Table 4.

Associations of Nurse Outcomes and Nurse-Reported Patient Outcomes With the Proportion of Nonpermanent Nurses, Before and After Adjustment^*

	Adjusted for Nurse Characteristics		Adjusted for Nurse Characteristics and Staffing-Resource Adequacy
	OR (95% CI)	P value	OR (95% CI)	P value
Nurse outcomes
Dissatisfaction with job	1.19 (1.07–1.33)	.001	1.00 (0.92–1.09)	.94
Plans to leave job within 1 y	1.39 (1.24–1.57)	<.001	1.28 (1.14–1.42)	<.001
High burnout	1.10 (0.98–1.23)	.10	0.91 (0.83–0.99)	.03
Nurse-reported patient outcomes
Wrong medication	0.92 (0.82–1.04)	.18	0.82 (0.74–0.92)	<.001
Nosocomial infection	1.08 (0.94–1.24)	.25	0.99 (0.86–1.13)	.84
Complaints from patients/families	1.05 (0.95–1.15)	.34	0.91 (0.84–0.99)	.04
Patient falls	1.06 (0.89–1.26)	.51	0.89 (0.76–1.05)	.16
Work-related injuries	1.16 (1.04–1.29)	.008	1.01 (0.91–1.14)	.75
Verbal abuse directed toward nurses	1.02 (0.92–1.13)	.68	0.90 (0.82–0.99)	.03

Abbreviations: OR, odds ratio; CI, confidence interval.

The odds ratios represent the increase (when >1.0) or decrease (when <1.0) in the odds of nurses reporting the outcome with each 10% increase in the proportion of nonpermanent nurses. They were estimated using logistic regression models with robust standard errors.

^*Analysis of 1999 Pennsylvania Hospital Outcomes Study data.