The Gerontologist

Temkin-Greener,, Helena;Szydlowski,, Jill;Intrator,, Orna;Olsan,, Tobie;Karuza,, Jurgis;Cai,, Xueya;Gao,, Shan;Gillespie, Suzanne, M

doi: 10.1093/geront/gny174pmid: 30657887

Abstract Background and Objectives Previous studies have shown that staff perception of team effectiveness is related to better health outcomes in various care settings. This study focused on the Veterans Health Administration’s Home-Based Primary Care (HBPC) program. We examined variations in HBPC interdisciplinary teamwork (IDT) and identified modifiable team and program characteristics that may influence staff perceptions of team effectiveness. Research Design We used a broadly validated survey instrument to measure perceived team effectiveness, workplace conditions/resources, group culture, and respondents’ characteristics. Surveys were initiated in January and completed in July, 2016. Methods Team membership rosters (n = 249) included 2,852 IDT members. The final analytical data set included 1,403 surveys (49%) from 221 (89%) teams. A generalized estimating equation model with logit link function, weighted by survey response rates, was used to examine factors associated with perceived team effectiveness. Results Respondents who served as primary care providers (PCPs) were 8% more likely (p = .0044) to view team’s performance as highly effective compared to other team members. Teams with nurse practitioners serving as team leader reported 6% higher likelihood of high-perceived team effectiveness (p = .0234). High team effectiveness was 13% more likely in sites where the predominant culture was characterized as group/developmental, and 7%–8% more likely in sites with lower environmental stress and better resources and staffing, respectively. Conclusions and Implications Team effectiveness is an important indirect measure of HBPC teams’ function. HBPC teams should examine their predominant culture, workplace stress, resources and staffing, and PCP leadership model as part of their quality improvement efforts. Interdisciplinary teamwork, Primary care model, Performance Background and Objectives The Veterans Health Administration’s (VHA) Home-Based Primary Care (HBPC) program was designed to provide comprehensive care to community-based frail Veterans with complex, chronic care needs. The program provides services ranging from primary to rehabilitation, palliative care, long-term home care, disease management, and coordination. In existence for more than 40 years, HBPC is currently available at almost 400 sites across the United States serving more than 50,000 Veterans annually. The program was designed to improve the coordination of care, help Veterans remain independent at home, and reduce preventable emergency room visits, hospitalization, and long-term care institutional placement (Beales & Edes, 2009). At the core of the HBPC model is an interdisciplinary care team, which is expected to assess the enrollees, develop a plan of care, provide all necessary primary care services, and refer enrollees to other services as needed. In addition to physicians, the HBPC interdisciplinary teams on average comprise nine disciplines including nurse practitioners (NPs) or physician assistants (PAs), nurses, social workers, rehabilitation therapists, psychologists, dieticians, pharmacists, and administrative personnel (Karuza et al., 2017). Interdisciplinary teams have been implemented in many health care settings and have been shown to be particularly critical in caring for frail older patients with complex care needs. Studies of teamwork in nursing homes have demonstrated that facilities with better team processes (e.g., communication, conflict resolution, cohesion) receive fewer quality of care deficiency citations, especially those putting residents in immediate jeopardy (Temkin-Greener, Zheng, Cai, Zhao, & Mukamel, 2010). Residents in facilities with better team processes also have been shown to experience better health outcomes, such as fewer fractures, lower risk of pressure ulcers, reduced risk of urinary incontinence, and higher quality of life (Anderson, Issel, & McDaniel, 2003; Gittell, Weinberg, Pfefferle, & Bishop, 2008; Temkin-Greener et al., 2010; Temkin-Greener, Cai, Zheng, Zhao, & Mukamel, 2012). In community-based long-term care, when interdisciplinary teams are used, for example in the Program of All-Inclusive Care for the Elderly (PACE), team functioning has also been shown to be associated with health outcomes. PACE enrollees in programs where staff members reported higher perceived team effectiveness had better functional outcomes and improved urinary incontinence (Mukamel et al., 