HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS REGISTER		Author Keyword(s) Vol Page [Advanced]

This Article Full Text Full Text (PDF) web only appendix Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Add article to my folders Download to citation manager Request Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Hilden, J. Search for Related Content PubMed PubMed Citation Articles by Hilden, J. Topic Collections EBM Notebook

How should clinicians interpret results reflecting the effect of an intervention on composite end points?

University of Copenhagen, Department of Biostatistics
Copenhagen, Denmark

Key Words: diabetes (gestational) • hypoglycaemic agents • insulin • pregnancy outcome • statistics

The first 150 words of the full text of this article appear below.

Evidence-Based Medicine recently published a thoughtful commentary by Montori et al,¹ addressing the above question, prompted by the Australian Carbohydrate Intolerance Study in Pregnant Women (ACHOIS),² which was abstracted in the same issue.³ The ACHOIS investigators compared a screen-and-treat programme for gestational diabetes with routine pregnancy care and used as their main end point a composite of 4 outcomes: shoulder dystocia, Duchenne-Erb palsy, fracture, and death—a natural choice given that a main purpose of gestational glycaemic control is to prevent the baby from being born too big. The topic is important and several points are worth noting.

Montori et al listed 3 questions to help decide the appropriateness of a composite outcome:

(a) Are the component outcomes of similar importance to the patients?

(b) Do the more and less important outcomes occur with similar frequency?

(c) Are the component outcomes likely to have similar relative reductions of the . . . [Full text of this article]