Results 1 to 4 of 4

Thread: [Evidence-Based Medicine] What is Number Needed to Treat and Number Needed to Harm

  1. #1

    Thumbs up [Evidence-Based Medicine] What is Number Needed to Treat and Number Needed to Harm

    TABLE 6-6 Number Needed to Treat and Number Needed to Harm
    In this example, the clinical question is whether or not the addition of clopidogrel to the regimen of a 65-year-old man with unstable angina who is already taking aspirin would prevent death or coronary event. A search of published trials and presented papers at scientific meetings uncovered only one relevant study (N Engl J Med 2001;345(7):494–502)
    In the trial:
    • 12,562 subjects with coronary syndrome were randomized to aspirin alone or aspirin plus clopidogrel.
    • On average, patients were followed for 9 months.
    • The primary end point was to prevent cardiovascular (CV) death, myocardial infarction (MI), or stroke.
    To calculate the number needed to treat (NNT), first calculate the absolute risk reduction (ARR). This is the absolute difference between the event rate in the control group (CER) minus the event rate in the experimental group (EER). The NNT is the inverse of the ARR.
    The trial reports that 11.47% of the aspirin-alone group (control group) had MI, stroke, or CV death. In contrast, 9.28% of the aspirin-plus-clopidogrel group (experimental group) had these events.
    Control Event Rate (Aspirin-Alone Group) Experimental Event Rate (Aspirin-Plus-Clopidogrel) RRR = (CER – EER)/CER ARR = (CER – EER) NNT = 1/ARR
    11.47% 9.28% 19% 2.19% 46
    Thus the NNT is 46. That is, treating 46 patients with unstable angina for 9 months with aspirin plus clopidogrel should prevent MI, stroke, or CV death in 1 patient. To balance risks against benefits of an intervention, we can generate a similar number needed to harm to express the risks associated to the intervention.
    The trial reports that 2.7% of the aspirin-alone group had major nonfatal bleeding events compared with 3.6% of subjects in the intervention group (aspirin plus clopidogrel).
    To calculate the number needed to harm (NNH), first calculate the absolute risk increase (ARI). This is the absolute difference between the event rate in the experimental group (EER) minus the event rate in the control group (CER). The NNH is the inverse of the ARI.
    Control Event Rate Experimental Event Rate ARI (Absolute Risk Increase) NNH
    2.7% 3.6% 0.9% 111
    The NNH is thus 111, meaning that treating 111 patients with both drugs for 9 months would result in one major nonfatal bleed. Combining the NNT and NNH and projecting the results to 1,000 patients would lead to this conclusion: This randomized, controlled trial suggests that treating 1,000 individuals with unstable angina with the combination of aspirin plus clopidogrel would prevent 21 patients from having a stroke, MI, or CV death at the cost of 9 major nonfatal bleeding events.
    Last edited by admin; Tue 20th May '14 at 11:18pm.
    Clinical Pharmacy Specialist - Hematology

  2. #2


    The relative risk reduction (RRR), as a measure of the magnitude of an intervention's effect, can be misleading. It does not discriminate between large and trivial absolute differences between the control and experimental groups. For example, an intervention may result in a 50% risk reduction for the adverse outcome, and this amount of decrease would sound impressive to most clinicians and patients. However, it might represent only a small difference in the risk of a rare event (e.g., 0.2% of patients in a placebo group died compared with 0.1% of patients on active drug). In contrast, a 50% risk reduction might reflect a much more meaningful difference, for instance, when 50% of placebo group died versus 25% of patients in the intervention group (an absolute difference of 25%). The RRR is the same for both examples, but the magnitude of the impact of the intervention is drastically different. The information provided by the RRR is incomplete because it does not take into account the baseline risk of subjects in the trial.
    Clinical Pharmacy Specialist - Hematology

  3. #3


    For example, in trauma patients who require massive transfusion (defined as transfusion of >= 10 units of RBCs), plasma infusion reduced the death rate by approximately 60% and also reduced the risk of multiorgan failure by approximately 60% in comparison with control.

    In the statement above, I think, if the both 60% reduction in mortality and multiorgan failure rate, respectively, are relative risk reduction (RRR) or absolute risk reduction (ARR), is uncertain sine I didn't see the acutal rate of death and multiorgan failure.

    However, the source article concluded that four patients needed to be treat with higher plasma:RBC ratio to prevent one death. That is to say to treat 4 patients with trauma requiring massive transfusion with strategy of a higher plasma:RBC ratio plasma transfusion would prevent death in 1 patient.


    1.Murad MH1, Stubbs JR, Gandhi MJ, Wang AT, Paul A, Erwin PJ, Montori VM, Roback JD. The effect of plasma transfusion on morbidity and mortality: a systematic review and meta-analysis. Transfusion. 2010 Jun;50(6):1370-83.
    Last edited by CheneyHsiung; Tue 20th May '14 at 1:10am.
    Clinical Pharmacy Specialist - Hematology

  4. #4
    PharmD Year 1 TomHsiung's Avatar
    Join Date
    Jan 2013
    Chengdu, Sichuan, China


    In that article, the results of several clinical trials of massive transfusion are below.

    Now let's calculate the ARR of intervention groups compared with control groups.

    The mortality of experiment vs control group:

    Borgman, 2007 19.14% vs 45.24% ARR 26.1% NNT 3.83 RRR 57.69%

    Cotton, 2009 43.20% vs 62.41% ARR 19.21% NNT 5.21 RRR 30.78%

    Holcomb, 2008 34.52% vs 47.66% ARR 13.14% NNT 7.61 RRR 27.57%

    Maegele, 2008 33.19% vs 45.87% ARR 12.68% NNT 7.89 RRR 27.64%

    Kashuk, 2008 38.98% vs 59.46% ARR 20.48% NNT 4.88 RRR 34.44%

    Teixeira, 2009 26.66% vs 65.61% ARR 38.95% NNT 2.57 RRR 59.37%

    Duchesne, 2008 26.76% vs 87.50% ARR 60.74% NNT 1.65 RRR 69.42%

    Dente, 2009 14.00% vs 57.14% ARR 43.14% NNT 2.32 RRR 75.50%
    Last edited by TomHsiung; Tue 20th May '14 at 2:04pm.
    B.S. Pharm, West China School of Pharmacy, Class of 2007, Health System Pharmacist, RPh. Hematology, Infectious Disease.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts