The first article in the series for learners of evidence-based medicine discusses the concept of number needed to treat (NNT).1 However, studies involving patients who need long-term follow-up, such as those with cancer or chronic cardiac conditions, commonly use time-to-event or survival analysis. It is important to realize that the NNT is not calculated in the same way for these studies.
NNT from survival analysis data should be estimated by the hazard ratio2 and is not based on the difference in event rates between treatment groups at the end of follow-up.3
Mario L. de Lemos British Columbia Cancer Agency Vancouver, BC
References
- 1.Barratt A, Wyer PC, Hatala R, McGinn T, Dans AL, Keitz S, et al. Tips for learners of evidence-based medicine: 1. Relative risk reduction, absolute risk reduction and number needed to treat. CMAJ 2004;171(4):353-8. [DOI] [PMC free article] [PubMed]
- 2.Altman DG, Andersen PK. Calculating the number needed to treat for trials where the outcome is time to an event. BMJ 1999;319 (7223): 1492-5. [DOI] [PMC free article] [PubMed]
- 3.de Lemos ML. How to survive the survival plots [letter]. Lancet 2002;360(9337):954. [DOI] [PubMed]