Risk differences per 10 000 patients for asthma-related deaths and intubations (adapted from GlaxoSmithKline30)
| Comparison | Salmeterol | Comparator | Asthma-related deaths RD (95% CI) | Salmeterol | Comparator | Asthma-related intubations RD (95% CI) | ||||
| n | N | n | N | n | N | n | N | |||
| Salmeterol vs placebo (no ICS) | 8 | 9463 | 0 | 8932 | 8.79 (−8.23 to 25.80) | 15 | 9463 | 10 | 8932 | 5.42 (−13.67 to 24.52) |
| Salmeterol + ICSBK vs ICSBK | 5 | 10264 | 3 | 10135 | 2.03 (−12.17 to 16.22) | 20 | 10264 | 13 | 10135 | 7.16 (−10.08 to 24.39) |
| Salmeterol + ICSSP vs ICS | 1 | 2841 | 0 | 3040 | 4.64 (−28.77 to 38.04) | 0 | 2841 | 0 | 3040 | 8.21 (−25.91 to 42.33) |
| Salmeterol + ICSSD vs ICS | 0 | 11437 | 0 | 11163 | Not estimable | 0 | 11437 | 0 | 11163 | Not estimable |
BK, background; ICS, inhaled corticosteroid; n, number of events; N, total sample; RD, risk difference; SD, single device; SP, separate inhaler devices.