Clinical trials in acute severe asthma: are type II errors important?

Abstract

Many studies in acute severe asthma have had low power. Fifteen randomised double blind studies investigating the treatment of acute severe asthma published during 1974-84 were analysed for power and 95% confidence limits. Twelve studies failed to detect a significant difference in bronchodilatation produced by two treatments and reported the treatments to produce equal effect. Each study had, however, less than a 60% probability of detecting a true 25% difference in bronchodilatation.