Assessing asthma in adult clinical trials of inhaled B2-agonists : a search for a standard primary outcome measure
Since the late 1960s the safety of inhaled B2-agonists has been questioned and the longtenn regular use of these drugs has been linked to increasing morbidity and mortality. National and international guidelines recommend that short acting inhaled B2-agonists should only be used on an "as needed" basis and yet the evidence for these recommendations is still unclear, one reason being the lack of common definition for an outcome. The Regular Use of Salbutamol Trial (TRUST) was designed to assess the risks and benefits of regular versus as needed salbutamol in mild to moderate asthma. In order to establish whether a common primary outcome measure could improve the comparability and interpretation of different trials, a systematic rcyiew of randomised controlled trials of long and short acting inhaled B2-agonists in asthmatic subjects was undertaken to identify well designed trials in this field and primary outcome measures used. The systematic review identified five different primary outcome measures from 26 trials of long and short acting inhaled B2-agonists. The TRUST definition of exacerbation was compared with the five primary outcome measures identified using the TRUST diary card data. In addition, the diary card variables (changes in PEF, symptom scores and medication use) were examined to determine the extent to which they predicted exacerbations according to the different definitions. The use of additional corticosteroids and an increase in daytime symptoms of two or more above baseline were the strongest predictors of all four definitions of exacerbation. A fall in morning PEF of 100 lImin was strongly associated with all definitions of exacerbation but was not a sensitive measure. In conclusion, exacerbations of asthma could be identified by use of additional corticosteroids and an increase in two or more of daytime symptoms. The specificity could be improved by including morning PEF but this may reduce patient compliance with study protocol in asthma trials.