Definition of Terms
It is important to note that although the terms team and teamwork are often closely linked in the research literature, they do not always exist in concert within organizations. While teams may be formally constituted, that is not sufficient to ensure that teamwork exists; likewise, teamwork behaviors can emerge without the formal designation of teams. Therefore, it is important to clarify that the research reported herein is about visible teamwork, not necessarily “teams” as often described in the literature.
Although the terms team and work group are often used interchangeably in the literature, it was necessary to distinguish them in this research because the study focused on interactions between workers, or teamwork, which does not necessarily occur within work groups. For the purposes of this study, a work group was defined as “employees working on the same shift who were assigned to the same residents and/or unit.” For example, all employees on the day shift assigned to residents on the first floor of a facility were considered a work group. These employees shared the same workspace, shared the same resources and supplies, and were responsible for a distinct group of residents. However, the work group may or may not have functioned as a team by utilizing teamwork.
For the purposes of this study, teamwork was defined as “working interdependently, coordinating actions effectively to meet the demands of other work group members, exchanging information, interacting dynamically, and working toward a common goal or goals” (adapted from Salas, Burke, & Cannon-Bowers, 2000).
Teams, both formal and informal, exist within larger organizations and thus are both influenced by and potentially influence those contexts. Kozlowski and Ilgen (2006) stated that teams should be thought of as existing in complex organizational systems and that the work of teams is driven by aspects of the broader organizational systems in which they are embedded. This concept was well articulated by Kahn (2005) in a study of teams in two Massachusetts hospitals. He concluded that teams “are microcosms: smaller, easily digestible representations of dynamics at play in their embedding systems” (p. 141). In other words, teams are representative of—and shaped by—the larger organizational contexts in which they are found. These theories, which point to the importance of organizational factors in teamwork, provided the conceptual framework for this study.
From the standpoint of the classic structure-process-outcome paradigm (Donabedian, 1966), organizational factors (e.g., task design, group boundaries) provide the structure through which the process of teamwork is achieved. However, few studies have examined the process of teamwork and the factors necessary for it. This is due, in part, to the multidimensional nature of teamwork, the difficulty defining it, and the difficulty in quantitatively measuring it (Marks, Mathieu, & Zaccaro, 2001). Instead, much previous research has focused on defining the different types of teams found in organizations (Sundstrom, McIntyre, Halfhill, & Richards, 2000) and identifying the structural elements necessary for positive team outcomes (Guzzo & Dickson, 1996), thus neglecting the process aspect of the paradigm.
This study sought to address this gap in the literature by first identifying LTC facilities that were successfully achieving the process of teamwork (i.e., DCWs were working interdependently, coordinating actions, exchanging information) and then identifying the organizational factors (structure) within facilities that enabled high amounts of teamwork among DCWs in some facilities and not others. Because this was a previously unexplored topic, the research questions that guided this inquiry were broad and descriptive:
- What organizational factors facilitate teamwork in LTC facilities?
- What organizational factors act as barriers to teamwork in LTC facilities?
A qualitative case study method was used; analysis of observations and semi-structured interviews were collected as part of a larger study of LTC facilities in eastern Massachusetts (Tyler et al., 2006). First, data from 40 field observations previously conducted in 20 LTC facilities were used to categorize facilities by teamwork level: high, moderate, or low. Next, qualitative analysis of 59 interviews previously conducted with employees from the high- and low-teamwork facilities was used to identify the organizational factors within facilities that acted as facilitators of or barriers to teamwork. Details of these methods are provided below.
Case study methodology (Yin, 1994), which allowed for comparisons to be made between facilities with high and low amounts of teamwork, was used to determine the organizational factors that contributed to high amounts of teamwork in LTC facilities. Purposive sampling allowed for interview data from facilities exhibiting both high and low levels of teamwork to be analyzed through the process of comparative analysis (Miles & Huberman, 1994). An examination of the organizational factors that affect teamwork in LTC facilities required comparison of high- and low-teamwork facilities because facilities differing on this dimension were expected to differ on other dimensions. Similarly, facilities with comparable amounts of teamwork were expected to be alike in important ways.
