According to the U.S. Department of Health and Human Services (DHHS), the direct health care workforce has approximately 2.3 million workers with 24.7% of these being nursing assistants (NAs) working in nursing home facilities (Squillace, Remsburg, Bercovitz, Rosenoff, & Branden, 2006). With estimates of a 34% growth in the number of health care support positions needed by 2020, NAs will be one of the fastest growing occupations in America and the fastest growing within health care (U.S. Department of Labor, 2012). A growing shortage of direct care workers compounds the increasing demand, with some facilities reporting position vacancies between 10% and 20% and estimated annual turnover rates as high as 100% (Charney & Schirmer, 2007; Donoghue, 2010; Institute of Medicine [IOM], 2008; Squillace et al., 2006). This high turnover rate is unsustainable as it results in staggering costs to organizations as well as negatively impacting quality patient care (Castle & Engberg, 2007; Donoghue, 2010). With increasing worker shortages, these numbers highlight the critical issue of direct care worker (DCW) recruitment and retention as a key factor in meeting the future health care needs of society’s aging population (U.S. Department of Labor, 2012; IOM, 2008).
The nursing home work environment and tasks typical of the direct care occupations are additional contributing factors to the difficulties of both recruiting and retaining these workers, particularly NAs (Temple, Dobbs, & Andel, 2010). NAs are usually responsible for completing the most physically demanding patient care tasks, including lifting and moving patients. In addition to physically strenuous tasks, NAs are also subject to the intense emotional stressors associated with patient care. Because of the physical demands and emotional strain of their work, combined with the significant staffing shortages and high patient loads, NAs are an occupational group particularly vulnerable to workplace injury, job dissatisfaction, and turnover intent (Schoenfisch & Lipscomb, 2009; Trinkoff, Johantgen, Muntaner, & Le, 2005).
Given these challenges, it is important to investigate the nature of worker injury and determine how cumulative injury contributes to negative outcomes for health care organizations and their employees and patients. Furthermore, it must be determined what factors, if any, can successfully mitigate NA injury rates. As such, the purpose of this study is to explore the relationship between NA injury rates and key outcomes, such as job satisfaction and turnover intent, while also examining if positive workplace factors, such as supervisory support, employee engagement, and injury prevention training, are associated with lower rates of workplace injury and subsequent outcomes.
The Occupational Health and Safety Hazards Experienced by Care Providers
Direct care workers are an occupational subgroup with one of the highest rates of work-derived illness and injury, reporting double the national average of all occupations combined (U.S. Department of Labor, 2010). Within this subgroup, NAs report significantly higher rates of occupational injury than nurses or other health care workers (Alamgir, Cvitkovich, Shicheng, & Yassi, 2007; Pompeii, Lipscomb, Schoenfisch, & Dement, 2009; Rodriquez-Acosta et al., 2009; Schoenfisch & Lipscomb, 2009). Studies have shown that injury within the health care workforce is related to high turnover rates, workers leaving the health care workforce permanently, poor quality of care, and suboptimal patient outcomes (Charney & Schirmer, 2007; de Castro, Hagan, & Nelson, 2006; Stone, 2004).
Direct care workers, particularly NAs, are continuously presented with a variety of workplace factors that create unsafe workplaces and may pose personal risks, including exposure to communicable diseases, musculoskeletal injuries, violence stemming from patient or family assault, psychosocial work stress, and emotional demands (Charney & Schirmer, 2007; Mittal, Rosen, & Leana, 2009; Schoenfisch & Lipscomb, 2009; Stone, Du, & Gershon, 2007). Studies have found that health care workers who perceive an unsafe work environment are less engaged in their jobs and more susceptible to injury, dissatisfaction, and turnover (Castle, Engberg, Mendeloff, & Burns, 2009; Stone et al., 2007). Elements of the work environment have been found to exert a strong influence on health care workers and hold the potential to negatively influence organizational commitment, work performance, workplace safety, and turnover decisions (IOM, 2008; Mittal et al., 2009; Rathert, Ishqaidef, & May, 2009; Stone et al., 2007).
In terms of workplace injuries, patient handling may represent the greatest risk to NAs, as workers in this category are twice as likely to report patient handling injuries as other health care workers (Charney & Schirmer, 2007; Pompeii et al., 2009). Moreover, injuries from moving or transferring patients often lead to organizational and employee losses related to lost productive work time, sick pay, and worker compensation costs (Charney & Schirmer, 2007; Pompeii et al., 2009). A recent study found one-third of all musculoskeletal injuries experienced by employees of a large medical staff were attributable to patient handling, and NAs accounted for a large proportion of the injured workers (Pompeii et al., 2009). Many studies frequently note the ergonomics of lifting/moving/assisting patients as an important safety risk, with many direct care workers worrying this issue may lead to personal injury (Castle et al., 2009; Markkanen et al., 2007; Sherman et al., 2008). Moreover, a small yet growing trend in direct care worker injury derives from the phenomenon of patient-on-care provider violence, with patient violence growing as a source of care provider injury (Galinsky et al., 2010).
As health care institutions are forced to provide more streamlined care, the resultant changes include lower staffing levels and higher patient loads (Castle & Engberg, 2007; Trinkoff et al., 2005). The combination of lower staffing levels and high workloads present safety hazards that may influence NAs to perform duties in ways that contradict or disregard occupational safety training, which in turn may increase negative outcomes for employees such as increased rates of injury (Castle & Engberg, 2007; Trinkoff et al., 2005).