2006). To date, HBPC research within the VHA has focused largely on the program’s impact on the cost of care and on clinical quality (Edes et al., 2014; Edwards, Prentice, Simon, & Pizer, 2014; Hughes et al., 2000; Kinosian, Edes, Becker, & Hossain, 2014; Percy, Gilmore, & Goldberg, 2007). The results have suggested high variability in these outcomes across the HBPC programs perhaps reflecting differences in the interdisciplinary teams’ functioning. Although teamwork is at the core of the HBPC model, there has been very little research into how these teams are structured and organized, and the extent to which HBPC team and program characteristics may influence staff perceptions of team performance. The objective of this study is to contribute to the understanding of factors that are potentially modifiable and may be important in influencing perceived team effectiveness within the HBPC. Study Overview This study is part of a larger project, titled “Factors Associated with Institutional Use by Veterans in Home Based Primary Care.” Funded by the VHA’s Health Services Research and Development, the project was designed to investigate the relationships between HBPC organizational/structural characteristics, interdisciplinary team practice, and hospital and nursing home utilization. The 3-year study began in 2015 with Phase 1 deploying a Background/Structure Portfolio (BSP) survey addressed to program directors of all 394 HBPC sites, located at Veterans Affairs (VA) medical center facilities and at community-based outpatient clinics. Findings from Phase 1 focusing on HBPC program characteristics have been reported elsewhere (Karuza et al., 2017). This study is part of Phase 2, designed to examine the interdisciplinary team care processes. In Phase 3, the relationship between program structure, team performance, and Veterans’ health care utilization will be examined. The study was reviewed and approved by the Central VA Institutional Review Board (IRB). Research Design: Conceptual Model The model presented in Figure 1 depicts the relationship between HBPC team structure, organization, management or group culture, and team performance measured by perceived team effectiveness. In prior studies of interdisciplinary teams, perceived effectiveness has been shown to influence patients’ health outcomes in a variety of care settings (Flood, 1994; Gittell et al., 2008; Mukamel et al., 2006; Shortell, Rousseau, Gillies, Devers, & Simons, 1991; Temkin-Greener et al., 2012). Studies have also documented the relationship between perceived team effectiveness and other domains of teamwork, including leadership, communication, conflict resolution, and team cohesion (Shortell et al., 1991; Temkin-Greener, Gross, Kunitz, & Mukamel, 2004). These team processes are, in turn, expected to influence perceived team performance, measured with regard to technical quality of care provided and the ability to meet the needs of both patients and their family members (Temkin-Greener, Zheng, Katz, Zhao, & Mukamel, 2009). Figure 1. View largeDownload slide Home-Based Primary Care (HBPC) team structure, organization and management style, and team performance: a conceptual model. Figure 1. View largeDownload slide Home-Based Primary Care (HBPC) team structure, organization and management style, and team performance: a conceptual model. Our study objective, to identify factors that influence perceived work effectiveness, is mostly exploratory as studies of HBPC teams in the VHA are scant and have not focused on the relationship between team characteristics, the environment within which they operate and performance. On the basis of other studies of teamwork, largely in long-term care, we propose to test three hypotheses (Table 1) focusing on factors that may be modifiable and that may affect team performance, such as workplace stress, resources/staffing availability (H1), group culture (H2), and HBPC primary care provider (PCP) model type (H3). Hypothesis 1 is supported by literature showing that workplace conditions are associated with perceived team effectiveness in both community and in institutional long-term care teams (Temkin-Greener et al., 2004; Temkin-Greener et al., 2009). For hypothesis 2, prior studies have shown that teams with more collegial and dynamic organizational cultures perform better than teams with more formal and bureaucratic cultures (Brazil et al., 2010; Körner et al., 2015; Strasser et al., 2002; Temkin-Greener et al., 2009). Research on team performance stresses the importance of team members sharing a common mental model. A more homogeneous perception of collaboration within a team has been shown