Choi, JiSun PhD, RN; Boyle, Diane K. PhD, RN
Registered nurses (RNs) are the largest group of clinical care providers in healthcare systems, and their role has been well recognized as crucial for better patient outcomes and safety.1,2 For these essential care providers working in acute care hospital settings, job satisfaction not only is important for their retention3,4 but also can affect patient outcomes. Recent evidence links job satisfaction to nurses’ ratings of care quality and patient satisfaction with hospital care.5,6 Few studies have been conducted to investigate the direct relationship between nurse job satisfaction and patient clinical outcomes. Moreover, little is known about how unit-level RN job satisfaction (ie, RN workgroup job satisfaction) affects unit patient outcomes. This study examined the relationship between RN workgroup job satisfaction and patient falls on 4 types of acute care hospital units (step-down, medical, surgical, and combined medical-surgical [MS] units).
Patient Falls in Acute Care Hospitals
Despite the increased attention of hospital executives and staff to patient safety, falls are 1 of the most commonly occurring adverse events among hospitalized patients, with about 30% of falls resulting in injury disability or death.7 Overall hospital fall rates reported in previous studies have remained relatively constant, ranging from 2.2 to 4.1 per 1000 patient days.8 The fall rates, however, varied significantly by the type of patient care unit, such as medical, surgical, and rehabilitation. In a study using 2002 unit-level data from the National Database of Nursing Quality Indicators (NDNQI), the observed patient fall rates were the highest on medical units and the lowest on critical care (CC) units among the most common nursing units (CC, step-down, medical, and surgical).9 In another study using 2004 NDNQI data, the estimates of fall rate per 1000 patient days on the selected units in Magnet®-designated hospitals were 1.12 for intensive care units, 4.35 for medical units, and 6.84 for rehabilitation units; those in non-Magnet hospitals were 1.30 for CC units, 4.54 for medical units, and 7.15 for rehabilitation units.10 The literature includes numerous studies on patient falls focusing on the identification of risk factors and the development of fall prevention strategies. Much of the research on patient falls, however, has been conducted at the hospital level, not the patient care unit level.
Nursing Factors Associated With Patient Falls
Although patient falls have been advanced as a nursing-sensitive outcome indicator,11 few researchers have investigated nursing factors (eg, nurse staffing, nursing interventions, and RN experience) that might affect patient falls. Nurse staffing has been a predominant focus of investigation in relation to patient falls in acute care hospitals. Consistent with findings from a systematic review of studies about nurse staffing and patient outcomes that demonstrated the contribution of higher RN staffing to better patient outcomes,12 several researchers reported a significant relationship between nurse staffing and patient fall rates.8-10,13 In addition to nurse staffing as an important factor in patient falls, RN experience, defined as average years of RN experience in nursing, was found to be related to patient falls; units with more experienced RNs had fewer patient falls.13
Because nurses provide 24-hour direct care at the patient’s bedside, their work attitudes must be assessed in relation to patient outcomes. Recent findings indicated that nurses with direct patient care responsibility were much less satisfied than those with no patient care responsibility.8 Within hospitals, Boyle and colleagues14 found a significant difference in overall unit-level RN job satisfaction across 10 unit types, including MS, step-down, CC, pediatric, and emergency departments. Registered nurse workgroups in MS, step-down, and CC units had moderate levels of satisfaction. Among these 3 unit types, RN workgroups in combined MS units reported the lowest satisfaction. Thus, patient outcomes might be affected by the different levels of job satisfaction among RN workgroups who provide direct patient care on units where adverse events occur. No studies wherein researchers examined the effect of RN workgroup job satisfaction on patient falls were found in the literature.
