Results of interrupted time series analyses
| Outcome | Pre-intervention slope | Slope change ratio | Post-intervention slope | |
| All-cause readmissions within 7 days Poisson regression: slopes are monthly incidence rate ratios | Control site | 0.96 (0.92 to 0.99) p=0.01* | 1.06 (1.01 to 1.11) p=0.02* | 1.01 (1.00 to 1.03) p=0.17 |
| Intervention site | 0.98 (0.98 to 0.99) p<0.001*** | 1.00 (0.97 to 1.02) p=0.66 | 0.99 (0.98 to 1.00) p=0.04* | |
| All-cause readmissions within 7–30 days Poisson regression: slopes are monthly incidence rate ratios | Control site | 0.98 (0.95 to 1.00) p=0.08 | 1.02 (0.98 to 1.05) p=0.31 | 0.99 (0.98 to 1.01) p=0.38 |
| Intervention site | 0.99 (0.99 to 1.00) p<0.001*** | 0.98 (0.97 to 1.00) p=0.01* | 0.97 (0.97 to 0.98) p<0.001*** | |
| All-cause readmissions within 30–90 days Poisson regression: slopes are monthly incidence rate ratios | Control site | 0.98 (0.95 to 1.00) p=0.07 | 1.01 (0.97 to 1.04) p=0.62 | 0.98 (0.97 to 1.00) p=0.02* |
| Intervention site | 0.99 (0.99 to 0.99) p<0.001*** | 0.98 (0.97 to 1.00) p=0.004** | 0.97 (0.97 to 0.98) p<0.001*** | |
| Length of stay Log-linear regression: slopes are monthly ratios | Control site | 0.98 (0.97 to 0.99) p<0.001*** | 1.02 (1.01 to 1.03) p=0.005** | 1.00 (1.00 to 1.01) p=0.61 |
| Intervention site | 1.00 (1.00 to 1.00) p=0.62 | 1.00 (1.00 to 1.00) p=0.64 | 1.00 (1.00 to 1.00) p=0.32 | |
| In-hospital mortality Poisson regression: slopes are monthly incidence rate ratios | Control site | 0.95 (0.92 to 0.98) p<0.001*** | 1.06 (1.02 to 1.10) p=0.007** | 1.00 (0.99 to 1.02) p=0.60 |
| Intervention site | 0.99 (0.98 to 0.99) p<0.001*** | 0.96 (0.95 to 0.98) p<0.001*** | 0.95 (0.95 to 0.96) p<0.001*** |
Results of interrupted time series analyses for 7-day, 30-day and 90-day all-cause readmission rates, length of stay and in-hospital mortality for patients discharged following an admission with heart failure. All models enforce continuity in time trends across the interruption. For each model, the following estimates are shown with 95% CIs and p values (*0.01≤p<0.05, **0.001≤p<0.01, ***p<0.001): pre-intervention slope compared with no trend, difference in slope from pre-intervention to post-intervention, post-intervention slope compared with no trend. The resulting slopes for the control site post-intervention, the intervention site during pre-intervention and the intervention site post-intervention are also shown, along with the corresponding differences in slope. Since all models were multiplicative due to the log link, all differences are ratios.