to be associated with higher perceived team effectiveness (Burtscher et al., 2011; Lemieux-Charles & McGuire, 2006; Lim & Klein, 2006). Therefore, in hypothesis 3, we posit that because nurses account for the largest proportion of HBPC team members (Karuza et al., 2017) NP-led teams will report higher perceived team effectiveness, compared to teams with leaders from other disciplines. Table 1. Home-Based Primary Care (HBPC) Team Structure and Organization as Predictors of Perceived Work Effectiveness Hypothesis Literature-based rationale Key independent variables H1: HBPC teams with higher availability of staff and resources and lower reported workplace stress report higher perceived team effectiveness Workplace conditions have been shown to impact perceived team effectiveness in both community and institutional long-term care teams (Temkin- Greener et al., 2004; Temkin-Greener et al., 2009) Workplace stress Availability of resources and staffing H2: HBPC teams characterized by management culture that is more participatory, collegial (group focused) and innovative (developmental) report higher perceived team effectiveness. Teams with more collegial and dynamic organizational cultures have been shown to perform better than teams with more formal and bureaucratic cultures (Brazil, Wakefield, Cloutier, Tennen, & Hall 2010; Körner, Wirtz, Bengel, & Göritz, 2015; Strasser, Smits, Falconer, Herrin, & Bowen 2002; Helena Temkin-Greener et al., 2009) Culture type: group/developmental versus hierarchical/rational H3: HBPC teams with a nurse practitioner primary care provider model report higher perceived team effectiveness Team performance models stress the importance of team members sharing a common mental model. A more homogeneous perception of collaboration within a team will result in higher perceived team effectiveness (Burtscher, Kolbe, Wacker, & Manser 2011; Lemieux-Charles & McGuire, 2006; Lim & Klein, 2006) Primary care HBPC provider model led by nurse practitioner versus other Hypothesis Literature-based rationale Key independent variables H1: HBPC teams with higher availability of staff and resources and lower reported workplace stress report higher perceived team effectiveness Workplace conditions have been shown to impact perceived team effectiveness in both community and institutional long-term care teams (Temkin- Greener et al., 2004; Temkin-Greener et al., 2009) Workplace stress Availability of resources and staffing H2: HBPC teams characterized by management culture that is more participatory, collegial (group focused) and innovative (developmental) report higher perceived team effectiveness. Teams with more collegial and dynamic organizational cultures have been shown to perform better than teams with more formal and bureaucratic cultures (Brazil, Wakefield, Cloutier, Tennen, & Hall 2010; Körner, Wirtz, Bengel, & Göritz, 2015; Strasser, Smits, Falconer, Herrin, & Bowen 2002; Helena Temkin-Greener et al., 2009) Culture type: group/developmental versus hierarchical/rational H3: HBPC teams with a nurse practitioner primary care provider model report higher perceived team effectiveness Team performance models stress the importance of team members sharing a common mental model. A more homogeneous perception of collaboration within a team will result in higher perceived team effectiveness (Burtscher, Kolbe, Wacker, & Manser 2011; Lemieux-Charles & McGuire, 2006; Lim & Klein, 2006) Primary care HBPC provider model led by nurse practitioner versus other View Large Table 1. Home-Based Primary Care (HBPC) Team Structure and Organization as Predictors of Perceived Work Effectiveness Hypothesis Literature-based rationale Key independent variables H1: HBPC teams with higher availability of staff and resources and lower reported workplace stress report higher perceived team effectiveness Workplace conditions have been shown to impact perceived team effectiveness in both community and institutional long-term care teams (Temkin- Greener et al., 2004; Temkin-Greener et al., 2009) Workplace stress Availability of resources and staffing H2: HBPC teams characterized by management culture that is more participatory, collegial (group focused) and innovative (developmental) report higher perceived team effectiveness. Teams with more collegial and dynamic organizational cultures have been shown to perform better than teams with more formal and bureaucratic cultures (Brazil, Wakefield, Cloutier, Tennen, & Hall 2010; Körner, Wirtz, Bengel, & Göritz, 2015; Strasser, Smits, Falconer, Herrin, & Bowen 2002; Helena Temkin-Greener et al., 2009) Culture type: group/developmental versus hierarchical/rational