Many studies have shown that the relationship of injury to health care worker job performance is such that instances of injury are negatively associated with perceptions of patient outcomes and quality of care (Aiken, Clarke, Sloane, Sochalski, & Silber, 2002; Castle et al., 2009; Charney & Schirmer, 2007). We argue that the effect of injury on the understudied NA workforce requires greater examination, and employee outcomes such as job satisfaction, turnover intent, and quality of care perceptions are likely to have a negative relationship with injury incidents. Therefore, we offer our first set of hypotheses:
H1: Higher rates of reported NA workplace injuries will be negatively related to (a) job satisfaction, (b) willingness to recommend the facility as a place to work, (c) willingness to recommend the facility as a place to seek care, and positively related to (d) turnover intentions.
National Institute for Occupational Health and Safety: An Agenda for Safer Workplaces
The National Institute for Occupational Health and Safety (NIOSH) has crafted the National Occupational Research Agenda (NORA) to encourage innovative research directed at improving organizational practices and work processes to minimize the negative effects of worker exposure to workplace hazards (Sauter et al., 2002). Within NORA, the Healthcare and Social Assistance Sector is identified as a predominantly high-injury industry and sector research is directed at understanding how to improve the occupational health and safety of the health care provider work force (NORA Healthcare and Social Assistance Sector Council, 2009). The NIOSH/NORA Work Organization Framework identities specific areas of research and organizational directions that focus on ways to minimize workplace injuries (Sauter et al., 2002; see Figure 1). Organization of work incorporates job characteristics (such as safety climate, task attributes, and work roles). Occupational safety and health programs incorporate organizational interventions that promote the health and safety of workers (such as injury prevention training). Physical and psychological hazards reflect workplace factors that potentially contribute to worker injury. As such, the framework offers a simple yet comprehensive guide to investigating how work environments contribute to fostering workplace hazards and what interventions might serve to ameliorate the subsequent risk of injury within health care organizations. Using the framework as a guide, the following sections identify specific areas within the organization of work as well as the safety and health programs that are germane to investigate in order to minimize NA workplace injuries.
Organization of Work
As evidenced above, the organization of direct care workplaces include job characteristics that potentially contribute to the creation of or minimization of workplace hazards. Within the work context, both social-relational aspects of work and worker role/development are two areas identified as positive facets of work organization and are reflected, respectively, by supervisor support and employee engagement.
This study examines the role of supervisor support for NAs, given evidence that supervisor support is an important variable in other health care workforce populations. Studies have found that many of the detrimental effects of poor work environments are ameliorated by the presence of a supportive supervisor (IOM, 2008; Mittal et al., 2009; Rathert & Fleming, 2008). For example, support from supervisors is consistently associated with decreased levels of workplace stress, decreased turnover intent, improved worker safety, and fewer on-the-job injuries (Donoghue, 2010; Lee, Coustasse, & Sikula, 2011). Rathert and Fleming (2008) found that continuous quality improvement leadership (emphasizing interpersonal skills, support, psychological safety) positively moderated the relationship between the work environment and teamwork among acute care nurses, whereas Lee and colleagues (2011) found that NAs whose supervisors exhibited positive leadership qualities were significantly less likely to report workplace injuries or injury-related absenteeism. This evidence supports the proposition that supervisor support is related to positive employee outcomes including reduced workplace injury rates; thus, we offer our second hypothesis:
H2: NAs who report low levels of supervisor support will report more injuries than NAs who report higher levels of supervisor support.
Employee engagement refers to a positive, fulfilling state of mind characterized by vigor, dedication, and absorption in the workplace (Schaufeli, Taris, & van Rhenen, 2008) and is frequently conceptualized as the opposite of employee burnout. Engaged employees are characterized by energy, involvement, and effectiveness, whereas burnt-out employees are characterized by exhaustion, cynicism, and inefficacy (Laschinger & Leiter, 2006; Maslach & Leiter, 2008). Engagement encompasses an employee’s emotional and intellectual commitment to an organization, the amount of discretionary effort an employee is willing to perform at work, as well as the employee’s perception of organizational and supervisory support for personal fulfillment and growth (Schaufeli et al., 2008). Engagement has been shown to be a significant predictor of job satisfaction and turnover intent (Schaufeli et al., 2008), two key outcomes that have also been linked with occupational injury (Castle et al., 2009; Stone et al., 2007). Moreover, care providers who report working in a negative environment (e.g., poor leadership, poor teamwork) are more likely to report lower levels of engagement and higher rates of injury, job dissatisfaction, and turnover intentions (Laschinger & Leiter, 2006; Stone et al., 2007). On the other end of the engagement spectrum, studies have offered evidence of a relationship between health care worker burnout and poor patient outcomes including mortality rates, infection rates, adverse events, and decreased patient satisfaction (Aiken et al., 2002; Halbesleben, Wakefield, Wakefield, & Cooper, 2008; Laschinger, & Leiter, 2006).
