The combination of an inhaled corticosteroid (ICS), budesonide, and a rapid long-acting β(2)-agonist (LABA), formoterol, in a single inhaler for use as maintenance and reliever therapy (Symbicort Turbuhaler SMART™) effectively achieves a high level of asthma control and reduces exacerbations and asthma-related hospitalizations. The COSMOS study, a multinational, 12-month study (N = 2143), compared budesonide/formoterol maintenance and reliever therapy with salmeterol/fluticasone propionate plus as-needed salbutamol, allowing physicians to modify maintenance doses of both combinations according to routine clinical practice. The aim of this post hoc sub-group analysis of the COSMOS study is to provide focused data on budesonide/formoterol maintenance and reliever therapy compared with salmeterol/fluticasone propionate plus as-needed salbutamol in patients (aged ≥16 years) enrolled across Asian countries, specifically China, Korea, Taiwan and Thailand. This sub-analysis of the COSMOS study concerns all 404 randomized patients ≥16 years of age (mean forced expiratory volume in 1 second [FEV(1)] 69.1%) who were recruited from Asian countries. Patients received either budesonide/formoterol (Symbicort Turbuhaler SMART™, n = 198), starting dose 160 mg/4.5 mg two inhalations twice daily (bid) [plus additional as-needed inhalations], or salmeterol/fluticasone propionate (Seretide(®) Diskus(®), n = 206), starting dose 50 mg/250 mg bid (plus salbutamol [Ventolin(®)] as needed). Maintenance doses could be titrated by clinicians after the first 4 weeks (budesonide/formoterol maintenance plus as needed, n = 198; salmeterol/fluticasone propionate plus salbutamol, n = 206). To allow for free adjustment in maintenance doses in both arms, the trial was performed open-label; maintenance doses could be titrated by clinicians after the first 4 weeks. The time to first severe exacerbation (defined as deterioration in asthma resulting in hospitalization/emergency room treatment, oral corticosteroids for ≥3 days or unscheduled visit leading to treatment change) was the primary variable. The time to first severe exacerbation was prolonged in patients using maintenance plus as-needed budesonide/formoterol compared with salmeterol/fluticasone propionate plus salbutamol (log-rank p = 0.024). The risk of a first exacerbation was reduced by 44% (hazard ratio 0.56; 95% confidence interval [CI] 0.32, 0.95; p = 0.033) in patients using the adjusted budesonide/formoterol regimen versus titrated salmeterol/fluticasone propionate. The overall exacerbation rates were 0.16 versus 0.26 events/patient-year, respectively, with a 38% reduction (rate ratio 0.62/patient/year; 95% CI 0.41, 0.94; p = 0.024) in favour of the budesonide/formoterol regimen. Compared with baseline, both regimens provided clinically relevant improvements in asthma control, quality of life and FEV(1); no statistically significant differences between the treatment groups were observed. Mean adjusted (standard deviation) ICS dose (expressed as beclomethasone dose equivalents) during treatment, including as-needed budesonide doses, was 944 (281) and 1034 (394) μg/day, respectively, in patients using maintenance plus as-needed budesonide/formoterol compared with salmeterol/fluticasone propionate. In patients (aged ≥16 years) enrolled from Asian countries as part of the COSMOS study, the budesonide/formoterol maintenance and reliever regimen was associated with a lower future risk of exacerbations versus the physicians' free choice of salmeterol/fluticasone propionate dose plus salbutamol. Single inhaler combination treatment with maintenance plus as-needed budesonide/formoterol was also at least as efficacious as salmeterol/fluticasone propionate dose plus salbutamol in improving current asthma control.