For this study, we used linked 2009 data from 2 sources: the NDNQI and the NDNQI RN Survey. The NDNQI is part of the American Nurses Association Patient Safety and Nursing Quality Initiative and provides participating hospitals with national comparison reports at the unit level. Quarterly data regarding hospital characteristics (eg, teaching status and Magnet status) and nursing quality indicators (eg, nurse staffing and patient fall rates), as well as annual RN survey data on work attitudes and RN characteristics (eg, job satisfaction, education, and tenure) directly from RNs employed by NDNQI member hospitals, are collected by NDNQI. More details on the NDNQI, including data collection methods, are described in other studies.9,14 This study for the secondary analysis of NDNQI data was approved by the University of Kansas Medical Center’s institutional review board.
The patient care unit was the unit of analysis, not the individual patient or nurse. For this study, the units must have reported both nurse staffing and patient falls to NDNQI and participated in the NDNQI RN survey in 2009. We included only units that reported data on both nurse staffing and patient falls for at least 9 months during the 2009 calendar year. Also, units with at least 5 RNs participating in the RN survey and a 50% or greater response rate were included in our analysis to ensure an adequate representation of the unit RN workgroup. As a result, the final analytical sample consisted of 2763 inpatient units (576 hospitals; 63,034 RNs). All units were adult population–based units that included step-down (574 units), medical (708 units), surgical (537 units), and combined MS (944 units). Critical care units were not included in this study because they were different from other selected unit types in terms of patient acuity level and required nurse staffing levels.
Characteristics of hospitals with units selected for this study were teaching hospitals (47.2%), Magnet hospitals (33%), and hospitals with more than 300 staffed beds (36.3%). For RN respondents in our sample units, the average age was 39 years, and the average unit tenure was 5 years. Slightly more than half of the RNs (51.8%) had a bachelor’s or higher degree in nursing.
A patient fall, as defined by the NDNQI,15 is an unplanned descent to the floor with or without an injury to the patient. For this study, the total number of patient falls on the unit per year was calculated for the year 2009. To calculate a patient fall rate, the total number of falls on a unit for the year 2009 was divided by the total number of patient days on that unit for the same period of 2009 and then multiplied by 1000.
RN Workgroup Job Satisfaction
The 7-item NDNQI-adapted Job Enjoyment (JE) scale was used to measure overall job satisfaction at the unit level. Registered nurses were asked to assess each item based on the stem “nurses with whom I work on the unit would say that they...” For example, 1 item on this scale was “nurses with whom I work on the unit would say that they are fairly well satisfied with their job.” Items were scored using a 6-point Likert-type scale: 1 (strongly disagree) to 6 (strongly agree). For this study, we calculated the mean JE score for each RN and then the average of these scores for each unit. The potential mean scores ranged from 1 to 6, and higher scores indicated higher RN workgroup job satisfaction. Reliability and validity of the JE scale have been established at both the individual RN and unit workgroup level.14,16 In the current study, the individual RN-level internal consistency reliability coefficient for the scale was 0.92. The unit-level reliability was assessed by estimating intraclass correlation coefficients (ICCs). An ICC (2) value of 0.85, an estimate of the reliability of the group means, indicated adequate unit-level reliability. Also, an ICC (1) value of 0.20, an index of interrater reliability, indicates adequate within-group agreement to establish construct validity. That is, the individual RN responses about their workgroup job satisfaction were very similar to each other within a unit.
Variables included other nursing unit characteristics (nurse staffing, RN education, and RN unit tenure) and hospital characteristics (teaching status, Magnet status, and hospital size). Two nurse staffing variables were included in the analysis: total nursing hours per patient day (NHPPD) and RN skill mix. The average total NHPPD for the year 2009 was calculated; the number of hours worked by RNs, licensed practical nurses (LPNs), and unlicensed assistive personnel (UAP) assigned to the unit was added for the year 2009 and then divided by the sum of patient days for that unit during the same period. The average RN skill mix for the year 2009 was also calculated as the proportion of nursing hours worked by RNs among all nursing personnel in a unit during the year 2009. The education level of the RN, defined as the highest level of nursing education, was measured as the percentage of RNs with a bachelor’s degree or higher on the unit. Registered nurse tenure on the unit was measured as the average number of years the RNs had worked on the current unit.
