Cox proportional hazard regression of mortality regressed on patient turnover on units, and measures of low registered nurse and nursing support staffing (78 303 patients, 133 742 unique patient-unit-shift observations)
| Variable | HR | 95% CI | P value | |
| Lower | Upper | |||
| Model 1: Number of shifts patient experienced with low RN staffing or high turnover | ||||
| Number of low RN shifts | 1.023 | 1.011 | 1.035 | <0.001 |
| Number of high turnover shifts | 0.984 | 0.955 | 1.013 | 0.270 |
| Model 2: Number of shifts patient experienced with low registered nurse staffing or high turnover in second to fifth days of admission | ||||
| Number of low RN shifts | 1.091 | 1.059 | 1.125 | <0.001 |
| Number of high turnover shifts | 0.970 | 0.924 | 1.017 | 0.210 |
| Model 3: Number of shifts patient experienced with low registered nurse or nursing support staffing or high turnover | ||||
| Number of low RN shifts | 1.027 | 1.002 | 1.053 | 0.035 |
| Number of low nursing support shifts | 1.030 | 1.017 | 1.042 | <0.001 |
| Number of shifts with both low RN and nursing support | 1.025 | 1.008 | 1.043 | 0.004 |
| Number of high turnover shifts | 0.990 | 0.962 | 1.020 | 0.515 |
| Model 4: Number of shifts patient experienced with low registered nurse or nursing support staffing or high turnover in second to fifth days of admission | ||||
| Number of low RN shifts | 1.048 | 0.998 | 1.100 | 0.061 |
| Number of low nursing support-shifts | 1.032 | 1.008 | 1.057 | 0.010 |
| Number of shifts with both low RN and nursing support | 1.136 | 1.089 | 1.185 | <0.001 |
| Number of high turnover shifts | 0.976 | 0.93 | 1.024 | 0.314 |
Results from four separate Cox proportional hazard regressions of mortality. All regressions include 78 303 patients and 453 429 unique patient-unit-day observations. All regressions include measures of patient age, gender and Charlson index score, calendar month and year.
RN, registered nurses.