Table 2.
Primary outcomes (model-based estimates)
| Inhaled budesonide (95% BCI) | Usual care (95% BCI) | Estimated benefit median time to recovery or hospital admission or death rate (95% BCI) | Hazard ratio or odds ratio (95% BCI) | Probability of superiority | |
|---|---|---|---|---|---|
| Primary analysis—SARS-CoV-2-positive participants | |||||
| Number of participants | 787 | 1069 | .. | .. | .. |
| Time to first reported recovery, days* | 11·8 (10·0 to 14·1) | 14·7 (12·3 to 18·0) | 2·94 (1·19 to 5·11) | 1·21 (1·08 to 1·36) | >0·999 |
| Hospital admission or death at 28 days† | 6·8% (4·1 to 10·2) | 8·8% (5·5 to 12·7) | 2·0% (−0·2 to 4·5) | 0·75 (0·55 to 1·03) | 0·963 |
| Secondary analysis—all participants | |||||
| Number of participants | 990 | 1858 | .. | .. | .. |
| Time to first reported recovery, days* | 10·9 (8·9 to 13·2) | 13·3 (11·1 to 16·7) | 2·54 (1·00 to 4·54) | 1·18 (1·07 to 1·30) | >0·999 |
| Hospital admission or death at 28 days† | 5·8% (3·4 to 8·6) | 7·3% (4·5 to 10·6) | 1·5% (−0·3 to 3·6) | 0·78 (0·57 to 1·04) | 0·953 |
| Sensitivity analysis—concurrent randomisation population | |||||
| Number of participants | 787 | 838 | .. | .. | .. |
| Time to first reported recovery, days* | 11·7 (9·8 to 14·2) | 15·0 (12·5 to 18·3) | 3·26 (1·46 to 5·43) | 1·24 (1·10 to 1·39) | >0·999 |
| Hospital admission or death at 28 days† | 6·6% (3·8 to 10·1) | 8·9% (5·2 to 13·1) | 2·2% (0·0 to 4·9) | 0·73 (0·53 to 1·00) | 0·975 |
BCI=Bayesian credible interval.
Estimated benefit in median times to recovery are derived from a Bayesian piecewise exponential model adjusted for age and comorbidity at baseline, with 95% BCI; a positive value in estimated benefit in median time to recovery (or hazard ratio >1) corresponds to a reduction in time to recovery in days with budesonide compared with usual care; treatment superiority is declared if probability of superiority is ≥0·99 versus usual care.
Estimated absolute percentage differences in hospital admission or death were derived from a Bayesian logistic regression model adjusted for age and comorbidity at baseline, with 95% BCI; a positive value in the estimated percentage difference (or odds ratio <1) favours budesonide; treatment superiority is declared if probability of superiority is ≥0·975 versus usual care.