Poisson regression models predicting initiation/improved adherence to bundle components (N=73)
| Model A: QI methods | Model B: hospital engagement | Model C: QI methods and hospital engagement | Model D: QI methods and type of hospital engagement | |||||
|---|---|---|---|---|---|---|---|---|
| IRR | 95% CI | IRR | 95% CI | IRR | 95% CI | IRR | 95% CI | |
| Adherence to Project JOINTS QI methods | 1.16 | (0.96 to 1.40) | 1.04 | (0.87 to 1.26) | 1.04 | (0.86 to 1.26) | ||
| Hospital engagement | 1.12 | (1.03 to 1.22)** | 1.11 | (1.02 to 1.21)* | ||||
| Interactive campaign activities | 1.08 | (0.91 to 1.30) | ||||||
| Campaign materials and tools | 1.13 | (1.01 to 1.27)* | ||||||
| Wald χ2 | 148.05** | 133.72** | 137.52** | 141.89** | ||||
| McFadden's R2 | 0.2299 | 0.2431 | 0.2436 | 0.2439 | ||||
Results from Poisson regression with robust standard errors. All models controlled for state, rural/urban location, hospital size, surgery volume, previous Institute for Healthcare Improvement experience, previous surgical site infection experience and number of prior components.
The Wald χ2 statistic is used to determine whether at least one of the predictors’ regression coefficient is not equal to zero.
R2 indicates the proportion of the variation in the outcome variable due to the independent variables included.
*significant at p≤0.05; **significant at p≤0.01.
JOINTS, Joining Organizations IN Tackling SSIs; SSI, surgical site infection; QI, quality improvement; IRR, incidence rate ratio.