We conceptualize employee engagement using Macey and Schnieder’s (2008) framework that articulates three types of employee engagement: (a) trait engagement reflects one’s individual views of life and includes personality, affect, and conscientiousness; (b) state engagement reflects feeling of workplace involvement and includes autonomy, commitment, empowerment, and satisfaction; and (c) behavioral engagement reflects extra-role behavior in the workplace such as organizational citizenship and role expansion. Here we focus on employee state engagement because psychological state is a factor that is subject to environmental priming; thus, organizations and management have the capacity to influence employee state engagement (Macey & Schnieder, 2008; Wright, Cropanzano, & Meyer, 2004). In alignment with the NIOSH/NORA Work Organization Framework (Sauter et al., 2002), we argue that state engagement should result in more vigilance and focus, which should be related to reduced injuries. Therefore, we offer our third hypothesis:
H3: NAs who report lower levels of employee engagement will report higher rates of injury than NAs who report higher levels of employee engagement.
Occupational Safety and Health Programs
Despite the prevalence of workplace injuries, the number of quality studies describing injury prevention or injury reduction strategies for NAs in nursing homes is limited. Many of these studies have omitted findings related to NAs as a specific group of care providers (Dawson et al., 2007). A recent systematic review of injury prevention interventions found only 16 studies that analyzed the NAs and RNs separately and concluded that a tremendous need exists for more high-quality research to accurately examine the effects of interventions designed to prevent injuries among all care providers, including NAs (Dawson et al., 2007).
Furthermore, investigators have examined the effect of various training modalities and mechanical interventions on reducing work-related injuries among nurses and NAs with mixed results. Evanoff, Wolf, Aton, Canos, and Collins (2003) evaluated the effect of mechanical patient lifts on reducing musculoskeletal injuries in both hospitals and long-term care facilities (LTC). Overall, the use of mechanical lifts seemed to reduce the severity of the work-related injuries but not the quantity of these injuries. However, Collins, Wolf, Bell, and Evanoff’s (2004) study of workplace training interventions showed a reduction in worker compensation claims post intervention as well as the incidence of repeat injuries. The small number of studies, coupled with the varying types of interventions and inconsistent findings, underscores the need for additional studies identifying elements of the work environment that help mitigate injuries among NAs.
Occupational health and safety reviews in other industries have identified a significant relationship between safety training and safety outcomes such as safety knowledge, safety performance, and hazard awareness (Burke et al., 2006, 2011; Cohen & Colligan, 1998). These findings suggest that interventions and training targeted at improving direct care worker knowledge, skills, and abilities regarding workplace safety should be effective in reducing injury rates. Within the NIOSH/NORA Work Organization Framework (Sauter et al., 2002), occupational safety and health programs represent safety training and the evidence supports the proposition that safety training will be positively related to lower employee injury rates. Thus, we offer our final hypothesis:
H4: NAs who rate their injury prevention training as poor/fair will report higher injury rates than NAs who rate their training as good/excellent.
In summary, empirical research has shown relationships among injury and poor individual and organizational outcomes in many worker populations. This study set out to examine these relationships in NAs working in nursing homes. In addition, this study aims to examine relationships among some important work environment variables and outcomes in this population.
Data and Sampling
Data for this study are from the 2004 National Nursing Assistant Survey (NNAS), the first national probability survey of certified NAs sponsored by the U.S. DHHS and conducted by the National Center for Health Care Statistics (U.S. DHHS, 2004). The survey was conducted using stratified, multistage, probability sampling and is a supplement to the National Nursing Home Survey (NNHS; U.S. DHHS, 2004). The 2004 NNHS/NNAS was the seventh cycle in this series of nationally representative surveys of U.S. nursing homes, their services, their staff, and their residents.
NAs employed at each participating facility were randomly selected to participate in interviews, and all interviews were conducted via telephone. Interviews provided data on a wide variety of work-related and demographic factors. The survey instrument included sections on recruitment, job history, education/training/licensure, management and supervision, organizational commitment and job satisfaction, workplace environment, work-related injuries, and demographics (see U.S. DHHS, 2004, for complete data and sampling information). Items used in these health care workforce studies have been validated across multiple workforces and multiple surveys.
Number of Injuries
NAs were asked to indicate the number of times they had been injured at the facility where they are employed in the previous year using open ended coding. For the purpose of our analysis, the responses were then recoded into four categories: (a) no injuries, (b) one injury, (c) two injuries, and (d) three or more injuries.
Overall Rating of Training
A single item was used to measure training, “How well did your initial NA training prepare you to prevent work injuries?” The response scales were 1 = poor/fair, 2 = good, and 3 = excellent.
This scale was developed using 10 items about supervisor supportiveness (Christian, Bradley, Wallace, & Burke, 2009), which included questions such as the following: My supervisor: (a) …is supportive of progress in my career, such as further training; (b) …helps me with my job tasks when help is needed; (c) …listens to me when I am worried about a resident’s care; and (d) …tells me when I am doing a good job (see Squillace et al., 2006, for a complete list of the survey questions). The 4-point response scale included the following: 1 = strongly disagree, 2 = somewhat disagree, 3 = somewhat agree, and 4 = strongly agree. Factor analysis was used to confirm the dimensionality of the 10 items, one factor was extracted (eigenvalue > 1), accounting for 55.71% of the variance. KMO (.95) and Bartlett’s test of sphericity (p <.001) indicate acceptable data value distribution and multivariate normality; coefficient alpha = .91.
