The difficulty of assessing nebulizer responses in chronic obstructive pulmonary disease (COPD) has been demonstrated before. This study aims to re-examine both the role of domiciliary nebulizers in COPD and also bronchodilator (BD) assessment in individuals. In a double-blind, randomized, cross-over trial, 19 stable patients with severe COPD were given the following medication 6-hourly for 2-week periods: (1) nebulized salbutamol 2.5 mg with ipratropium 0.5 mg and placebo inhalers (MDI) with spacer; (2) placebo nebules and inhaled salbutamol 400 microg with ipratropium 80 microg via MDI with spacer; (3) inhaled salbutamol 400 microg with ipratropium 80 microg via MDI with spacer (but no placebo nebulized drugs). Both nebulized and MDI drugs produced highly significant improvements in forced expiratory volume in 1 sec (FEV1), forced vital capacity (FVC), specific airways conductance, 6-min walking distance (6MWD) and residual volume. There were no significant differences between BD responses obtained after active nebulized and active MDI BDs. From the diary cards, 2 weeks of active nebulized BDs produced a slightly higher median peak expiratory flow (PEF) than active MDI BDs (236 and 219 l m(-1), respectively, P=0.01) and slightly less extra inhaler use (0.8 and 1.1 puffs, respectively, P<0.05) but no significant difference in dyspnoea or quality of life (QOL) scores. There were significant correlations between domiciliary PEF and acute BD-induced changes in FVC and 6MWD, and also between domiciliary dyspnoea scores and acute changes in both total lung capacity and 6MWD. In conclusion, nebulized medication conferred little clinical advantage over the regular use of inhalers with spacers in this group of patients with severe COPD. However, acute changes in total lung capacity, FVC and 6MWD may be useful predictors of the longer-term effects of nebulized BDs in individual patients.