Baseline characteristics for the entire cohort and the propensity-matched population.
| Entire cohort (n=35 691) | Propensity-matched population* (n=7986) | |||
| Standard particles ICS† (n=34 220) | Extrafine particles ICS† (n=1471) | Standard particle size ICS† (n=6654) | Extrafine particles ICS† (n=1332) | |
| Age, years, median (IQR) | 70.7 (62.9–77.8) | 69.1 (61.5–76.2) | 69.0 (61.3–76.6) | 69.1 (61.4–75.9) |
| Male, n (%) | 15 481 (45.2) | 646 (43.9) | 2934 (44.1) | 590 (44.3) |
| BMI, kg/m2, median (IQR) | 25.0 (21.0–29.0) | 25.0 (21.5–29.0) | 25.0 (22.0–29.0) | 25.0 (21.5–29.0) |
| BMI unknown, n (%) | 2460 (7.2) | 89 (6.1) | – | – |
| Active | 10 818 (31.6) | 404 (27.5) | 1958 (29.4) | 393 (29.5) |
| Former | 19 960 (58.3) | 885 (60.2) | 4266 (64.1) | 858 (64.4) |
| Never | 1066 (3.1) | 84 (5.7) | 430 (6.5) | 81 (6.1) |
| Unknown | 2375 (6.9) | 98 (6.7) | – | – |
| FEV1 (%), median (IQR) | 45.0 (33.0–59.0) | 46.0 (34.0–60.0) | 46.0 (33.0–60.0) | 45.0 (33.5–60.0) |
| GOLD stages 1–4 according to FEV1 (%), n (%) | ||||
| ≥ 80 | 1465 (4.3) | 70 (4.8) | 351 (5.3) | 62 (4.7) |
| 79–50 | 11 643 (34.0) | 520 (35.4) | 2495 (37.5) | 500 (37.5) |
| 49–30 | 12 145 (35.5) | 520 (35.4) | 2488 (37.4) | 510 (38.3) |
| <30 | 6377 (18.6) | 268 (18.2) | 1320 (19.8) | 260 (19.6) |
| Unknown | 2590 (7.6) | 93 (6.3) | – | – |
| AECOPD-Hosp, n (%) | ||||
| 0 | 22 354 (65.3) | 1044 (71.0) | 4350 (65.4) | 943 (70.8) |
| 1 | 7376 (21.6) | 241 (16.4) | 1413 (21.2) | 220 (16.5) |
| ≥2 | 4490 (13.1) | 186 (12.6) | 891 (13.4) | 169 (12.7) |
| ≥1 All-cause-hosp, n (%) | 21 702 (63.4) | 837 (56.9) | 4204 (63.2) | 759 (57.0) |
| OCS treatments, n(%) | ||||
| 0 | 16 948 (49.5) | 601 (40.9) | 2720 (40.9) | 536 (40.2) |
| 1 | 4087 (11.9) | 180 (12.2) | 805 (12.1) | 168 (12.6) |
| ≥2 | 13 185 (38.5) | 690 (46.9) | 3129 (47.0) | 628 (47.1) |
| Prophylactic OCS, n (%) | 3501 (10.2) | 185 (12.6) | 820 (12.3) | 161 (12.1) |
Additional baseline characteristics can be found in online supplemental table 5.
*Propensity matched cohort. Propensity matching was done using a greedy match algorithm matching up to five controls receiving standard particle size ICS per case receiving extrafine particles. Matching was done on the following variables: age, sex, smoking status, forced expiratory volume in the first second, BMI, daily use of ICSs in budesonide equivalent dose, number of oral corticosteroids treatments and overlapping asthma.
†Devices containing beclomethason hydrofluoroalkane solutions or ciclesonide sprays were categorised as extrafine particles, all other ICS devices were categorised as standard particle size.
AECOPD, acute exacerbation of chronic obstructive pulmonary disease; BMI, body mass index; FEV1, forced expiratory volume in the first second; GOLD, Global Initiative for Chronic Obstructive Lung Disease; Hosp, hospitalisation; ICS, inhaled corticosteroids; OCS, oral corticosteroids.