbfleisch J.D.
        • Anderson S.
        A consequence of omitted covariates when estimating odds ratios.
        J Clin Epidemiol. 1991; 44: 77-81
        • Hauck W.W.
        • Anderson S.
        • Marcus S.M.
        Should we adjust for covariates in nonlinear regression analyses of randomized trials?.
        Control Clin Trials. 1998; 19: 249-256
        • Heller R.F.
        • Dobson A.J.
        • Attia J.
        • Page J.
        Impact numbers: Measures of risk factor impact on the whole population from case-control and cohort studies.
        J Epidemiol Community Health. 2002; 56: 606-610
      2. Gehrmann U, Kuss O, Wellmann J, Bender R. Logistic regression was preferred to estimate risk differences and numbers needed to be exposed adjusted for covariates. J Clin Epidemiol; under review.