Three hospital characteristics were included and dichotomized: hospital size, defined as the number of beds (<300 beds, 43%; ≥300 beds, 57%); teaching status (teaching hospital, 59%; nonteaching hospital, 41%); and Magnet status (Magnet hospital, 43%; non-Magnet hospital, 57%).
The data on study variables were summarized by selected unit type using descriptive statistics. The outcome variable for this study, patient falls, was a count variable. The nature of the data was hierarchical; units were nested in hospitals. Generalized linear mixed modeling is a preferable statistical technique to handle the nonnormal and nested data. Thus, random-intercept negative binomial regression analysis was performed using Stata version 11 (StataCorp LP, College Station, Texas) to examine the relationship between RN workgroup job satisfaction and patient falls while accounting for other nursing unit and hospital characteristics. The number of patient falls was modeled with a negative binomial distribution when the number of patient days was included in the model as an exposure variable. In the model, a random intercept was included to adjust for clustering units within hospitals. The patient care unit type was also included to statistically control for the differences in unit patient acuity levels.
The descriptive statistics on the 2 aspects of nurse staffing (total NHPPD and RN skill mix), RN unit tenure, RN workgroup job satisfaction, and fall rates are summarized in Table 1. The mean number of total NHPPD ranged from 8.64 for medical units to 10.29 for step-down units. The mean percentage of hours worked by RNs (RN skill mix) ranged from 66.44% on combined MS units to 73.79% on step-down units. Both total NHPPD and RN skill mix were higher in step-down units compared with the other 3 unit types (medical, surgical, and combined MS units), whereas the average tenure of RNs on the unit was shorter in step-down units than in other unit types. Registered nurse workgroup satisfaction across all unit types was moderate, with a range of 3.68 to 3.76. In 2009, the average number of falls per 1000 patient days across all units included in this study was 3.47, ranging from 2.87 on surgical units to 4.06 on medical units.
Table 2 presents the results of the random-intercept negative binomial regression model in which the relationship between RN workgroup job satisfaction and patient falls on the units was examined. When controlling for the other unit and hospital variables, RN workgroup job satisfaction was significantly associated with patient falls. For each 1-unit increase in RN workgroup job satisfaction, patient fall rates decreased by 5.9%. Among other nursing unit characteristics, RN skill mix and RN unit tenure were significantly associated with patient falls. The fall rate was 0.3% lower for each percentage point increase in the percentage of nursing hours provided by RNs among all nursing personnel. Fall rates were also 1.7% lower for each 1-year increase in the average unit tenure of the RNs on the units. Of the 3 hospital characteristics, only teaching status was significantly associated with patient falls. The patient fall rates on units in teaching hospitals were 9.7% higher than those in nonteaching hospitals.
Job satisfaction has long been studied as the strongest predictor of nurse retention, and its positive effect on patient outcomes has been posited in studies of nurse job satisfaction. Our study was conducted to investigate the direct relationship of RN job satisfaction to patient falls at the patient care unit level. Findings indicate that higher RN workgroup job satisfaction is significantly associated with fewer patient falls on the selected units, as well as other nursing unit characteristics, such as RN skill mix and RN unit tenure.
The patient care unit environment is complex and dynamic. The patient population and nurse staffing levels vary by unit and unit type. Such variations affect nursing care processes and likewise will affect staff satisfaction at the unit level. In our study, unit RN workgroups were moderately satisfied with their jobs and reported similar levels of job satisfaction across all 4 selected unit types. Among these unit types, however, RN workgroups in combined MS units were the least satisfied with their jobs, followed by RN workgroups in step-down units. The results were consistent with those of previous research that RN workgroups in MS and step-down units had the least satisfaction among study unit types, including CC, pediatric, rehabilitation, psychiatric, and maternal-newborn units.14 This finding may be due to the nature of nursing care delivery and work environment on these units. The importance of the link between working conditions and individual nurse job satisfaction in acute care hospitals has been well documented.17-19 In general, CC units are recognized as 1 of the most stressful places to work. However, nurses in combined MS units may be equally challenged to care for a large number of patients with various diagnoses on a daily basis. More research is needed to identify work environment factors influencing RN workgroup job satisfaction based on unit types.