Observation data previously collected in 20 LTC facilities were used to categorize the facilities by teamwork level: high, moderate, or low. Participating facilities were located throughout eastern Massachusetts and were recruited through professional contacts of the two authors, at industry meetings, and by snowball sampling based on recommendations of the early participants. Twenty-five facilities were invited to participate, and 5 refused for reasons unrelated to the research, such as union organizing within the facility, added stress on facilities due to financial difficulties, and the pending sale of the facility. The refusing facilities ranged in size from 50 to 350 beds. In these facilities, CNA hours per resident day ranged from 2 to 3 (mean = 2.48), 3 were for profit and 2 were nonprofit, 3 were chain affiliated, 3 were located in suburban areas, and 2 were in urban areas.
In 19 of the 20 participating facilities, the two authors conducted observations on two different shifts and/or units. (Due to scheduling conflicts, the first author conducted both observations at one facility.) Due to Institutional Review Board requirements, a specific DCW (nurse or CNA) consented to be observed and was the focus of each observation. These DCWs were selected by convenience from among the DCWs who happened to be working on that given shift. One CNA and one nurse were observed in each facility for 5 to 6 hours each. However, while one DCW was chosen for shadowing during each observation period, the interactions between the DCW and his or her coworkers and those with facility residents were the main focus of the observations. In other words, while one DCW consented for observation, the activities of the entire unit were under observation. Each observation lasted 5 to 6 hours to encompass all of the important events that take place during the shift, including shift change, medication passes, mealtime, and getting residents out of bed or putting them to bed.
Two types of notes were taken during the observations: a general timeline of what was happening on the unit during each observation and notes about specific topics of interest. These topics were formal and informal meetings about resident care, interactions across hierarchical levels (DCWs and managers), performance feedback, staff-resident interaction and feedback, staff interactions concerning the planning/providing of resident services, staff comments regarding the timing/sequencing of work, teamwork and assisting, and questions and comments for interview follow up. After each observation, notes were also taken regarding general impressions and a description of the physical layout of the unit and facility. The use of a semi-structured observation guide and conferencing by the two authors were used to increase reliability across observations.
Building from the definition of teamwork adopted for this study, the facility categorization flow chart (FCFC), shown in Figure 1, was developed and used to categorize the 20 facilities by teamwork level. The FCFC was used to systematize analysis of the observation data by asking questions of the observation notes from each facility regarding both the physical tasks and communication that took place during the observations. In this way, it was methodically determined whether groups of DCWs in each facility consistently used teamwork. For example, Question 3 on the FCFC asks about CNA participation in shift-change meetings. This question was meant to determine not just whether or not information was being exchanged, but also to measure the quality of the information exchange (i.e., whether members were interacting dynamically). Only those facilities in which work groups exhibited all of the elements of teamwork—as defined above—were categorized as high teamwork, and only those exhibiting none of these elements were categorized as low teamwork. Those exhibiting some elements of teamwork were categorized as moderate teamwork and dropped from the interview analysis. Table 1 includes examples of the elements of teamwork and how they were observed.
Figure 1. Facility Categorization Flow Chart.Note. CNA = Certified Nursing Assistant.
Table 1: Factors Differentiating High- and Low-Teamwork Facilities
FCFCs were completed for each of the 20 facilities in this same manner. Four of the 20 facilities, which had affirmative answers for all of the questions on the FCFC, were categorized as high teamwork, and the 5 facilities that had negative answers to all questions on the FCFC were categorized as low teamwork. Eleven of the facilities could not be categorized as either high or low teamwork and were categorized as moderate-teamwork facilities. These facilities exhibited some, but not all, elements found in the high-teamwork facilities. For example, in a moderate-teamwork facility, CNAs may have worked together to provide physical care to residents, but information exchange did not take place. Thus, the high- and low-teamwork facilities represented the two extremes of a teamwork continuum, which allowed for comparative analysis (Miles & Huberman, 1994).
After categorization, Online Survey Certification and Reporting data from the Centers for Medicare & Medicaid Services (CMS) website were reviewed to ensure that high- and low-teamwork facilities were not significantly different in terms of profit status, location, size, or nursing hours. With regard to profit status, both for-profit and nonprofit facilities were among those categorized as high and low teamwork. Specifically, 3 of the 4 high-teamwork facilities were for profit and 1 was nonprofit. Among the 5 low-teamwork facilities, 2 were for-profit and 3 were nonprofit. Two of the high-teamwork facilities were chain affiliated, as were 3 of the low-teamwork facilities.
The locations of facilities were also similar, with both urban and suburban facilities among those categorized as high and low teamwork. With regard to high-teamwork facilities, 2 facilities were located in urban areas and 2 were located in suburban areas. In the low-teamwork group, 2 facilities were in urban areas and 3 were in suburban areas. Regarding size, high-teamwork facilities ranged from 118 to 161 beds; low-teamwork facilities ranged from 113 to 142 beds. Individual units contained 35 to 45 residents.
Finally, no differences existed between facilities categorized as high and low teamwork with regard to CNA hours per resident day or nurse hours per resident day. These numbers represent the daily number of CNA or nursing hours scheduled each day divided by the number of residents at a facility. In other words, these numbers are intended to show how much direct CNA or nurse time each resident could theoretically be receiving based on a facility’s staffing. However, because nurses and CNAs perform other tasks besides direct resident care (e.g., paperwork), this number does not truly represent the hours of hands-on care residents receive (Harrington, Zimmerman, Karon, Robinson, & Beutel, 2000). Again, these numbers were similar across high- and low-teamwork facilities. High-teamwork facilities ranged from 2.1 to 2.9 CNA hours per resident day (mean = 2.40), and low-teamwork facilities ranged from 1.9 to 3.0 (mean = 2.46). Nurse hours per resident day were also similar, with facilities categorized as high teamwork ranging from 1.2 to 1.5 (mean = 1.43) and low teamwork ranging from 1.0 to 1.8 (mean = 1.44).
Interviews and Analysis
Both DCWs and nurse managers were interviewed for this study. Six or seven interviews were conducted in each facility: two with DCWs (one nurse and one CNA) and four or five with managers, depending on the size of the facility, for a total of 59. All interviews were conducted individually. Interviews with managers focused on the design of nursing jobs, including the number of nurses and CNAs employed on each unit; job descriptions for these positions; information about interactions among staff; information about formal and informal opportunities for feedback among CNAs and nurses; as well as general information about the size and structure of individual units. Participants in these interviews included the director of nursing (DON), the assistant director of nursing (ADON), the staff development (education) coordinator, and the unit managers (i.e., all nursing managers whenever possible).
The DCWs interviewed were the same nurses and CNAs who were observed. DCWs were interviewed about their job history, job duties, feedback, and teamwork. In addition to these more structured questions, all informants were asked open-ended questions that allowed them to discuss the issues they believed were important to their work experiences (i.e., “If you could change one thing about the work here, what would it be?” and “What is the hardest thing about your job?”). At the conclusion of the interviews, informants were also invited to discuss any topic they believed was important but not addressed during the interview to that point.
Interviews were analyzed in the order they were originally conducted, beginning with the process of microanalysis (Strauss & Corbin, 1998), where each line was assigned a code. Preliminary codes were created, notes were taken regarding each facility, and a set of master notes were created regarding how each facility’s findings related to the research questions. The first two sets of interviews analyzed were from low-teamwork facilities. Notes were taken regarding the ways in which these two facilities were similar and how these similarities may have been related to the research questions. These early ideas and concepts were then tested against data from the next facility, a high-teamwork facility, using the constant comparative method. The refined set of concepts was then tested against the next facility until all interviews had been analyzed.
Use of the constant comparative method helped ensure reliability of the data analysis. Each new line of interview text was analyzed in terms of the existing coding scheme and labeled with the appropriate code, if one existed. If an appropriate code did not exist, a new code was created. The entire data set was analyzed twice, first to create the codes and their related code groups and a second time so that all interviews could be coded using the full set of codes. Table 2 includes the list of codes and code groups. Although the first author coded all interviews, both authors met regularly to discuss code development and preliminary findings, and the larger research group was involved through discussion of the codes and findings. An audit trail kept track of all decisions about codes, code groups, and how these were applied to all interviews.
Table 2: Codes, Code Groups, and Definitions Used in the Current Study
Data analysis revealed that high- and low-teamwork LTC facilities differed in three organizational areas: management style, training, and feedback and recognition. Specifically, high-teamwork facilities were found to have managers with flexible and pragmatic styles who included DCWs in decision making. They provided individualized training that was based on DCW needs and had systems in place for employee feedback and recognition. These findings are detailed below and illustrated in Figure 2. In contrast, managers in low-teamwork facilities were found to be disciplinarian and authoritarian. Low-teamwork facilities provided the same training to employees new to the field as they did those with more experience, and they did not provide employees with timely feedback or positive recognition. As shown in Table 2, a number of coding categories (those listed as “People” and “Non-Grouped” codes) were not found to be related to teamwork among DCWs in LTC facilities, such as DON tenure, corporate ownership, or primary nursing. For example, resident care organized in a primary nursing model—where nurses are responsible for a resident’s total care (as opposed to a task nursing model where one nurse may provide treatments and another dispense medications to the same resident)—was found in both high- and low-teamwork facilities.
Figure 2. High-Teamwork Model.
Managers in high-teamwork facilities were found to be flexible, pragmatic, interactive, and visible. In contrast, low-teamwork facility managers were found to be disciplinarian and authoritarian (Lewin, Lippitt, & White, 1939). Managers in high-teamwork facilities described supervising DCWs in more practical and pragmatic terms, whereas managers in low-teamwork facilities used terms such as “watching” or “catching” when describing their supervisory styles. For example, managers in low-teamwork facilities repeatedly spoke of DCWs as children and described their roles as supervisors using terms such as “baby sit” and “school teacher.” As the ADON in one low-teamwork facility said after recounting a story about what he regarded as poor behavior by two direct-care nurses:
I (Informant): That’s the kind of stuff…that I hate. That—That’s the worst. Can you imagine doing that? These are adults, they’re grown-ups.
A (indicates either first or second author): Yeah.
I: You know what I mean? It’s like being a second grade school teacher.
Similarly, managers in low-teamwork facilities were seemingly unable to respond to the specific needs of their facilities in the ways that high-teamwork facility managers were able to. Managers in high-teamwork facilities were able to assess the unique needs of their facilities and their employees, such as staffing or training needs. They were able to make changes in their facilities on the basis of these needs and, if necessary, used trial and error to determine the best solution to any problem or issue. As one high-teamwork facility DON explained:
But you need to, you know, you look to see who has strengths in what areas. And who could pick up different tasks that medical records was doing. To tell you the truth, sometimes they stay where they were assigned because it seems to be working better.
On the other hand, low-teamwork facility managers seemed unable to respond to the unique needs of their facilities and their employees. In fact, managers in the low-teamwork facilities were much more likely to report that they believed all LTC facilities were the same. As one low-teamwork facility DON stated: “As far as the overall way things are done, it’s pretty much the same in every building.”
In some cases, the authoritarian styles of low-teamwork managers resulted in animosity between themselves and the DCWs. As one CNA stated: “Like sometimes we get stressed out. Like when you’re caring for the resident and being paged, ‘We need an assistant.’ What do you want us to do? Why don’t you do [care for] the resident or answer the page?”
To avoid this animosity, low-teamwork facility managers employed more middle managers who seemed to act as buffers between themselves and the DCWs. Therefore, they could avoid interacting with DCWs and what they reported as constant complaints from them. As one manager stated:
Well, I have a lead CNA who’s responsible for the CNAs. I give her, um, some tasks and then she divides it up among the other ones. Um, and of course if there’s any problem they come to me and I tell them, “Well, you’re going to do it anyway.”
While senior managers in low-teamwork facilities avoided communicating directly with frontline staff by communicating through various middle managers, unit managers also avoided interacting with DCWs by using written communication and by isolating themselves physically by staying in their offices rather than walking through and around the unit:
We have a sheet that we put up daily with the assignments on them. I write the, the nurse aide’s name on it. And under it is the list of residents that she’ll care for. There’s a printout. Because the assignments really don’t change. They have the same room number, the same ones all the time. Um, along with that, I write down whose shower day it is. Any temperatures that need to be taken, anything special for the day, an appointment. And the sheet goes on the desk for the aides to see in the morning. And then she’ll check that, um, for the day. Um, we have our list of I & Os [intake and output] taken out, to be put on the counter as well. And our ambulation list so they know who to walk, who they need to walk, daily or twice a day.
In contrast, high-teamwork facility managers did not attempt to avoid communicating with DCWs and made rounds on the units or were even assigned regular duties on the units, such as administering medications. High-teamwork facility managers did not use middle managers as buffers between themselves and the DCWs as low-teamwork facility managers did. In fact, they often sought input from DCWs before instituting changes in the facility. One DON described how changes in her facility were decided on: “Sometimes we’ve had to form teams to better see who could better do a job that might come up. And sometimes we have to change tasks.”
Substantial differences were also found between high- and low-teamwork facilities with regard to employee training. High-teamwork facilities employed more experienced staff development (education) coordinators who valued the training of DCWs, geared their training to the needs of individual employees, and mentored new employees. For example, in high-teamwork facilities, the orientation period—both its length and substantive content—was individualized based on the new employee’s previous work experience and prior training. As one high-teamwork facility staff developer described: “For instance I now have two new nurses and we’ve extended their orientation because they are new grads. So, I said ten day [training period], but we do it according to the individual.”
This was not the case among low-teamwork facilities, which provided the same training to experienced DCWs as they did to those new to the field. New employee training generally covered only the topics mandated by the State Department of Public Health and often only involved having DCWs watch videotapes. The following exchange is how a staff developer from a low-teamwork facility described her new employee orientation:
I: It’s a day and a half long…go through, through everything from resident abuse… to infection control, all the mandatory…
A: Um hmm.
I: Blood-borne pathogens. Um, then the second day…it’s a half day…. They watch some tapes…on lifts..the lifts that’s used to get the residents up.
High-teamwork facilities were also found to include a period of mentorship for their new employees that low-teamwork facilities did not. New employees were given time to work on the units with an experienced DCW; this was true for both nurses and CNAs. This period of mentorship seemed to allow the facility to model good teamwork behaviors for new employees and reinforced the fact that teamwork was an expected part of working in the facility. In high-teamwork facilities, the length of this mentorship period varied based on the needs of the new employee, and the new employee was not allowed to work alone until he or she was deemed ready by the facility’s staff developer:
A: So 2 to 6 weeks?
I: Depending on how they feel. What will happen is, I will talk to the scheduler. We set up a mentor with them of CNAs that we know are experienced and that we trust to be…above average CNAs. And we’ll have them work with them. And then after 2 weeks, I will interview the CNA and the preceptor and say, “How do you feel, both of you?” Independent on where they’re both at. It’s all individual, there’s no set standard.
Further, if at some time in the future the new employee was found to be deficient in an area, he or she would be expected to receive further training from the staff developer:
And I also go up on the stairs, up on the floors with them for 3 days, um, showing them how I want things done and how I expect things to be done, and if I see them done any differently, you get to sit with me again.
In low-teamwork facilities, new employees were allowed only short periods working with a mentor, if any mentorship was provided at all. In these cases, low-teamwork facility staff developers expected the new employee’s mentor to train the new employee and even held the mentor responsible if the new employee was later found to have any deficiencies:
I mean, if the, um, CNA goes out and does something wrong, I go back on them [mentors]. Because the CNAs have a checklist and if it’s checked off they know how to do it. If they don’t, I go back on the CNAs that did their orientation.
Related to this last point was the training and experience of the staff developers themselves. In high-teamwork facilities, staff developers were much more experienced in providing training to others; each had more than 4 years of experience in the position. This was not the case in low-teamwork facilities, where the longest tenured staff developer had been in the position 18 months. In fact, in one low-teamwork facility, the staff developer had never even worked in a LTC facility prior to accepting her position, yet she was expected to train others to work in a LTC facility:
A: So you came here into that job, staff development?
A: OK. And had you done staff development anyplace else?
I: No. This was a whole new world for me…long-term care.
A: This is the first time you were in long-term care?
Finally, attitudes about training DCWs were very different between managers in high- and low-teamwork facilities. In high-teamwork facilities, managers stated that providing DCWs with training about the medical conditions of their residents and the symptomology of these conditions would help DCWs perform their jobs better and provide better care to facility residents:
Some people believe the CNAs should be kept in the dark. The CNAs should not know anything about the chart or the patient. I’m just the opposite. I believe that the more they know about the disease process, about why someone has a stroke and how that affects them, the better they’ll be for the resident.
Feedback and Recognition
High- and low-teamwork facilities also differed in the area of employee feedback and recognition. Specifically, high-teamwork facilities were found to have systems in place for providing employee feedback on a regular basis and formal mechanisms for recognizing the work done by DCWs. Low-teamwork facilities did not have such systems, and managers generally described only the mechanisms available for negative feedback or employee discipline.
Three of the four high-teamwork facilities had formal and public ways of recognizing the good work done by DCWs, and the fourth facility was actively planning a program of this kind. In one facility, employees praised each other during their weekly staff meetings, and one employee was named employee of the week. In another facility, written praise for employees was posted throughout the facility, and in another facility employees were recognized at monthly luncheons and again at a yearly banquet. As one unit manager described:
What they do is, um, if they do something, even if we just see them doing something like one of their patients looks a little extra special, if they’re going out and the aide did a little extra, we’ll give them a fish [fish-shaped note for positive comments about employees that are posted on the walls throughout the facility] to say “Thank you for the good care.”
No similar programs were found at low-teamwork facilities and, in fact, when asked about mechanisms for employee feedback during interviews, low-teamwork facility managers were much more likely to mention ways that negative feedback was provided to employees:
Um, I think they get some feedback. Negative. Sometimes, um, I may come up from a meeting. I may be very angry and call the nurses to a closed area to say, “This is what just happened. How did this happen?”
They also spoke mostly of verbal and written warnings:
If it’s not done, if I catch that person not doing it the exact same way, then I’ll give you a verbal warning. And I would make notation that I did give you a verbal warning. That’s how we give feedback.
When managers at low-teamwork facilities did speak of providing positive feedback to employees, they stated that this took place when employees “went above and beyond” or did “extraordinary” work: “Uh, if a if a worker is, uh, has done…an extra-ordinary work, I personally I compliment the person.”
This can be contrasted with a comment from a high-teamwork facility manager:
I try to let them know that they’ve done a good job and I appreciate everything they do. I’ll bring in doughnuts or bagels and it just gives you that little bit of extra time. And, just a thank you and praise them for the work that they do. They have a very hard job.
DCWs themselves also reported receiving little or no feedback in low-teamwork facilities:
A: OK. So, is there any way for you to get feedback?
I: From management, every once in a while, “Oh, you’re doing a great job, you’re doing a great job.” But as far as on a daily thing. Nobody really says nothing to nobody.
The findings of this study, which point to the importance of management style, training, and feedback and recognition for DCW teamwork in LTC facilities, are consistent with the findings of other studies of LTC facilities regarding employee outcomes, such as job satisfaction and turnover among DCWs. Management style, training, and feedback and recognition have all previously been linked to these important outcomes for DCWs. It is therefore possible that teamwork may be the process through which these organizational factors (i.e., structure) affect the outcomes of employee satisfaction and turnover. Although several studies have identified important links between management style, training, and feedback and employee outcomes, none has identified the pathway(s) or process through which these links occur. The findings of this study suggest that teamwork is one possible pathway.
For example, in one study of how management practices affect DCW turnover in LTC facilities, Brannon, Zinn, Mor, and Davis (2002) found that facilities with flatter organizational structures and fewer middle managers were more likely to have very low turnover. Similarly, Scott-Cawiezell et al. (2005) found that better management of LTC facilities was related to lower staff turnover. Another study also found that turnover was related to management style (Anderson, Corazzini, & McDaniel, 2004). In that study, managers who used reward-based administrative styles with open patterns of communication saw much lower turnover among their DCWs. Bowers, Esmond, and Jacobson (2003) found that lack of recognition caused turnover among CNAs. All of these studies found that the same management practices shown to influence teamwork in the current study were similarly shown to affect turnover of DCWs in LTC facilities.
Previous studies have also shown that training is related to employee outcomes in LTC. For example, Shemansky (1997) found that implementing a mentorship program for DCWs in one LTC facility reduced turnover. Hegeman (2005) found similar results in the facilities she examined; those facilities that implemented a peer mentoring program had increased DCW retention in relation to a comparative group of facilities. Chesteen, McClain, and Smith (2003) found that DCW job satisfaction was related to training. Specifically, they found that better trained workers were more able to recognize and apply appropriate standards for themselves as well as others. This is an important finding because it suggests the way in which training may affect teamwork. When the DCW is properly trained, he or she is better able to perform tasks but is also better able to realistically evaluate the work of fellow employees. In this way, the DCW is not disappointed by coworkers’ inability to live up to unrealistic expectations. This may help foster teamwork among adequately trained DCWs.
Unlike studies that have linked employee and resident outcomes with factors that facility managers have no control over, such as facility size or profit status (Harrington et al., 2000; Hillmer, Wodchis, Gill, Anderson, & Rochon, 2005), all of the recommendations provided here are achievable in existing facilities. As demonstrated by the high-teamwork facilities in this study, facilities have the ability to provide better management, training, and feedback to their DCWs. Making these improvements in LTC facilities may result in better teamwork and better outcomes for DCWs.