H3: HBPC teams with a nurse practitioner primary care provider model report higher perceived team effectiveness Team performance models stress the importance of team members sharing a common mental model. A more homogeneous perception of collaboration within a team will result in higher perceived team effectiveness (Burtscher, Kolbe, Wacker, & Manser 2011; Lemieux-Charles & McGuire, 2006; Lim & Klein, 2006) Primary care HBPC provider model led by nurse practitioner versus other Hypothesis Literature-based rationale Key independent variables H1: HBPC teams with higher availability of staff and resources and lower reported workplace stress report higher perceived team effectiveness Workplace conditions have been shown to impact perceived team effectiveness in both community and institutional long-term care teams (Temkin- Greener et al., 2004; Temkin-Greener et al., 2009) Workplace stress Availability of resources and staffing H2: HBPC teams characterized by management culture that is more participatory, collegial (group focused) and innovative (developmental) report higher perceived team effectiveness. Teams with more collegial and dynamic organizational cultures have been shown to perform better than teams with more formal and bureaucratic cultures (Brazil, Wakefield, Cloutier, Tennen, & Hall 2010; Körner, Wirtz, Bengel, & Göritz, 2015; Strasser, Smits, Falconer, Herrin, & Bowen 2002; Helena Temkin-Greener et al., 2009) Culture type: group/developmental versus hierarchical/rational H3: HBPC teams with a nurse practitioner primary care provider model report higher perceived team effectiveness Team performance models stress the importance of team members sharing a common mental model. A more homogeneous perception of collaboration within a team will result in higher perceived team effectiveness (Burtscher, Kolbe, Wacker, & Manser 2011; Lemieux-Charles & McGuire, 2006; Lim & Klein, 2006) Primary care HBPC provider model led by nurse practitioner versus other View Large Other variables that may affect perceived team effectiveness are also considered in the analyses and they are age of the survey respondent, gender, race, HBPC team tenure, and primary profession. Methods Survey Development The survey we used to assess interdisciplinary team performance was based on the original work of the Shortell group first tested in intensive care units (Shortell et al, 1991). Subsequently, this survey tool was modified and psychometrically validated for use in community and in institutional long-term care settings (Temkin-Greener, 2009; 2004). We further pilot-tested the tool’s face and content validity in three HBPC sites with 23 team members. The final survey instrument consisted of items assessing team care domains (leadership, communication/coordination, conflict management, cohesion, and perceived team effectiveness), workplace stress and resources, management/group culture, and respondents’ background characteristics. Questions related to team care domains and workplace stress and resources were based on items using a 5-point Likert scale. The key team domain of interest in this study was perceived team effectiveness, which consists of seven items. Other survey domains of interest were workplace stress (four items) and resources and staffing (five items; Supplementary Table 1). Also included in the survey was a 20-item scale used to measure competing values to determine the predominant culture type of the HBPC program. This measure, developed and validated by Zammuto and Krakower (1991), has been used in health care and in other settings (Marchand, Haines, & Dextras-Gauthier, 2013; Zazzali, Alexander, Shortell, & Burns, 2007). The measure assesses the degree to which team members perceive their programs to value innovation and risk taking versus tradition and established hierarchy. Questions about the respondents’ demographics and professional and work experience were also included in the survey instrument. The full survey instrument is available on request. Survey Procedures The VA Central Office identified 394 HBPC teams. The original procedure was to survey all program directors to obtain detailed program information and rosters for all HBPC team members (see flow diagram in Supplementary Figure 1). However, some program directors were found to manage up to 10 HBPC teams and asking them to provide separate information on all their team members would have constituted a very high response burden. Therefore, for program directors who managed more than two teams, we randomly selected two of their teams and asked the directors to provide us with a list of members for these team. We received