Research has demonstrated that job satisfaction was positively and moderately associated with job performance.20 Indeed, nurses repeatedly have been found to be satisfied with their work when they provided high-quality care to patients.21 Not surprisingly, we found a negative relationship between RN workgroup job satisfaction and patient falls. Although successful fall prevention programs require interdisciplinary team efforts, nurses are key direct care providers in implementing unit-specific strategies. The individual RN’s attitude toward these fall prevention interventions may be affected by the workgroup’s job attitude on the unit where they work. Satisfied RN work groups may be more willing to change their nursing practices to prevent patient falls on their units, such as responding faster to patient call lights. Although further research is needed to investigate whether RN workgroup job satisfaction can serve as a proxy for nursing staff attitudes toward unit-specific quality improvement initiatives, assessing and benchmarking the level of RN workgroup job satisfaction to nationally comparable unit types would be beneficial to improve patient outcomes. Strategies to improve the satisfaction of RN workgroups should be considered based on specific units and patient population needs. Unit nurse managers are crucial in developing such strategies and implementing changes needed to improve unit-level practice. More importantly, efforts made by unit mangers to improve RN workgroup satisfaction would require organizational support and effective nursing leadership.
In addition to RN workgroup job satisfaction, findings show that other nursing unit characteristics, such as nurse staffing and RN unit tenure, are also important factors in reducing patient fall rates on the units. As expected, we found that longer RN unit tenure was associated with lower fall rates. With regard to nurse staffing, however, our findings were not consistent with a previous unit-level study in identifying a significant and inverse relationship between 2 aspects of nurse staffing (total NHPPD and RN skill mix) and patient falls.4 In the current study, only RN skill mix was associated with patient falls; a higher proportion of nursing hours provided by RNs was related to lower fall rates. In a study where researchers examined the effects of nursing hours provided by 3 separate nursing personnel (RNs, LPNs, and UAP) on patient falls using 2004 NDNQI data,10 interestingly, RN hours were inversely associated with patient falls, whereas LPN and UAP hours were positively associated with patient falls. Thus, to reduce the patient fall rate on units, it appears that experienced RN hours are more important than total nursing hours. This staffing pattern at the patient care unit level provides additional evidence that both quantity and quality of nurse staffing do matter to achieve patient safety. Nursing administrators and managers should give careful consideration to unit staffing plans by having RNs with adequate skill and knowledge to meet unit patient population needs and by retaining experienced RNs on the unit.
Several limitations of our study are acknowledged. We used cross-sectional data, and therefore, it was not possible to infer a causal relationship between RN workgroup job satisfaction and patient falls. Although unit-level data from a large and national sample of hospitals were used, the study sample was limited to NDNQI member hospitals, which are voluntary participants in NDNQI. These facilities tend to be larger and are more likely to be nonprofit than all hospitals nationwide.9,10 Findings should be generalized with caution. In our study, only unit type was included to control for the differences of patient population and acuity level. Adjustments limited to patient care unit type may be insufficient. More specific controls for other factors that might affect patient falls, such as patient-level risk factors and acuity, should be included in further research to explore these models.
Our study is among the 1st to examine the relationship between RN job satisfaction and clinical patient outcomes at the patient care unit level. The finding that RN workgroup job satisfaction is significantly and inversely associated with patient falls provides needed evidence for the relationship between nurse job satisfaction and better patient outcomes. Unit RN workgroups must be considered as key care providers to achieve patient safety in an effective way. Strategies to improve RN workgroup job satisfaction should be tailored on a unit-by-unit basis.
