Sensitivity analyses for the primary outcome of mortality
| Factor | Subjects n | Adjusted model | p-value |
| HR (95% CI) | |||
| Baseline MVPA (closest to index hospitalisation) | |||
| Patients with spirometry#,¶ | 1292 | ||
| 0 min·week−1 | 1.0 | ||
| 1–149 min·week−1 | 0.88 (0.60–1.28) | 0.50 | |
| ≥150 min·week−1 | 0.53 (0.28–1.01) | 0.06 | |
| Patients with smoking history# | 2066 | ||
| 0 min·week−1 | 1.0 | ||
| 1–149 min·week−1 | 0.71 (0.52–0.97) | 0.03 | |
| ≥150 min·week−1 | 0.57 (0.35–0.90) | 0.02 | |
| Any MVPA# (closest to index hospitalisation) | 2370 | ||
| 0 min·week−1 | 1.0 | ||
| ≥1 min·week−1 | 0.66 (0.51–0.85) | <0.001 | |
| Median/mode MVPA+ | 2370 | ||
| 0 min·week−1 | 1.0 | ||
| 1–149 min·week−1 | 0.61 (0.43–0.88) | <0.01 | |
| ≥150 min·week−1 | 0.62 (0.37–1.03) | 0.07 | |
| Highest MVPA§ | 2370 | ||
| 0 min·week−1 | 1.0 | ||
| 1–149 min·week−1 | 0.69 (0.55–0.85) | <0.001 | |
| ≥150 min·week−1 | 0.59 (0.47–0.76) | <0.001 |
Cox regression models adjusted for age, marital status, race/ethnicity, body mass index, smoking status, disease severity (medications, oxygen use, Charlson comorbidity index, pulmonary hypertension, depression, any hospitalisation in previous 12 months), ever attended pulmonary rehabilitation in the 3 years prior to the index hospitalisation, palliative care consultation during the index hospitalisation, length of stay, discharge disposition and duration of time from assessment of physical activity to index hospitalisation. MVPA: moderate to vigorous physical activity. #: included on indicated models; ¶: forced expiratory volume in 1s % predicted was included in the adjusted model and was not a significant predictor of death; +: based on all physical activity readings over 12 months prior to index hospitalisation; §: highest physical activity reading over 12 months prior to index hospitalisation.