Employee State Engagement
Following Macey and Schneider’s (2008) employee engagement framework, we identified survey questions that align with state engagement (workplace attributes that include involvement, commitment, empowerment, and satisfaction). Five items measuring the level of employee state engagement included (a) I am appropriately respected or rewarded by my nursing facility for my work; (b) I am involved in challenging work; (c) I have a chance to gain new skills and knowledge on the job; (d) I have the opportunity to work in teams; and (e) I am trusted to make resident care decisions. The 4-point response scale included the following: 1 = strongly disagree, 2 = somewhat disagree, 3 = somewhat agree, and 4 = strongly agree. Factor analysis was used to confirm the dimensionality of the five items; one factor was extracted (eigenvalue > 1) accounting for 44.99% of the variance. KMO (.76) and Bartlett’s test of sphericity (p <.001) indicate acceptable data value distribution and multivariate normality; coefficient alpha = .69.
Worker and Organizational Outcomes
The two worker outcome measures were job satisfaction and turnover intent. The two organizational outcome measures were willingness of the NA to (a) recommend her or his facility as a place to work and (b) recommend her or his facility as a place to receive care. Each of the above variables was measured using a single item as follows: Overall job satisfaction was measured using a 4-point scale (1 = extremely dissatisfied, 2 = somewhat dissatisfied, 3 = somewhat satisfied, 4 = extremely satisfied), turnover intent was measured on a 3-point scale (1 = not at all likely, 2 = somewhat likely, 3 = very likely), and finally, willingness to recommend their facility as a place to work and as a place to seek care were both measured using a 4-point scale (1 = definitely not recommend, 2 = probably not recommend, 3 = probably recommend, 4 = definitely recommend).
The scarcity of research about the nature of work and NAs in nursing homes was the impetus to test the need to control for the range of likely influential demographic and organizational factors. The following were identified as being important control variables: age (1 = less than 25 years, 2 = 25–34 years, 3 = 35–44 years, 4 = 45–54 years, 5 = 55 years or above), race (1 = Caucasian, 2 = African American, 3 = Hispanic, 4 = all others), marital status (0 = married/partner, 1 = single/widow/separated/divorced), experience (1 = less than 1 year, 2 = 1 to less than 2 years, 3 = 2 to 5 years, 4 = 6–10 years, 5 = 11–20 years, 6 = more than 20 years), and education (1 = some school, 2 = high school graduate, 3 = some college, 4 = college graduate; Castle, Engberg, Anderson, & Men, 2007; Castle et al., 2009; IOM, 2008; Stone, & Wiener, 2001). Whether the NA has paid sick days (0 = no, 1 = yes) and is participating in the organization’s health insurance plan (0 = no, 1 = yes) were also included. Finally, organizational characteristics found to have a significant influence on organization/worker outcome relationships and workforce retention in other direct care worker studies (Castle et al., 2009; Temple et al., 2010) include profit status (0 = for-profit, 1 = all others) and size of the organization (1 = less than 50 beds, 2 = 50–99 beds, 3 = 100–199 beds, 4 = 200 or more beds).
As part of the analysis, two regression approaches were used. The first utilized hierarchical regression analysis (Tabachnick & Fidell, 2001) to determine the relationships between NA injury and worker outcomes (job satisfaction and turnover intent) and between NA injury and organizational outcomes (recommendations for care and job). Hierarchical regression is a useful technique to examine relationships among the variables while delineating the effects of the control, independent, and dependent variables. The second regression utilized multinomial logistic regression (MLR) to examine the relationship between the reported number of injuries and NAs’ ratings of their supervisor support, their level of engagement, and their workplace injury-prevention training. MLR is a useful technique because it analyzes responses in multiple categories (categories of injuries) and it is well suited to this type of analysis (Tabachnick & Fidell, 2001). SPSS version 20.0 was used for the analysis.
A total of 3,017 NAs from 592 nursing homes participated in the survey, yielding a 71% response rate. Table 1 shows the basic descriptive statistics of the variables and correlations. Approximately 92% of the sample are female, and 50.5% reported being married or living with a partner. Fifty-four percent of the sample are Caucasian, 30% are African American, and 10% are Hispanic. Forty-five percent of the sample has completed high school, whereas 24.5% have some or have completed a college education. The mean age of the sample is 36.9 years. For facility size, 38.4% of the facilities are in the 50- to 99-bed range whereas 48.1% are in the 100- to 199-bed size range. Finally, 58% of the facilities operate as “for-profits.”
Tables 2 and 3 offer detailed information as to the reported type of injury as well as the reported source of injury. Scratches/cuts/open wound injury type are the leading reported cause of NA injury yet there are numerous reports of more severe injury types including back injuries, bruises, and bites (see Table 2). Many NAs with a single injury reported experiencing either a back injury or a non-back muscle sprain. As injury frequency increases, there is a greater prevalence of human bite injuries. Table 3 offers the reported injury source, with provision of patient/resident care activities or patient/resident aggression being the dominant sources of most reported NA injuries.
In terms of the number of injuries, 37.9% of NAs did not report any workplace injuries, (18.9%) reported having had one injury, 12.5% reported two injuries, and 23.9% reported three or more workplace injuries. More than half rated their workplace injury prevention training as “excellent” (48.1%) or “good” (37.2%). Table 4 shows while 55.7% of the NAs without an injury rated their training as “excellent,” only 43.8% of the NAs who had been injured three times or more rated their injury prevention training as “excellent.” In other words, as injury quantity increased, ratings of injury training effectiveness decreased.