1. Institute of Medicine. Keeping Patients Safe: Transforming the Work Environment of Nurses. Washington, DC: National Academies Press; 2003.
2. Institute of Medicine. The Future of Nursing: Leading Change, Advancing Health. Washington, DC: National Academies Press; 2010.
3. Hayes LJ, O’Brien-Pallas L, Duffield C, et al. Nurse turnover: a literature review. Int J Nurs Stud. 2006; 43 (2): 237–263.
4. Lu H, Barriball KL, Zhang X, While AE. Job satisfaction among hospital nurses revisited: a systematic review. Int J Nurs Stud. 2012; 49 (8): 1017–1038.
5. Djukic M, Kovner CT, Brewer CS, Fatehi FK, Cline DD. Work environment factors other than staffing associated with nurses’ ratings of patient care quality. Health Care Manage Rev. 2013; 38 (2): 105–114.
6. McHugh MD, Kutney-Lee A, Cimiotti JP, Sloane DM, Aiken LH. Nurses’ widespread job dissatisfaction, burnout, and frustration with health benefits signal problems for patient care. Health Aff. 2011; 30 (2): 202–210.
7. Currie L. Fall and injury prevention. In: Hughes RG, ed. Patient Safety and Quality: An Evidence-Based Handbook for Nurses. Rockville, MD: Agency for Healthcare Research and Quality; 2008: 195–250.
8. Lake ET, Cheung RB. Are patient falls and pressure ulcers sensitive to nurse staffing? West J Nurs Res. 2006; 28 (6): 654–677.
9. Dunton N, Gajewski B, Taunton RL, Moore J. Nurse staffing and patient falls on acute care hospital units. Nurs Outlook. 2004; 52 (1): 53–59.
10. Lake ET, Shang J, Klaus S, Dunton N. Patient falls: association with hospital Magnet status and nursing unit staffing. Res Nurs Health. 2010; 33: 413–425.
12. Kane RL, Shamliyan TA, Mueller C, Duval S, Wilt TJ. The association of registered nurse staffing levels and patient outcomes: systematic review and meta-analysis. Med Care. 2007; 45: 1195–1204.
13. Dunton N, Gajewski B, Klaus SF, Pierson B. The relationship of nursing workforce characteristics to patient outcomes. Online J Issues Nurs. 2007; 12 (3). doi:10.3912/OJIN.Vol12No03Man03.
14. Boyle DK, Miller PA, Gajewski BJ, Hart SE, Dunton N. Unit type differences in RN workgroup job satisfaction. West J Nurs Res. 2006; 28: 622–640.
15. National Database of Nursing Quality Indicators. Guidelines for data collection on the American Nurses Association’s National Quality Forum endorsed measures. https://www.nursingquality.org/data.aspx
. Accessed January 17, 2013.
16. Taunton RL, Bott MJ, Koehn ML, et al. The NDNQI-adapted index of work satisfaction. J Nurs Meas. 2004; 12: 101–122.
17. Kelly LA, McHugh MD, Aiken LH. Nurse outcomes in Magnet® and non-Magnet hospitals. J Nurs Adm. 2011; 41: 428–433.
18. Aiken LH, Clarke SP, Sloane DM, Lake ET, Cheney T. Effects of hospital care environment on patient mortality and nurse outcomes. J Nurs Adm. 2008; 38: 223–229.
19. Patrician PA, Shang J, Lake ET. Organizational determinants of work outcomes and quality care ratings among Army Medical Department registered nurses. Res Nurs Health. 2010; 33: 99–110.
20. Judge TA, Thoresen CJ, Bono JE, Patton GK. The job satisfaction-job performance relationship: a qualitative and quantitative review. Psychol Bull. 2001; 127 (3): 376–407.
21. Hayes B, Bonner A, Pryor J. Factors contributing to nurse job satisfaction in the acute hospital setting: a review of recent literature. J Nurs Manag. 2010; 18: 804–814.