Peer led programme for asthma education in adolescents

doi:10.1136/bmj.323.7304.110

BMJ 2001;323:110 ( 14 July )

Letters

Peer led programme for asthma education in adolescents

Issues of design and analysis are crucial in cluster randomised trials

Papers describing cluster randomised trials must be peer reviewed by statisticians

Authors' reply

Issues of design and analysis are crucial in cluster randomised trials

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

EDITOR $---$ We have concerns about the design and analysis of Shah et al's cluster randomised trial of a peer led education programmefor asthma.¹ Neither the printed nor the (longer) electronicversion mentioned how clustering was accounted for in the trialdesign. The sample size was not justified $---$ neither the number ofclusters (six) nor numbers of children in them. This may seemunimportant since confidence intervals were provided for the comparisonsbetween arms, but the omission iscrucial.

The authors did not specify the magnitude of differences considered in advance as clinically important. The small intraclustercorrelations observed could just be fortuitous. With so few clusters,any estimate of between-cluster variance (and hence intraclustercorrelation) will be extremely imprecise. Without proper detailsof trial design, the danger of publication bias remains, wherea study with low power is more likely to be published when significanceis attained. The . . . [Full text of this article]

CiteULike

Complore

Connotea

Del.icio.us Add to Digg

Digg

StumbleUpon Add to Technorati

Technorati What's this?

Relevant Article

Effect of peer led programme for asthma education in adolescents: cluster randomised controlled trial: Smita Shah, Jennifer K Peat, Evalynn J Mazurski, Han Wang, Doungkamol Sindhusake, Colleen Bruce, Richard L Henry, and Peter G Gibson
BMJ 2001 322: 583. [Abstract] [Full Text] [PDF]