Significant associations of the control variables differ by dependent variables (see Table 5), and only the control variables having significant relationships with three or four of the dependent variables are highlighted here. Having paid sick leave is positively related to job satisfaction (β = .096, p < .01), willingness to recommend facility for job (β = .080, p < .01), and for care (β = .071, p< .05), while being negatively related to turnover intent (β = −.081, p < .01). Years of experience as a NA are negatively related to job satisfaction (β = −.070, p < .05), and willingness to recommend facility for job (β = −.079, p < .05), whereas being positively related to turnover intent (β = −.087, p < .01).
A series of four hierarchical linear regression models were run to test the hypotheses that NAs’ reported injuries would have a significant relationship with their job satisfaction, turnover intent, willingness to recommend their facility as a place to work, and willingness to recommend their facility as a place to seek care services, above and beyond the control variables. Regression analysis showed that the number of injuries reported by NAs is: negatively related to (H1a) job satisfaction (β = −.223, p < .001), (H1b) willingness to recommend their agency as a place to work (β = −.176, p < .001), and (H1c) as a place to seek care (β = −.141, p < .001), and positively related to (H1d) turnover intent (β = .108, p < .001; Table 5). Therefore, all of the H1 hypotheses were supported.
MLR was employed to examine the relationship between injury frequency and ratings of supervisor support, employee engagement, and injury prevention training, Results show negative and significant relationships between workplace injuries and NA ratings of supervisor support, employee engagement, and injury prevention training (Table 6). Note the referral group for all of following analysis is the noninjured group.
Analysis indicated that injury prevention training is negatively and significantly related to being injured twice (B = −.215, exp β = .807, p = .021) as well as being injured 3+ times (B = −.259, exp β = .771, p = .001). As injury prevention training ratings decline, the odds of being injured a second time (compared to not being injured) increase by 1.24 (1/expB) and the odds of being injured 3+ times increases by 3.86 (Tabachnick & Fidell, 2001). There was no significant relationship with the single injury category. It appears that training is not associated with one injury, but it is associated with two or more injuries. Therefore, H4 was partially supported.
With regard to supervisor support, the regression results show that supervisor support is significantly related to being injured once (B = −.322, exp β = .724, p = .001) as well as being injured 3+ times (B = −.496, exp β = .609, p< .001). As supervisor support decreases, the odds of being injured once (compared to not being injured) increase by 3.11 (1/expB) and the odds of being injured 3+ times increases by 2.02. There was no significant relationship with the category of two injuries. Therefore, H2 was partially supported.
The MLR results also show that employee engagement is significantly related to being injured twice (B = −.348, exp β = .706, p = .006) as well as being injured 3+ times (B = −.368, exp β = .692, p< .001). As employee engagement decreases, the odds of being injured twice (compared to not being injured) increase by 2.87 (1/expB) and the odds of being injured 3+ times increases by 2.72. There was no significant relationship with the single injury category. Therefore, H3 was partially supported.
Overall, these findings offer partial support for the hypotheses, that is, NAs who rate their injury prevention training poor, supervisor support low, and/or report low employee engagement are at increased odds of reporting multiple workplace injuries—ranging from 1.24 to 3.86 times more likely.
Using the NIOSH/NORA Work Organization Framework for Occupational Illness and Injury (Sauter et al., 2002) as a basis, this study examined a number of occupational factors involving worker injuries and worker/organizational outcomes among NAs employed in nursing homes. The current study replicated findings of prior studies that have linked work environment variables to worker and organizational outcomes in other health care workforces among NAs (Benjamin & Matthias, 2004; Blair & Glaister, 2005; Burke et al., 2006; Dawson & Surpin, 2001; Yamada, 2002). In addition, this study found new evidence that supervisor support and employee engagement may play key roles in NA injuries, job satisfaction, and commitment. In a workforce noted for its high worker injury rates, this study found that NAs who report higher levels of supervisor support and engagement tended to experience fewer injuries. Individual and organizational implications will be considered below.
As hypothesized, reporting more injuries was associated with lower job satisfaction, higher turnover intentions, and less willingness to recommend the facility as a place to work or as a place to seek care. These findings suggest that, not surprisingly, injuries have negative implications for nursing homes as well as NA individual well-being. In addition, however, the study found that supervisor support and state engagement were negatively related to higher odds of reporting more injuries.
NAs who reported more supportive supervisors were significantly less likely to be injured, and as supervisor support decreased the odds of being injured once compared with not being injured at all increased by 3.11. Supervisor support could play a role in helping NAs recall and implement safety procedures during their work, or supervisors could intervene with additional resources if needed to utilize the safety protocols. It is notable that the number of injuries inflicted by patients is high, particularly among NAs who reported three or more injuries. It could be that greater supervisor support plays a role in mitigating interpersonal conflicts that can escalate into patient aggression. Providing greater supervisor support and increased worker training may help reduce worker stress and injury risks when dealing with patients with behavioral problems or cognitive impairment (Benjamin & Matthias, 2004; Dawson & Surpin, 2001; Yamada, 2002). In other LTC settings, supervisor support has been shown to reduce employee turnover and enhance care provided to residents (Dawson & Surpin, 2001; Stone & Dawson, 2008; Stone & Weiner, 2001). Our findings that supervisor support is related to fewer injuries add to this literature as it seeks to identify independent variables in these relationships.
Importantly, employee engagement was also found to be related to NA injuries, in that those who were less engaged reported experiencing more injuries. This is an important new finding, as previous employee engagement research has yet to investigate the relationship of engagement to injury. Prior research has shown that engagement is related to individual and organizational outcomes such as job satisfaction, organizational commitment, and turnover intentions (Saks, 2006; Schaufeli et al., 2008). In one study, acute care nurses who reported higher levels of engagement also reported that their work units were more patient-centered, and there were fewer patient safety problems on their units (Rathert et al., 2009). That state engagement is related to fewer injuries suggests that NAs who are more focused and vigilant, and likely less burned-out, could avoid situations that lead to injuries. Identifying engagement as a mitigating factor could lead to organizational interventions that increase engagement, such as job redesign and increasing levels of autonomy. Given that nursing homes will continue to struggle to increase structural resources such as staffing and updated equipment, gaining understanding of how to increase NA state engagement seems imperative and may result in improvement in some important outcomes.
It is notable that all of our regression models found significant relationships among the independent variables (IVs) and either being injured once compared with not at all or being injured three times or more, but there was variability in relationships with the middle categories. This suggests that there are numerous dimensions related to worker injury among NAs, and learning from experience may combine with supervisor support or engagement to influence subsequent injury prevalence. For example, recall that NAs who reported more engagement were less likely to be injured two or three times or more, but there was no difference between those who were injured once compared with not at all. It could be that an experiential learning component is at play, in that those who are more engaged may get injured once, but their vigilance and focus facilitates their learning from the situation to the extent that they have fewer injuries in the future.
This study’s findings mirror studies with similar findings in other direct care worker populations that link injuries to poor individual and organizational outcomes (Benjamin & Matthias, 2004; Dawson & Surpin, 2001; Yamada, 2002). This suggests that these findings are robust and likely to be generalizable. If health care work environments can be designed to be more supportive and engaging, this may lead to fewer on-the-job injuries and thus help to curb high turnover rates in this worker population.
Implications for Management
The future challenges of providing care to an aging population include the dual problem of workforce shortages coupled with difficult working conditions that impede employee well-being (Blair & Glaister, 2005; Dawson & Surpin, 2001; IOM, 2008; Stone & Dawson, 2008; Stone & Wiener, 2001; Yamada, 2002). As occupational health and safety studies have shown, worker injuries are detrimental to the worker, the organization, and the organization’s performance outcomes (Burke et al., 2006; Lee et al., 2011; Vredenburgh, 2002). When direct care workers are injured, organizations bear the costs of workers’ compensation, replacement staff, overtime, and sick pay among other expenses. Reducing these costs will benefit the organization directly (injury costs and expenses) and indirectly (quality of care, referrals, and reputation). This study showed that the link between NA injury and negative outcomes for individuals and their organizations is significant. Given that health and safety programs have been shown to achieve positive worker outcomes and reduce the incidents of injury (Burke et al., 2006; Stone & Wiener, 2001; Vredenburgh, 2002), organizations will benefit financially from investing in creating effective occupational health and safety training programs, coupled with supervisor support and work environments that are engaging.
Nursing homes that pursue an employee development agenda with NA training as the foundation will most certainly reap the benefits of enhanced worker satisfaction, decreased stress, and decreased turnover (Blair & Glaister, 2005; Dawson & Surpin, 2001; Stone & Wiener, 2001). However, our results suggest that just as important may be managing the work environment so that NAs feel more supported by their supervisors and more engaged in their work. Engagement is consistently associated with organizational commitment (Rathert et al., 2009), and commitment has been shown to be more strongly related to turnover than is job satisfaction (Gees, Manojlovich, & Warner, 2008). Engagement may be a particularly important variable to cultivate in NAs, given that people pursue care work as a calling or vocation more than simply a job (Rathert et al., 2009).
Limitations and Future Directions
There are several limitations to this study. First, the data derived from the 2004 NNAS is several years old and may not represent current NAs or the current state of LTC workplaces. However, despite this potential limitation, this survey is the first national probability survey of certified NAs, offering a rich data set from which to understand the work experience of NAs to improve recruitment and retention of the paraprofessional LTC workforce. A key value in using this data lays in the National Center for Health Statistics use of this questionnaire as a basis for collecting survey data on other care provider workforces, essentially creating a nationally representative, comparative repository of health care provider workforce surveys.
Next, our findings are derived from a secondary analysis of a national survey. The study design and indices used in the survey were not specifically targeted at evaluating the effects of worker injury and training (U.S. DHHS, 2004). For example, NAs rated the effectiveness of their training in preventing injury, but there is no associated time line as to when they received the training or at what location, as such there is no way to rule out the influence of other endogenous factors in the relationships examined in this study. As such, although the findings are relevant and important, they may not be generalizable outside of this study. Moreover, the design was cross-sectional, which prohibits causal inferences. Specifically to the training rating/injury probability relationship, future research over multiple time periods and using multiple data sources will help establish the direction and temporal ordering of these relationships and to separate the substantive relationships from common method variance. In addition, the constructs in this study share a common method of measurement, and as such, common method bias could be a concern resulting from the single source data (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). However, the differentiated relationships between the variables reduce the possibility of this common method bias.
Further research is needed to specifically examine the relationship of worker injury to workplace outcomes for NAs, as direct relationships between injury and worker outcomes are understudied (IOM, 2008; Stone & Wiener, 2001). It is difficult to determine the temporal ordering of training evaluation and injury; are post injury perceptions influenced by the injury itself? Before and after injury data gathering would be optimal to tease out the role of injury on perceptions postinjury. Another interesting question that arises from our finding is the relationship of paid sick leave and all of the dependent variables. The provision of paid sick leaves appears to have a role in job satisfaction, willingness to recommend the facility and in lower turnover intentions. Thus, it suggests that further examination of sick leave with our care providers and care facilities may offer insightful evidence to foster strategic HR policies for the direct care workforce. Importantly, answering such research questions should include examination of what types of work environments and variables influence employee engagement. Additional research is also needed to better understand the most efficacious types of worker training, to determine what forms of safety specific training reduce incidents of worker injury (Alamgir et al., 2007), and how safety training is translated into on-the-job activities and behaviors. Researchers should pursue comprehensive intervention studies targeting worker injury reduction that have stakeholder support and a likelihood of success. Incorporating mixed methodologies, such as combining qualitative and quantitative data, will provide the opportunity for richer exploration of the relationships between workplace injury and organizational outcomes and longitudinal designs with multisource data would provide stronger tests of the hypotheses.
The NIOSH/NORA Work Organization Framework for Occupational Illness and Injury (Sauter et al., 2002) offers a framework from which to develop occupational health and safety interventions tailored to the specific work environment. The framework can be used as a tool for minimizing workplace hazards by enhancing the work environment and subsequently reducing worker injury. This study offers empirical support for the framework in the nursing home direct care worker population and provides a guideline to improve the quality of the work environment for NAs. It is evident from the findings that employee safety training is related to lower reported employee injury rates but is not enough. Our findings also highlight the importance of addressing work environment factors such as supervisor support and employee engagement to help reduce the risk of employee injury. Health care organizations pursuing an improved work environment are more likely to become an employer of choice and foster greater job satisfaction and lower turnover intentions among their NA resource pool.
We would like to thank Nick Turner and the anonymous reviewers at Academy of Management 2011 for their helpful suggestions and comments on earlier drafts of this manuscript.
Aiken L. H., Clarke S. P., Sloane S. P., Sochalski J., Silber J. H. (2002). Hospital nurse staffing and patient mortality, nurse burnout, and job dissatisfaction. Journal of American Medical Association, 288, 1987–1993.
Alamgir H., Cvitkovich Y., Shicheng Y., Yassi A. (2007). Work-related injury among direct care occupations in British Columbia, Canada. Occupational and Environmental Medicine, 64, 769–775.
Benjamin A. E., Matthias R. E. (2004). Work-life differences and outcomes for agency and consumer-directed home-care workers. The Gerontologist, 44, 479–488.
Blair C. E., Glaister J. E. (2005) Making the most of nursing homes’ number one resource. Health Care Management Review, 30 (2), 110–115.
Burke M. J., Salvador R. O., Smith-Crowe K., Chan-Serafin S., Smith A., Sonesh S. (2011). The dread factor: How hazards and safety training influence learning and performance. Journal of Applied Psychology, 96 (1), 46–70.
Burke M. J., Sarpy S. A., Smith-Crowe K., Chan-Serafin S., Salvador R. O., Islam G. (2006). Relative effectiveness of worker safety and health training methods. American Journal of Public Health, 96, 315–324.
Castle N. G., Engberg J. (2007). The influence of staffing factors on quality of care in nursing homes. Health Services Research, 42, 1822–1847.
Castle N. G., Engberg J., Anderson R., Men R. (2007). Job satisfaction of nurse aides in nursing homes: Intent to leaver and turnover. The Gerontologist, 47 (2), 193–204.
Castle N. G., Engberg J., Mendeloff J., Burns R. (2009). A national view of workplace injuries in nursing homes. Health Care Management Review, 34 (1), 92–103.
Charney W., Schirmer J. (2007). Nursing injury rates and negative patient outcomes—Connecting the dots. American Association of Occupational Health Nurses Journal, 55 (11), 470–475.
Christian M. S., Bradley J. C., Wallace J. C., Burke M. J. (2009). Workplace safety: A meta-analysis of the roles of person and situation factors. Journal of Applied Psychology, 94 (5), 1103–1127.
Collins J. W., Wolf L., Bell J., Evanoff B. (2004). An evaluation of a “best practices” musculoskeletal injury prevention program in nursing homes. Injury Prevention, 10 (4), 206–211.
Dawson A. P., McLennan S. N., Schiller S. D., Jull G. A., Hodges P. W., Stewart S. (2007). Interventions to prevent back pain and back injury in nurses: A systematic review. Occupational and Environmental Medicine, 64 (10), 642–650.
de Castro A. B., Hagan P. B., Nelson A. (2006). Prioritizing safe patient handling: The American Nurses Associations handle with care campaign. The Journal of Nursing Administration, 36 (7/8), 363–369.
Donoghue C. (2010). Nursing home staff turnover and retention: An analysis of national level data. Journal of Applied Gerontology, 29, 89–106.
Evanoff B., Wolf L., Aton E., Canos J., Collins J. (2003). Reduction in injury rates in nursing personnel through introduction of mechanical lifts in the workplace. American Journal of Industrial Medicine, 44 (5), 451–457.
Galinsky T, Feng H. A., Streit J., Brightwell W., Pierson K., Parsons K., Proctor C. (2010). Risk factors associated with patient assaults of home healthcare workers. Rehabilitation Nursing, 35 (5), 206–215.
Gees E., Manojlovich M., Warners S. (2008). An evidence-based protocol for nurse retention. Journal of Nursing Administration, 38 (10), 441–447.
Halbesleben J. R. B., Wakefield B. J., Wakefield D. S., Cooper L. B. (2008). Nurse burnout and patient safety outcomes. Western Journal of Nursing Research, 30 (5), 560–577.
Institute of Medicine. (2008). Retooling for an aging America: Building the health care workforce. Washington, DC: National Academy Press.
Laschinger H. K. S., Leiter M. P. (2006). The impact of nursing work environments on patient safety outcomes: The mediating role of burnout/engagement. Journal of Nursing Administration, 36, 259–267.
Lee D., Coustasse A., Sikula A. (2011) Transformational leadership and workplace injury and absenteeism: Analysis of a National Nursing Assistant Survey. Health Care Management Review, 36 (4), 380–387.
Macey W. H., Schneider B. (2008). The meaning of employee engagement. Industrial and Organizational Psychology, 1, 3–30.
Markkanen P., Quinn M., Galligan C., Chalupka S., Davis L., Laramie A. (2007). There’s no place like home: A qualitative study of the working conditions of home health care providers. Journal of Occupational and Environmental Medicine, 49, 327–337.
Maslach C., Leiter M. P. (2008) Early predictors of job burnout and engagement. Journal of Applied Psychology, 93, 498–512.
Mittal V., Rosen J., Leana C. (2009). A dual-driver model of retention and turnover in the direct care workforce. The Gerontologist, 49, 623–634.
Podsakoff P. M., MacKenzie S. B., Lee J., Podsakoff N. P. (2003). Common variance biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88, 879–903.
Pompeii L. A., Lipscomb H. J., Schoenfisch A. L., Dement J. M. (2009). Musculoskeletal injuries resulting from patient handling tasks among hospital workers. American Journal of Industrial Medicine, 52, 571–578.
Rathert C., Fleming D. A. (2008). Hospital ethical climate and teamwork in acute care: The moderating role of continuous quality improvement leadership. Health Care Management Review, 33 (4), 323–331.
Rathert C., Ishqaidef G., May D. R. (2009). Improving work environments in health care: Test of a theoretical framework. Health Care Management Review, 34 (4), 334–343.
Rodriguez-Acosta R. L., Richardson D. B., Lipscomb H. J., Chen J. C., Dement J. M., Myers D. J., Loomis D. P. (2009). Occupational injuries among aides and nursing in acute care. American Journal of Industrial Medicine, 52, 953–964.
Saks A. (2006). Antecedents and consequences of employee engagement. Journal of Managerial Psychology, 21, 600–619.
Sauter S. L., Brightwell, W. S., Colligan, M. J., Hurrell, J. J., Katz, J. M., LeGrande, D. E., … Tetrick, L. E. (2002). The changing organization of work and the safety and health of working people: Knowledge gaps and research directions
. NIOSH Publication No. 2002-116. Retrieved from http://www.cdc.gov/niosh/docs/2002-116/pdfs/2002-116.pdf
Schaufeli W. B., Taris T. W., van Rhenen W. (2008). Workaholism, burnout, and work engagement: Three of a kind or three different kinds of employee well-being? Applied Psychology: An International Review, 57 (2), 173–203.
Schoenfisch A. L., Lipscomb H. J. (2009). Job characteristics and work organization factors associated with patient-handling injury among nursing personnel. Work, 33, 117–128.
Sherman M. F., Gershon R. R. M., Samar S. M., Pearson J. M., Canton A. N., Damsky M. R. (2008). Safety factors predictive of job satisfactyion and job retention among home health aides. Journal of Occupational and Environmental Medicine, 50, 1430–1441.
Stone P. W., Du Y., Gershon R. R. M. (2007). Organizational climate and occupational health outcomes in hospital nurses. Journal of Occupational and Environmental Medicine, 49, 50–58.
Stone R. I. (2004). The direct care worker: The third rail of home care policy. Annual Review of Public Health, 25, 521–537.
Stone R. I., Dawson S. L. (2008). The origins of better jobs better care. The Gerontologist, 48, 5–13.
Tabachnick B. G., Fidell L. S. (2001). Using multivariate statistics. Fourth edition. Boston, MA: Allyn and Bacon.
Temple A., Dobbs D., Andel R. (2010). The association between organizational characteristics and benefits offered to nursing assistants: Results from the National Nursing Home Survey. Health Care Management Review, 35 (4), 324–332.
Trinkoff A. M., Johantgen M., Muntaner C., Le R. (2005). Staffing and worker injury in nursing homes. American Journal of Public Health, 95, 1220–1225.
U.S. Department of Health and Human Services. (2004). National Nursing Home Survey: The national nursing assistant survey (NNAS).
Retrieved from http://www.cdc.gov/nchs/nnas.htm
U.S. Department of Labor. (2010). Incidence rates of nonfatal occupationalinjuries and illnesses by case type and ownership, selected industries, 2010.
Bureau of Labor Statistics. Retrieved from http://www.bls.gov/news.release/osh.t01.htm
Vredenburgh A. G. (2002). Organizational safety: Which management practices are most effective in reducing employee injury rates? Journal of Safety Research, 33, 259–276.
Wright T. A., Cropanzano R., Meyer D. G. (2004). State and trait correlates of job performance: A tale of two perspectives. Journal of Business and Psychology, 18, 365–383.
Yamada Y. (2002). Profile of home care aides, nursing home aides, and hospital aides: Historical changes and data recommendations. The Gerontologist, 42 (2), 199–206.