This Article Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Aaron, S. D. Search for Related Content PubMed PubMed Citation Articles by Aaron, S. D.

Avoiding Mistakes in Calculating the Number Needed to Treat in Severe COPD

Dr. Kardos and colleagues have completed an impressive clinical trial demonstrating that combination therapy with salmeterol/fluticasone reduces the frequency of moderate/severe COPD exacerbations compared with treatment with salmeterol (1). However, there is a small mistake in the article when they report the number of patients needed to treat. The number needed to treat represents the estimated number of patients who need to be treated with the new treatment rather than the standard treatment for one additional patient to benefit (2). It is defined as the reciprocal of the absolute risk reduction. It can only be obtained for trials that report a binary outcome; multiple events occurring in the same patient are not included in the number needed to treat (3).

As described in Kardos and coworkers' trial, 241 of 487 patients treated with salmeterol (49.5%) and 210 of 507 patients treated with salmeterol/fluticasone (41.4%) had at least one exacerbation over the 44-week trial. Therefore, the absolute risk reduction = 49.5% – 41.4% = 8.1%, and the number of patients needed to treat with salmeterol/fluticasone (rather than salmeterol) to prevent one additional patient from experiencing an exacerbation in 44 wk = 1/0.081 = 12.3. In contrast, Dr. Kardos reports a number needed to treat of 2.08.

The mistake is that the authors have calculated the number needed to treat from the difference in mean rates of exacerbations per patient-year. These data are not binary, since patients may have more than one exacerbation per year, and be counted numerous times. The number needed to treat can only be calculated from the percentage of patients who have had one or more exacerbations during the trial period.

Shawn D. Aaron

University of Ottawa, Ottawa, Ontario, Canada

FOOTNOTES

Conflict of Interest Statement: S.D.A. sat on Canadian Advisory Boards for Boehringer Ingelheim and Astra Zeneca and was paid $3,000 in total for these activities; he helped prepare teaching slides for Boehringer-Ingelheim Canada and was paid $5,000.

REFERENCES

1. Kardos P, Wencker M, Glaab T, Vogelmeier C. Impact of salmeterol/fluticasone propionate versus salmeterol on exacerbations in severe chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2007;175:144–149.[Abstract/Free Full Text]
2. Cook RJ, Sackett DL. The number needed to treat: a clinically useful measure of treatment effect. Br Med J 1995;310:452–454.[Free Full Text]
3. Altman DG. Confidence intervals for the number needed to treat. Br Med J 1998;317:1309–1312.[Free Full Text]

This Article Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Aaron, S. D. Search for Related Content PubMed PubMed Citation Articles by Aaron, S. D.