Advertisement

Common problems related to the use of number needed to treat

  • Andreas Stang
    Correspondence
    Corresponding author. Institute of Clinical Epidemiology, Medical Faculty, Martin-Luther-University of Halle-Wittenberg, Magdeburger Strasse 8, 06097 Halle, Germany. Tel.: +49-345-557-3596; fax: +49-345-557-3565.
    Affiliations
    Institute of Clinical Epidemiology, Medical Faculty, Martin-Luther-University of Halle-Wittenberg, Magdeburger Strasse 8, 06097 Halle, Germany
    Search for articles by this author
  • Charles Poole
    Affiliations
    Department of Epidemiology, University of North Carolina, Chapel Hill, NC, USA
    Search for articles by this author
  • Ralf Bender
    Affiliations
    Institute for Quality and Efficiency in Health Care, Department of Medical Biometry, Cologne, Germany

    Faculty of Medicine, University of Cologne, Cologne, Germany
    Search for articles by this author
Published:November 02, 2009DOI:https://doi.org/10.1016/j.jclinepi.2009.08.006

      Abstract

      Objective

      To illustrate basic issues that have to be taken into account when study results are presented by means of the number needed to treat (NNT).

      Study Design and Setting

      This article presents an overview of common problems related to the NNT with corresponding explanations.

      Results

      Without stating the direction of the effect, the alternative treatment, the treatment period, and the follow-up period, information in terms of NNTs is uninterpretable. The naive use of person-time data for the calculation of NNTs is frequently inappropriate. Rounding NNTs to the next upward integer may obscure differences among therapies.

      Conclusions

      The basic information about which treatments are compared, the treatment period, the follow-up period, and the direction of the effect should be given when study results are presented in terms of NNTs. Adequate methods should be used for point and interval estimation of NNTs. Unnecessary rounding of NNTs should be avoided. In more complicated situations of confounding or varying follow-up times, the use of more sophisticated methods is required with increasing potential for misinterpretation.

      Keywords

      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:

      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

      References

        • McAlister F.A.
        The “number needed to treat” turns 20—and continues to be used and misused.
        CMAJ. 2008; 179: 549-553
        • Laupacis A.
        • Sackett D.L.
        • Roberts R.S.
        An assessment of clinically useful measures of the consequences of treatment.
        N Engl J Med. 1988; 318: 1728-1733
        • Halvorsen P.A.
        • Kristiansen I.S.
        Decisions on drug therapies by numbers needed to treat: a randomized trial.
        Arch Intern Med. 2005; 165: 1140-1146
        • Hildebrandt M.
        • Vervolgyi E.
        • Bender R.
        Calculation of NNTs in RCTs with time-to-event outcomes: a literature review.
        BMC Med Res Methodol. 2009; 9: 21
        • Kristiansen I.S.
        • Gyrd-Hansen D.
        • Nexoe J.
        • Nielsen J.B.
        Number needed to treat: easily understood and intuitively meaningful? Theoretical considerations and a randomized trial.
        J Clin Epidemiol. 2002; 55: 888-892
        • Misselbrook D.
        • Armstrong D.
        Patients' responses to risk information about the benefits of treating hypertension.
        Br J Gen Pract. 2001; 51: 276-279
        • Halvorsen P.A.
        • Kristiansen I.S.
        • Aasland O.G.
        • Forde O.H.
        Medical doctors' perception of the “number needed to treat” (NNT). A survey of doctors' recommendations for two therapies with different NNT.
        Scand J Prim Health Care. 2003; 21: 162-166
        • Nexoe J.
        • Gyrd-Hansen D.
        • Kragstrup J.
        • Kristiansen I.S.
        • Nielsen J.B.
        Danish GPs' perception of disease risk and benefit of prevention.
        Fam Pract. 2002; 19: 3-6
        • Fahey T.
        • Griffiths S.
        • Peters T.J.
        Evidence based purchasing: understanding results of clinical trials and systematic reviews.
        BMJ. 1995; 311: 1056-1059
        • Sinclair J.C.
        • Bracken M.B.
        Clinically useful measures of effect in binary analyses of randomized trials.
        J Clin Epidemiol. 1994; 47: 881-889
        • Cook R.J.
        • Sackett D.L.
        The number needed to treat: a clinically useful measure of treatment effect.
        BMJ. 1995; 310: 452-454
        • Sackett D.L.
        On some clinically useful measures of the effects of treatment.
        Evid Based Med. 1996; 1: 37-38
        • Nuovo J.
        • Melnikow J.
        • Chang D.
        Reporting number needed to treat and absolute risk reduction in randomized controlled trials.
        JAMA. 2002; 287: 2813-2814
        • Altman D.G.
        • Schulz K.F.
        • Moher D.
        • Egger M.
        • Davidoff F.
        • Elbourne D.
        • et al.
        The revised CONSORT statement for reporting randomized trials: explanation and elaboration.
        Ann Intern Med. 2001; 134: 663-694
        • Smeeth L.
        • Haines A.
        • Ebrahim S.
        Numbers needed to treat derived from meta-analyses—sometimes informative, usually misleading.
        BMJ. 1999; 318: 1548-1551
        • Wu L.A.
        • Kottke T.E.
        Number needed to treat: caveat emptor.
        J Clin Epidemiol. 2001; 54: 111-116
        • Duncan B.W.
        • Olkin I.
        Bias of estimates of the number needed to treat.
        Stat Med. 2005; 24: 1837-1848
        • Hutton J.L.
        Number needed to treat: properties and problems.
        J R Stat Soc Ser A Stat Soc. 2000; 163: 403-415
        • Kristiansen I.S.
        • Gyrd-Hansen D.
        • Nexoe J.
        • Nielsen J.B.
        Goodbye NNT.
        J Clin Epidemiol. 2004; 57: 221
        • Walter S.D.
        Choice of effect measure for epidemiological data.
        J Clin Epidemiol. 2000; 53: 931-939
        • Bender R.
        Number needed to treat (NNT).
        in: Armitage P. Colton T. Encyclopedia of biostatistics. Wiley, Chichester, UK2005: 3752-3761
        • Altman D.G.
        • Bland J.M.
        Time to event (survival) data.
        BMJ. 1998; 317: 468-469
        • Altman D.G.
        Confidence intervals for the number needed to treat.
        BMJ. 1998; 317: 1309-1312
        • Hildebrandt M.
        • Bender R.
        • Gehrmann U.
        • Blettner M.
        Calculating confidence intervals for impact numbers.
        BMC Med Res Methodol. 2006; 6: 32
        • Maldonado G.
        • Greenland S.
        Estimating causal effects.
        Int J Epidemiol. 2002; 31: 422-429
        • de Craen A.J.
        • Vickers A.J.
        • Tijssen J.G.
        • Kleijnen J.
        Number-needed-to-treat and placebo-controlled trials.
        Lancet. 1998; 351: 310
        • Freeman E.W.
        • Rickels K.
        • Sammel M.D.
        • Lin H.
        • Sondheimer S.J.
        Time to relapse after short- or long-term treatment of severe premenstrual syndrome with sertraline.
        Arch Gen Psychiatry. 2009; 66: 537-544
        • Maughan T.S.
        • James R.D.
        • Kerr D.J.
        • Ledermann J.A.
        • McArdle C.
        • Seymour M.T.
        • et al.
        Comparison of survival, palliation, and quality of life with three chemotherapy regimens in metastatic colorectal cancer: a multicentre randomised trial.
        Lancet. 2002; 359: 1555-1563
        • De Gramont A.
        • Krulik M.
        • Cady J.
        • Lagadec B.
        • Maisani J.E.
        • Loiseau J.P.
        • et al.
        High-dose leucovorin and 5-fluorouracil bolus and infusion in advanced colorectal cancer.
        Eur J Clin Oncol. 1988; 24: 1499-1503
        • Lokich J.J.
        • Ahlgren J.D.
        • Gullo J.J.
        • Philips J.A.
        • Fryer J.G.
        • et al.
        A prospective randomised comparison of continuous infusion flourouracil with a conventional bolus schedule in metastatic colorectal carcinoma.
        J Clin Oncol. 1989; 7: 425-432
        • 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.
        • Hildebrandt M.
        • Gehrmann U.
        Estimating adjusted NNT measures in logistic regression analysis.
        Stat Med. 2007; 26: 5586-5595
        • Sabatine M.S.
        • Cannon C.P.
        • Gibson C.M.
        • Lopez-Sendon J.L.
        • Montalescot G.
        • Theroux P.
        • et al.
        Effect of clopidogrel pretreatment before percutaneous coronary intervention in patients with ST-elevation myocardial infarction treated with fibrinolytics: the PCI-CLARITY study.
        JAMA. 2005; 294: 1224-1232
        • Kleinbaum D.G.
        • Klein M.
        Logistic regression: a self-learning text.
        2nd edition. Springer, New York2002
        • Lubsen J.
        • Hoes A.
        • Grobbee D.
        Implications of trial results: the potentially misleading notions of number needed to treat and average duration of life gained.
        Lancet. 2000; 356: 1757-1759
        • Mayne T.J.
        • Whalen E.
        • Vu A.
        Annualized was found better than absolute risk reduction in the calculation of number needed to treat in chronic conditions.
        J Clin Epidemiol. 2006; 59: 217-223
        • Altman D.G.
        • Andersen P.K.
        Calculating the number needed to treat for trials where the outcome is time to an event.
        BMJ. 1999; 319: 1492-1495