HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Permissions Request Permissions Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Google Scholar Articles by Puhan, M. A. Articles by Schünemann, H. J. PubMed PubMed Citation Articles by Puhan, M. A. Articles by Schünemann, H. J.

How to increase the value of randomised trials in COPD research

¹ Dept of Epidemiology, Johns Hopkins School of Public Health, Baltimore, MD, USA. ² Horten Centre, University of Zurich, Zurich, Switzerland. ³ Clarity Research Group, Dept of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, ON, Canada. ⁴ Dept of Epidemiology, Italian National Cancer Institute "Regina Elena", Rome, Italy.

CORRESPONDENCE: H. J. Schünemann, Dept of Clinical Epidemiology and Biostatistics, Faculty of Health Sciences, McMaster University, 1200 Main Street W, Hamilton, ON, Canada. E-mail: schuneh{at}mcmaster.ca

Keywords: Chronic obstructive pulmonary disease, evidence-based medicine, study design

Received: June 24, 2008
Accepted March 23, 2009

Methodological criteria that increase the validity of randomised trials are often not considered in respiratory research, even in large chronic obstructive pulmonary disease (COPD) trials. We describe four important aspects in the design, analysis and reporting of randomised trials, selected based on their relevance to current COPD research and based on our judgments of importance for researchers and users of the literature.

First, to optimally control for confounding, where confounding refers to a factor that is associated with an exposure or intervention and influences the outcome, a clear definition of the main relationship between treatment and the primary outcome as well as identification of measurable confounders is required. In addition to randomisation per se as the key method to protect against confounding, restriction (excluding patients with specific characteristics that may introduce confounding), stratification (separate randomisation of patients with specific characteristics) and statistical adjustment are means to be considered to optimally control for confounding that simple randomisation may not achieve.

Secondly, the selection of the primary outcome should be guided by the importance to patients. Secondary outcomes provide hypotheses about the effects observed for the primary outcome and can provide important data for systematic reviews and meta-analyses, but should be interpreted with caution in single trials.

Thirdly, in study power calculations, not only the actual sample size, but the number of events, has a large influence on the power of the study and, often, unrealistic assumptions about event rates are made to increase the feasibility of trials.

Finally, essential steps to transfer results from research to practice include complete reporting of trials and developing tools, such as decision aids, to support patients and physicians in their shared decision making.