Patient counts for hospitalizations and ED visits for 30 and 90 days and within-group/between-group IRR based on Poisson regression models
| Pre-N/732 | Post-N/732 | Pre-IRR vs post-IRR (95% CI) (p value) | Risk change based on IRR | P value for difference in the risk change | |
| 30-day hospitalizations | |||||
| Cases (n=366) | 18 | 4 | 0.21 (0.07 to 0.60) (0.003) | −79% | 0.02 |
| Controls (n=366) | 6 | 8 | 1.14 (0.47 to 2.75) (0.77) | +14% | |
| Cases vs controls IRR (95% CI) (p value) | 2.71 (1.03 to 7.13) (0.043) | ||||
| Cases vs controls adjusted IRR (95% CI) (p value) | 0.53 (0.18 to 1.54) (0.24) | ||||
| 90-day hospitalizations | |||||
| Cases | 33 | 8 | 0.23 (0.11 to 0.50) (<0.001) | −77% | <0.001 |
| Controls | 9 | 18 | 1.58 (0.75 to 3.33) (0.23) | +58% | |
| Cases vs controls IRR (95% CI) (p value) | 3.25 (1.44 to 7.34) (0.005) | ||||
| Cases vs controls adjusted IRR (95% CI) (p value) | 0.4 (0.20 to 0.95) (0.04) | ||||
| 30-day ED visits | |||||
| Cases | 24 | 20 | 0.79 (0.42 to 1.47) (0.45) | −21% | 0.12 |
| Controls | 31 | 12 | 0.39 (0.22 to 0.71) (0.002) | −61% | |
| Cases vs controls IRR (95% CI) (p value) | 0.85 (0.49 to 1.46) (0.552) | ||||
| Cases vs controls adjusted IRR (95% CI) (p value) | 1.77 (0.80 to 3.80) (0.14) | ||||
| 90-day ED visits | |||||
| Cases | 60 | 39 | 0.62 (0.42 to 0.92) (0.016) | −38% | 0.85 |
| Controls | 55 | 33 | 0.66 (0.45 to 0.96) (0.03) | −34% | |
| Cases vs controls IRR (95% CI) (p value) | 1.20 (0.84 to 1.73) (0.316) | ||||
| Cases vs controls adjusted IRR (95% CI) (p value) | 1.08 (0.64 to 1.84) (0.77) | ||||
| 30-day hospitalizations and ED visits | |||||
| Cases | 41 | 24 | 0.55 (0.34 to 0.90) 0.018) | −45% | 0.88 |
| Controls | 36 | 20 | 0.53 (0.33 to 0.84) (0.006) | −47% | |
| Cases vs controls IRR (95% CI) (p value) | 1.18 (0.75 to 1.85) (0.486) | ||||
| Cases vs controls adjusted IRR (95% CI) (p value) | 1.21 (0.67 to 2.19) (0.53) | ||||
| 90-day hospitalizations and ED visits | |||||
| Cases | 98 | 46 | 0.49 (0.36 to 0.67) (<0.001) | −51% | 0.03 |
| Controls | 61 | 47 | 0.80 (0.58 to 1.10) (0.18) | −20% | |
| Cases vs controls IRR (95% CI) (p value) | 1.53 (1.09 to 2.13) (0.013) | ||||
| Cases vs controls adjusted IRR (95% CI) (p value) | 0.82 (0.54 to 1.26) (0.37) | ||||
Counts are shown for the number of cases and controls who had at least one hospitalization and/or ED visit. IRR estimates are shown within each group (comparing the preintervention to postintervention periods) and between groups (comparing cases with controls) in both the preintervention and postintervention periods. For the between-group comparisons in the postintervention period, the IRRs presented are adjusted for preintervention utilization, age, sex, baseline HbA1c and corrected for matching using the cluster option in Stata. The final column represents the significance of the comparison between the groups of their respective within-group risk change from preintervention to postintervention as determined by longitudinal Poisson models that include time and group interaction.
ED, emergency department; HbA1c, hemoglobin A1c; IRR, incidence risk ratio.