team rosters for 249 teams (63%) (Karuza et al., 2017). The combined rosters identified a total of 2,852 team members. These HBPC team members received E-mails explaining the study and requesting their participation. A unique URL, providing a link to the electronic version of the survey, was embedded in each E-mail. The survey was deployed using REDCap (Research Electronic Data Capture) hosted by the Veteran’s Information Resource Center. REDCap is a secure, web-based application designed to support data capture for research studies. The team surveys were initiated in January and completed in July of 2016. Nonrespondents received up to three follow-up reminders. Variable Construction Perceived team effectiveness This domain was constructed from the HBPC survey responses. For each item across this domain, a numerical score ranging from 1 (strongly disagree) to 5 (strongly agree) was assigned. Following previously established and validated methods, if two thirds or more of items in a domain were not completed, the response to the domain was considered missing (Shortell et al., 1991; Strasser et al., 2002; Temkin-Greener et al., 2004). We computed an average score for each domain by adding up the nonmissing values and dividing the sum by the number of nonmissing items. Only four respondents had any missing values on this domain. A domain score of 5 represents the most positive assessment of a domain and a score of 1 the most negative. In examining factors associated with the perceived team effectiveness, we dichotomized the dependent variable (team effectiveness) as high (score ≥ 4) versus low (score < 4). We did this because assessments of team effectiveness were skewed with a mean of 4.44 (SD = 0.33), median of 4.50, and mode of 4.71 (Figure 2). Sensitivity analyses were also performed using the domain score as a continuous variable, but as they did not produce measurably different estimates, for ease of interpretation, we present only results with the dichotomized team effectiveness. Figure 2. View largeDownload slide Distribution of perceived team effectiveness score in Home-Based Primary Care (HBPC) teams. Figure 2. View largeDownload slide Distribution of perceived team effectiveness score in Home-Based Primary Care (HBPC) teams. Personal characteristics Several individual-level factors were included in evaluating the variation in perceived team effectiveness: age (categorical), gender (bivariate), race (bivariate), respondents’ profession (categorical), and tenure on the HBPC team (continuous). Team characteristics Using the previously collected BSP survey data (Karuza et al., 2017), we used three team-level factors that may influence staff assessment of team effectiveness. Because team size may affect how team members interact with each other and with their patients, we included it as a covariate. In the multivariate analysis, each site’s team size was standardized based on the average and standard deviation for all teams combined. HBPC teams are organized around the PCP who is identified as an overall team leader (VHA, 2018). Of the five PCP models present in HBPC, the NP-led model was identified by program directors as the most common. We constructed a variable reflecting the HBPC primary care model led by NPs versus all other. Program characteristics Several work environment or program characteristics were also included as covariates. An average score ranging from 1 (low) to 5 (high) was assigned to indicate the level of workplace stress respondents reported for their team. Another average score reflected the adequacy of team-based resources/staffing. Both of these were included as continuous variables. A variable defining the respondents’ perceptions of culture type predominant at their site was created using a culture scale (Zammuto & Krakower, 1991). Four sets of statements reflecting different culture types were presented to survey respondents. Culture types were characterized as group (based on teamwork and participation), developmental (based on high risk-taking innovation and change), hierarchical (reflecting values and norms associated with bureaucracy), and rational (emphasizing efficiency and achievement). Respondents were asked to assign points (that summed to 100) across the four sets of organizational statements that best described their programs. Culture type score was first calculated for each individual respondent. At the team level, scores were computed using a mean value of the individual scores for a specific culture type. Teams were characterized by mixed culture type. For each team culture type was defined as primarily reflecting group/developmental or