The trial data were analysed on an intention to treat basis. The primary health outcomes assessed by the trial were normalised forced expiratory lung function in one second (FEV1) and improvement in quality of life (QOL) as measured by the St George's Respiratory Questionnaire (SGRQ).
FEV1 values, expressed in litres/second, were measured as follows:
baseline FEV1 plus the difference of treatment group contrasts relative to placebo of normalised area under the curve-FEV1 following the morning dose of study medications on the last day of treatment (12 weeks);
formoterol arms were further reduced by 0.04 and 0.02 litres, respectively (the amount by which the FEV1 measures for these arms exceeded the baseline FEV1 measures for placebo and ipratropium bromide).
Differences in QOL scores were measured between baseline and end-of-study SGRQ total scores. The baseline value for the ipratropium bromide arm was further adjusted by the amount it exceeded the other arms: 1.6 points.
Other efficacy variables were assessed in the trial but were not fully reported in this paper.
No details of the baseline characteristics of the patients included in the trial were reported, and no comment was made as to whether the groups were comparable.