2.3 | Sampling
Sample size was calculated basing on 2-sample equivalence model. The minimum sample size was calculated as 100 participants according to the significant level of α= 0.05, the statistical power of 80%, the mean difference less than 1.3 days to clinical stability (primary outcome) between two groups, and the standard deviation of 2 days in each group from pretest 17. Given an estimation drop-out rate (28%), including in-hospital mortality rates and the ward conversion rates, a total of 140 HF patients were enrolled.