hierarchical/rational type of culture. Statistical Analysis We obtained standardized Cronbach’s alpha to assess the internal consistency of the responses within the key domain of interest—perceived team effectiveness—as well as for the domains of workplace stress and resource and staffing. This metric takes on a value between 0 and 1, with values higher than 0.7 indicating high reliability of items in a domain. We estimated a generalized estimating equation model with logit link function, weighted by survey response rates to identify individual, team, and program-level factors that influence the probability of high-perceived team effectiveness. The model controlled for the Veteran Integrated Service Network (VISN) area where the HBPC team was located. Sensitivity analysis using the perceived team effectiveness as a continuous variable was also performed. To assess the possible response bias, we examined how HBPC teams that responded to the survey compared to all HBPC teams on the characteristics of the Veterans they served. We used the fiscal year 2016 VHA administrative and Medicare claims data approved under the study’s IRB. Results We received surveys from 1,606 respondents representing 245 teams. Teams were eliminated from the analytical data set if they met any of the following conditions: (a) only one team member responded (n = 17), (b) team response rate was less than 20% (n = 1), or (c) only one discipline was represented by survey respondents (n = 6). Surveys were assessed for item response completeness and 203 surveys (12.6%) were discarded. Discarded surveys were well distributed across the responding teams. The final analytical data set included 1,403 surveys (49%) from 221 (89%) HBPC sites. On the basis of the 2016 VHA administrative data, we compared responding HBPC teams to all teams with regard to the characteristics of Veterans they served (Supplementary Table 2). HBPC teams responding to the survey were not statistically significantly different from all HBPC teams, except for the proportion of Veterans who lived alone (29.0% vs 28.1%, p = .008, respectively). Characteristics of Survey Respondents Characteristics of the survey respondents and of the HBPC teams are presented in Table 2. The majority of the respondents (57.3%) were aged 40–59 years, female (77.7%), and non-Hispanic white (73.1%). The teams were also fairly homogenous in terms of education with 57.2% having completed an advanced degree (postbaccalaureate), and fewer than 5% not having completed college. The distribution of professional disciplines among team members was consistent with the high level of education. Nurses accounted for 33.5% of respondents and PCPs—including physicians, NPs, and PAs—comprised 15.3%. Other professions, such as social workers therapists, pharmacists, psychologists, and others accounted for 51.2% of the respondents. On average, the respondents tended to have significant tenure in the VA (7.6 years with SD of 6.8) and on the HBPC team (4.8 years; SD = 4.4). Table 2. Characteristics of Survey Respondents and Home-Based Primary Care (HBPC) Teams (N = 221). Level/category % or Mean (SD) Respondent characteristics Age 18–39 years 25.1% 40–59 years 57.3% 60+ years 17.6% Gender Female 77.7% Education level Less than college 4.9% College graduate 37.9% Postgraduate 57.2% Employment status Full time 92.7% Race, ethnicity White, non-Hispanic 73.1% Profession Primary care provider (physician, NP, physician assistant) 15.3% Registered nurse (RN) 33.5% Other professions 51.2% Professional experience Years in current profession 13.9 (10.7) VA experience Years at VA 7.6 (6.8) HBPC position experience Years in current position 4.8 (4.4) Team and program characteristics Team size 14.9 (7.12)  Teams with ≤9 members 23.1%  Teams with 10–15 members 37.5%  Teams with 16+ members 39.4% Provider model  NP is PCP 64.6%  Physician is PCP 4.3%  Physician/NP team as PCP 3.9%  Other model 6.9% Resources and staffing (score 1–5) 3.7 (0.9) Workplace stress (score 1–5) 3.3 (1.05) Program culture type  Group/developmental 48.9%  Rational/hierarchical 51.1% Level/category % or Mean (SD) Respondent characteristics Age 18–39 years 25.1% 40–59 years 57.3% 60+ years 17.6% Gender Female 77.7% Education level Less than college 4.9% College graduate 37.9% Postgraduate 57.2% Employment status Full time 92.7% Race, ethnicity White, non-Hispanic 73.1% Profession Primary care provider (physician, NP, physician assistant) 15.3% Registered nurse (RN) 33.5% Other professions 51.2% Professional experience Years in current profession 13.9 (10.7) VA experience Years at VA 7.6 (6.8) HBPC position experience Years in current