Research regarding the impact of special care units (SCUs) on the experience of staff provides a somewhat inconsistent picture. Several studies have examined the effect on staff of working in a facility that has an SCU without including a comparison with staff who work on conventional units. These studies have found generally positive results, including lower turnover among nurses. Despite the promise of SCUs, the degree to which they differ in important outcomes from conventional nursing home care has not been firmly established. Some comparative studies have found no clear differences in resident outcomes for SCUs. In addition, studies found that mere segregation of residents into an SCU without specialized programs and staff training does not seem to produce beneficial outcomes. However, SCUs that adhere to the spirit of specialized dementia care do appear to produce positive outcomes on residents.
This study examined job satisfaction and intention to quit for SCU and traditional-unit nursing staff.
This study combines data from 892 frontline nursing home staff collected as part of the Partners in Caregiving (PIC) study and the Partners in Caregiving in a Special Care Environment (PIC-SCU) study. Combining the baseline data of these two studies provides a unique opportunity to examine differences between SCU nursing staff and staff in traditional nursing home units with a large number of facilities and respondents. The two studies used identical recruitment and data collection methods and collected the same information from both groups of respondents. Participants in the PIC study included staff members in 20 nursing homes in the central New York region. All units were randomly selected from within the 20 participating facilities. All nursing staff members on the units, including registered nurses (RNs), licensed practical nurses (LPNs), and certified nursing assistants (CNAs), were recruited for the study. Interviews were attempted with 817 staff members, resulting in 655 respondents at Time 1, for a completion rate of 80 percent. The current comparison study includes only day- and evening-shift staff data (N = 545) to make the sample comparable with the PIC-SCU study. The PIC-SCU study included 20 facilities in three Connecticut counties randomly selected from all skilled nursing facilities with SCUs or dementia programs. The 20 participating facilities were randomly divided into control and intervention groups. The current article includes data from the 18 facilities with traditional SCUs. Of 377 eligible nurses and nursing assistants on the 18 units, 352 completed baseline interviews (93 percent response rate). Research staff of each study conducted telephone interviews with unit nursing staff at three time points (at baseline and at two and six months later).
The two dependent measures evaluated here include job satisfaction and intention to quit. These outcomes are two key indicators of staff well-being, which strongly affects the quality of resident care. Staff completed the Generic Job Satisfaction Scale, which includes nine items addressing different aspects of the job, such as receiving recognition for a job well done, good wages, and feeling secure about the job. Likelihood of leaving employment in the nursing home was measured by a single item that asked, “Thinking about the next 12 months, how likely do you think it is that you will decide to quit your job?” (very likely, likely, not too likely, not at all likely).
Independent variables comprised the socio-demographic indicators of age, race, job title (RN or LPN versus CNA), number of years they have worked in this nursing home, and number of years worked in any nursing home. Gender is not included, as more than 90 percent of both groups were female. Characteristics of the work environment included unit type (SCU versus traditional), facility and unit size, unionization, having enough time to accomplish all the required tasks (versus being somewhat or very short of time), and how much involvement staff reported in resident-care planning (none, a little, some, or alot).
Bivariate comparisons were conducted between the traditional and SCU staff on all independent and dependent measures. Statistical tests included t tests and chi-square tests. The authors tested the first hypothesis, that job satisfaction and commitment to continuing in the job will be higher on SCUs, via t tests comparing the means of the job satisfaction and intention to quit measures for staff on SCUs versus traditional units. A set of stepwise ordinary least square regression models were estimated for each of the two dependent variables to test the second hypothesis: that the higher levels of satisfaction and commitment can be explained by more satisfying and less conflictual relationships with coworkers, supervisors, and family members. The first step in each model included unit type, relevant socio-demographic indicators, and facility environment factors. A final step added variables assessing relationships with coworkers, supervisors, and families.
SCU staff reported significantly higher job satisfaction and less likelihood of quitting than those on traditional units. Multivariate analyses examining demographic, environmental, and personal relationship variables indicated that the benefits to staff of working on an SCU depended significantly on the positive relationships that staff have with supervisors, coworkers, and families.
The authors of the study identified the following limitations for the study:
1) There are systematic regional differences in the sample: The SCUs were all in Connecticut and the traditional units were all in New York state. However, there do not appear to be regulatory or demographic differences between the two states that are likely to have affected results in a significant way.
2) The traditional units were all drawn from not-for-profit facilities, whereas half of the SCUs were proprietary facilities. Although turnover rates may differ between proprietary and nonprofit facilities, turnover rates for these specific regions were not available.
3) A final limitation is the unknown impact of possible differences in resident case mix. The data collected did not include any rating of resident acuity; therefore, it is possible that the differences between unit types stem from the residents on the dementia unit being easier to care for in some way than those on the traditional units.