Changes in hospital nurse work environments and nurse job outcomes: an analysis of panel data

Abstract

Background: One strategy proposed to alleviate nursing shortages is the promotion of organizational efforts that will improve nurse recruitment and retention. Cross-sectional studies have shown that the quality of the nurse work environment is associated with nurse outcomes related to retention, but there have been very few longitudinal studies undertaken to examine this relationship.

Objectives: To demonstrate how rates of burnout, intention to leave, and job dissatisfaction changed in a panel of hospitals over time, and to explore whether these outcomes were associated with changes in nurse work environments.

Methods: A retrospective, two-stage panel design was chosen for this study. Survey data collected from large random samples of registered nurses employed in Pennsylvania hospitals in 1999 and 2006 were used to derive hospital-level rates of burnout, intention to leave current position, and job dissatisfaction, and to classify the quality of nurses' work environments at both points in time. A two-period difference model was used to estimate the dependence of changes in rates of nurse burnout, intention to leave, and job dissatisfaction on changes in nurse work environments between 1999 and 2006 in 137 hospitals, accounting for concurrent changes in nurse staffing levels.

Results: In general, nurse outcomes improved between 1999 and 2006, with fewer nurses reporting burnout, intention to leave, and job dissatisfaction in 2006 as compared to 1999. Our difference models showed that improvements in work environment had a strong negative association with changes in rates of burnout (β=-6.42%, p<0.01) intention to leave (β=-4.10%, p<0.01), and job dissatisfaction (β=-8.00%, p<0.01).

Conclusions: Improvements in nurse work environments over time are associated with lower rates of nurse burnout, intention to leave current position, and job dissatisfaction.

Fig. 1

Change in nursing organization characteristics from 1999 to 2006.

Figure 2

Comparing Mean Changes in Nurse Outcome Rates among Pennsylvania Hospitals from 1999 to 2006, by Change in Work Environment (n=137 hospitals) † †Mean change represents the least-square average changes from a linear regression model, adjusted for staffing and hospital characteristics (teaching status, bed size, and technology status)