Baseline characteristics of interest according to the cluster group
| Covariate | Cluster 1: diuretic (N=454) | Cluster 2: reversible, no diuretic (N=756) | Cluster 3: not reversible, no diuretic (N=333) | p Value |
|---|---|---|---|---|
| Age, median years (IQR) | 67 (62–74) | 64 (58–70) | 65 (59–71) | <0.0001 |
| Body mass index, median m/kg2 (IQR) | 28 (25–34) | 26 (23–30) | 25 (22–29) | <0.0001 |
| Smoking status (%) | ||||
| Former | 65 | 58 | 53 | 0.0024 |
| Current | 35 | 42 | 47 | |
| Smoking, mean pack-years (IQR) | 52 (40–77) | 50 (37–70) | 48.5 (36–70) | 0.0401 |
| FEV1% predicted (SD) | 33.9 (25.1–42.6) | 31.3 (23.9–39.4) | 37.7 (29.0–44.6) | <0.0001 |
| FEV1% reversibility (SD) | 18.55 (7.40–31.70) | 26.25 (18.60–38.20) | 4.50 (-1.00–8.70) | <0.0001 |
| Reversibility stratum* [no/yes] (%) | 60/40 | 41/59 | 100/0 | <0.0001 |
| Exacerbations requiring hospitalisation (past year) (%) | ||||
| 0 | 73.8 | 78.9 | 76.3 | ns |
| 1 | 23.6 | 19.5 | 19.8 | |
| 2 | 2.6 | 1.6 | 3.9 | |
| Exacerbations requiring oral steroids/antibiotics (past year) (%) | ||||
| 1 | 62.6 | 61.1 | 67.0 | ns |
| 2 | 20.0 | 25.0 | 19.2 | |
| 3 | 8.1 | 7.7 | 7.2 | |
| Baseline medications (%) | ||||
| Diuretics | 100 | 0.0 | 0.0 | <0.0001 |
| Anti-thrombotics | 50.7 | 32.0 | 40.2 | <0.0001 |
| ACE inhibitors | 50.0 | 26.7 | 30.6 | <0.0001 |
| Lipid modifiers | 49.3 | 28.7 | 33.6 | <0.0001 |
| Calcium channel blockers | 33.5 | 16.3 | 14.1 | <0.0001 |
| Psycholeptics | 32.6 | 21.4 | 24.0 | <0.0001 |
| Anti-histamines | 30.4 | 22.4 | 23.7 | 0.0062 |
| β-blockers | 24.0 | 10.8 | 12.9 | <0.0001 |
| Cardiac therapy | 23.1 | 8.6 | 7.2 | <0.0001 |
| Diabetes | 17.4 | 7.8 | 8.4 | <0.0001 |
| Anti-anaemics | 13.2 | 5.8 | 3.6 | <0.0001 |
| Anti-hypertensives | 7.3 | 3.2 | 1.5 | <0.0001 |
*Reversibility based on change in FEV1 from baseline following four puffs (360 µg) albuterol, defined as a ≥12% and ≥200 ml increase.
ACE, angiotensin converting enzyme; FEV1, forced expiratory volume in 1 s; IQR, interquartile range; ns, not significant; SD, standard deviation.