position 4.8 (4.4) Team and program characteristics Team size 14.9 (7.12)  Teams with ≤9 members 23.1%  Teams with 10–15 members 37.5%  Teams with 16+ members 39.4% Provider model  NP is PCP 64.6%  Physician is PCP 4.3%  Physician/NP team as PCP 3.9%  Other model 6.9% Resources and staffing (score 1–5) 3.7 (0.9) Workplace stress (score 1–5) 3.3 (1.05) Program culture type  Group/developmental 48.9%  Rational/hierarchical 51.1% PCP = primary care provider; NP = nurse practitioner Note: Standard deviations are only shown for continuous variables. Percentages may not add up to 100% due to missing data View Large Table 2. Characteristics of Survey Respondents and Home-Based Primary Care (HBPC) Teams (N = 221). Level/category % or Mean (SD) Respondent characteristics Age 18–39 years 25.1% 40–59 years 57.3% 60+ years 17.6% Gender Female 77.7% Education level Less than college 4.9% College graduate 37.9% Postgraduate 57.2% Employment status Full time 92.7% Race, ethnicity White, non-Hispanic 73.1% Profession Primary care provider (physician, NP, physician assistant) 15.3% Registered nurse (RN) 33.5% Other professions 51.2% Professional experience Years in current profession 13.9 (10.7) VA experience Years at VA 7.6 (6.8) HBPC position experience Years in current position 4.8 (4.4) Team and program characteristics Team size 14.9 (7.12)  Teams with ≤9 members 23.1%  Teams with 10–15 members 37.5%  Teams with 16+ members 39.4% Provider model  NP is PCP 64.6%  Physician is PCP 4.3%  Physician/NP team as PCP 3.9%  Other model 6.9% Resources and staffing (score 1–5) 3.7 (0.9) Workplace stress (score 1–5) 3.3 (1.05) Program culture type  Group/developmental 48.9%  Rational/hierarchical 51.1% Level/category % or Mean (SD) Respondent characteristics Age 18–39 years 25.1% 40–59 years 57.3% 60+ years 17.6% Gender Female 77.7% Education level Less than college 4.9% College graduate 37.9% Postgraduate 57.2% Employment status Full time 92.7% Race, ethnicity White, non-Hispanic 73.1% Profession Primary care provider (physician, NP, physician assistant) 15.3% Registered nurse (RN) 33.5% Other professions 51.2% Professional experience Years in current profession 13.9 (10.7) VA experience Years at VA 7.6 (6.8) HBPC position experience Years in current position 4.8 (4.4) Team and program characteristics Team size 14.9 (7.12)  Teams with ≤9 members 23.1%  Teams with 10–15 members 37.5%  Teams with 16+ members 39.4% Provider model  NP is PCP 64.6%  Physician is PCP 4.3%  Physician/NP team as PCP 3.9%  Other model 6.9% Resources and staffing (score 1–5) 3.7 (0.9) Workplace stress (score 1–5) 3.3 (1.05) Program culture type  Group/developmental 48.9%  Rational/hierarchical 51.1% PCP = primary care provider; NP = nurse practitioner Note: Standard deviations are only shown for continuous variables. Percentages may not add up to 100% due to missing data View Large Characteristics of HBPC Team There was considerable variability in team size (average = 14.9, SD = 7.1) with 23.1% of teams reporting 9 or fewer members, 37.5% reporting 10–15 members, and 39.4% reporting16 or more members (Table 3). We found no statistically significant (p = .458) correlation between team size and the reported daily census of Veterans, suggesting that larger teams may have had more part-time staff members. Table 3. Probability of High Team Effectiveness: Generalized Estimating Equation (GEE) With Logit Link Function Weighted by Response Rates. High perceived team effectiveness (score≥4) Variables Estimate Relative risk ratio p value Individual characteristics Age; ref = 40–59 years  18–39 0.0229 1.02 .4666  60+ −0.0103 0.99 .7698 Gender; ref = female  Male −0.0175 0.98 .5240 Race; ref = white  Minority −0.0300 0.97 .3156 Years on HBPC team 0.0013 1.00 .5777 Profession of respondent; ref: all other professions  Primary care provider 0.0796 1.08 .0044  Nurse 0.0214 1.02 .4195 Team characteristics Provider model; ref = all other model types  Nurse practitioner as PCP 0.0620 1.06 .0234  Team size 0.0065 1.01 .5699 Program characteristics* Workplace stress 0.0722 1.07 <.0001 Resources and staffing 0.0728 1.08 .0001 Predominant culture; ref = hierarchical/rational culture  Group/developmental 0.1238 1.13 <.0001 High perceived team effectiveness (score≥4) Variables Estimate Relative risk ratio p value Individual characteristics Age; ref = 40–59 years  18–39 0.0229 1.02 .4666  60+ −0.0103 0.99 .7698 Gender; ref = female  Male −0.0175 0.98 .5240 Race; ref = white  Minority −0.0300 0.97 .3156 Years on HBPC team 0.0013 1.00 .5777 Profession of respondent; ref: all other professions  Primary care provider 0.0796 1.08 .0044  Nurse 0.0214 1.02 .4195 