Download citation file:
From Harvard School of Public Health, Boston, Massachusetts, and University Medical Center Utrecht, Utrecht, the Netherlands.
Acknowledgment: The authors thank James Ware for his helpful comments.
Grant Support: By grant R01 ES017876 from the National Institutes of Health and by the Escher Project (T6-202).
Potential Conflicts of Interest: Dr. Knol: Grants received (money to institution): Top Institute Pharma. Disclosures can also be viewed at www.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNum=M10-2468.
Requests for Single Reprints: Tyler J. VanderWeele, PhD, Department of Epidemiology, Harvard School of Public Health, 677 Huntington Avenue, Boston, MA 02115; e-mail, email@example.com.
Current Author Addresses: Dr. VanderWeele: Department of Epidemiology, Harvard School of Public Health, 677 Huntington Avenue, Boston, MA 02115.
Dr. Knol: Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Huispost Str. 6.131Postbus 85500, 3508 GA Utrecht, the Netherlands.
Author Contributions: Conception and design: T.J. VanderWeele, M.J. Knol.
Analysis and interpretation of the data: T.J. VanderWeele.
Drafting of the article: T.J. VanderWeele.
Critical revision of the article for important intellectual content: T.J. VanderWeele, M.J. Knol.
Final approval of the article: M.J. Knol.
Statistical expertise: T.J. VanderWeele.
Obtaining of funding: T.J. VanderWeele.
In randomized trials with subgroup analyses, the primary treatment or intervention of interest is randomized, but the secondary factors defining subgroups are not. This article clarifies when confounding is an issue in subgroup analyses. If investigators are interested simply in targeting subpopulations for intervention, control for confounding is not needed. If investigators are interested in intervening on the secondary factors that define the subgroups to increase the treatment effect or in attributing the subgroup differences to the secondary factors themselves, then confounding is relevant and must be controlled for. The authors demonstrate this point by using examples from published randomized trials.
The estimates of treatment effect in male and female participants are valid, but these differences in treatment effect might be caused by age rather than sex.
Please read the other comments before posting. Contributors must reveal any conflict
Comments are moderated and will appear on the site at the discretion of The American
College of Physicians editorial staff. Please be sure your email address is
updated in your account, otherwise the American College of Physicians will not be
able to contact you about your comment.
Anyone can submit a comment any time after publication, but only those submitted within 4 weeks of an article’s publication will be considered for print publication. One month after publication, editors review all posted comments and select some for publication in the Letters section of the print version of Annals. (Not peer reviewed)
Authors: No more than 5
Text: Word Limit 400 (excludes references), 5 references, no figures or tables
* = Required Field
Disclosure of Any Conflicts of Interest*
(applies to the past 5 years and foreseeable future) Indicate any potential conflicts
of interest of each author below, including specific financial interests and relationships
and affiliations relevant to the subject matter or materials discussed in the manuscript
(eg, employment/affiliation, grants or funding, consultancies, honoraria, speakers
bureau, stock ownership or options, expert testimony, royalties, donation of medical
equipment, or patents filed, received, or pending). If all authors have none, check
"No potential conflicts or relevant financial interests" in the box below. Please
also indicate any funding received in support of this work. The information will
be posted with your response.
VanderWeele and Knol's attempts to clarify subgroup analyses for randomized trials deserves a strong warning(1).
Subgroup analyses should be viewed with caution and a particular attention must be warranted in industry funded trials when the primary outcome is not statistically significant(2).
Their demonstration using examples from published randomized trials have forgotten Sleigh's post hoc analysis of the ISIS- 2 trial which showed that aspirin therapy was significantly beneficial for all patients except those born under the astrological signs of Gemini or Libra(3).
1 Vanderweele TJ, Knol MJ. Interpretation of subgroup analyses in randomized trials: heterogeneity versus secondary interventions. Ann Intern Med 2011;154:680-3.
2 Sun X, Briel M, Busse JW et al. The influence of study characteristics on reporting of subgroup analyses in randomised controlled trials: systematic review. BMJ 2011 28;342:d1569
3 Sleight P. Subgroup analyses in clinical trials: fun to look at - but don't believe them! Curr Control Trial CardiovascMed 2000;1:125-127.
The In the Clinic® slide sets are owned and copyrighted by the American College
of Physicians (ACP). All text, graphics, trademarks, and other intellectual property
incorporated into the slide sets remain the sole and exclusive property of the ACP.
The slide sets may be used only by the person who downloads or purchases them and
only for the purpose of presenting them during not-for-profit educational activities.
Users may incorporate the entire slide set or selected individual slides into their
own teaching presentations but may not alter the content of the slides in any way
or remove the ACP copyright notice. Users may make print copies for use as hand-outs
for the audience the user is personally addressing but may not otherwise reproduce
or distribute the slides by any means or media, including but not limited to sending
them as e-mail attachments, posting them on Internet or Intranet sites, publishing
them in meeting proceedings, or making them available for sale or distribution in
any unauthorized form, without the express written permission of the ACP. Unauthorized
use of the In the Clinic slide sets will constitute copyright infringement.
to gain full access to the content and tools.
Learn more about subscription options