Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Correspondence
  • Published:

Inappropriate reporting and interpretation of subgroups in the AML-BFM 93 study

Leukemia volume 15, pages 1803–1804 (2001)Cite this article

The misinterpretation of subgroup analyses is one of the most common mistakes in the reporting of clinical trials.1 Creutzig et al2 fall into this trap in their recent paper on the AML-BFM 93 trial (Leukemia (2001) 348–354). They report that blast cell clearance by day 15 is significantly better for patients treated with idarubicin than with daunorubicin (P = 0.01), but go on to claim that the ‘difference was restricted to … high risk patients (P = 0.007)’. This conclusion is based on the inappropriate interpretation of P-values within individual subgroups.

Two important factors need to be considered when interpreting subgroup analyses. First, the smaller numbers in each subgroup lead to wider confidence intervals. Especially when, as here, the number of events in one of the groups is much smaller than the other, false negative results may arise: there is a benefit, but the sample size is too small to demonstrate it with statistical significance. Second, when the overall sample is broken down into smaller subgroups, the play of chance means that it is unlikely that the observed treatment effect in the two groups will be the same, even if there is no true difference in effect between them. Thus, in this example, one would expect the point estimate of treatment effect to be greater than the overall estimate in one group and smaller in the other.

This is a preview of subscription content, access via your institution

Access options

Buy this article

  • Purchase on Springer Link
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

Figure 1

References

  1. Collins R, McMahon S . Reliable assessment of the effects of treatment on mortality and major morbidity, I: clinical trials Lancet 2001 357: 373–380

    Article  CAS  Google Scholar 

  2. Creutzig U, Ritter J, Zimmermann M, Hermann J, Gadner H, Blütters Sawatzki D, Niemeyer CM, Schwabe D, Selle B, Boos J, Kühl J, Feldges A for the AML-BFM Study Group . Idarubicin improves blast cell clearance during induction therapy in children with AML: results of study AML-BFM 93 Leukemia 2001 15: 348–354

    Article  CAS  Google Scholar 

  3. Assmann SF, Pocock SJ, Enos LE, Kasten LE . Subgroup analysis and other (mis)uses of baseline data Lancet 2000 355: 1064–1069

    Article  CAS  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. University of Birmingham Clinical Trials Unit, Birmingham, UK

    K Wheatley & RK Hills

Authors
  1. K Wheatley
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. RK Hills
    View author publications

    You can also search for this author in PubMed Google Scholar

Rights and permissions

Reprints and permissions

About this article

Cite this article

Wheatley, K., Hills, R. Inappropriate reporting and interpretation of subgroups in the AML-BFM 93 study. Leukemia 15, 1803–1804 (2001). https://doi.org/10.1038/sj.leu.2402274

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/sj.leu.2402274

Search

Advanced search

Quick links