Team characteristics Provider model; ref = all other model types  Nurse practitioner as PCP 0.0620 1.06 .0234  Team size 0.0065 1.01 .5699 Program characteristics* Workplace stress 0.0722 1.07 <.0001 Resources and staffing 0.0728 1.08 .0001 Predominant culture; ref = hierarchical/rational culture  Group/developmental 0.1238 1.13 <.0001 HBPC = Home-Based Primary Care; PCP = primary care provider Note: Estimates for VISNs are not shown. View Large Table 3. Probability of High Team Effectiveness: Generalized Estimating Equation (GEE) With Logit Link Function Weighted by Response Rates. High perceived team effectiveness (score≥4) Variables Estimate Relative risk ratio p value Individual characteristics Age; ref = 40–59 years  18–39 0.0229 1.02 .4666  60+ −0.0103 0.99 .7698 Gender; ref = female  Male −0.0175 0.98 .5240 Race; ref = white  Minority −0.0300 0.97 .3156 Years on HBPC team 0.0013 1.00 .5777 Profession of respondent; ref: all other professions  Primary care provider 0.0796 1.08 .0044  Nurse 0.0214 1.02 .4195 Team characteristics Provider model; ref = all other model types  Nurse practitioner as PCP 0.0620 1.06 .0234  Team size 0.0065 1.01 .5699 Program characteristics* Workplace stress 0.0722 1.07 <.0001 Resources and staffing 0.0728 1.08 .0001 Predominant culture; ref = hierarchical/rational culture  Group/developmental 0.1238 1.13 <.0001 High perceived team effectiveness (score≥4) Variables Estimate Relative risk ratio p value Individual characteristics Age; ref = 40–59 years  18–39 0.0229 1.02 .4666  60+ −0.0103 0.99 .7698 Gender; ref = female  Male −0.0175 0.98 .5240 Race; ref = white  Minority −0.0300 0.97 .3156 Years on HBPC team 0.0013 1.00 .5777 Profession of respondent; ref: all other professions  Primary care provider 0.0796 1.08 .0044  Nurse 0.0214 1.02 .4195 Team characteristics Provider model; ref = all other model types  Nurse practitioner as PCP 0.0620 1.06 .0234  Team size 0.0065 1.01 .5699 Program characteristics* Workplace stress 0.0722 1.07 <.0001 Resources and staffing 0.0728 1.08 .0001 Predominant culture; ref = hierarchical/rational culture  Group/developmental 0.1238 1.13 <.0001 HBPC = Home-Based Primary Care; PCP = primary care provider Note: Estimates for VISNs are not shown. View Large The most common HBPC primary care model was with NP serving as PCP (64.4%), followed by physicians serving as PCP (24.3%), and a collaboration of physician/NP (3.9%). Slightly fewer than half (48.9%) of all teams were characterized, by the HBPC survey respondents, as having group/developmental work culture with 51.1% as rational/hierarchical. Assessments of resource/staffing availability and of the work environment stress were given average scores of 3.7 and 3.5 (respectively), based on a scale of 1–5 with higher scores indicating a more favorable assessment. Perceived Team Effectiveness The domain of perceived team effectiveness demonstrated high reliability with Cronbach’s alpha of 0.90, and the domains of workplace stress and resources and staffing had Cronbach’s alphas of 0.84 and 0.83, respectively (Supplementary Table 1). Findings focusing on the relationship between individual-, team-, and program-level characteristics and perceived team effectiveness are presented in Table 3. Only respondent’s profession was a statistically significant predictor of high team effectiveness. Respondents who served as PCPs were 8% more likely (relative risk ratio [RRR] = 1.08; p = .0044) to view their team’s performance as highly effective than all other professionals. Respondents who served as nurses did not rank their teams’ performance differently than all other professionals. In support of hypothesis 3, we found that teams in which NPs served as PCPs reported 6% (RRR = 1.06; p = .0234) higher likelihood of high-perceived team effectiveness, compared to models with other disciplines serving as PCP. In teams with predominant culture characterized as group/developmental, respondents were 13% (RRR = 1.13) more likely to report high team effectiveness than in teams with hierarchical/rational culture, thus supporting hypothesis 2. Furthermore, in teams with lower environmental stress and better resources/staffing, respondents were also more likely to report higher team effectiveness (RRR = 1.07 and 1.08, respectively), providing support for hypothesis 1. Discussion and Implications We used a previously developed and psychometrically validated tool for assessing interdisciplinary team performance to identify team characteristics associated with perceived effectiveness in the VA’s HBPC teams. HBPC teams were fairly homogeneous vis-à-vis educational attainment of their members (virtually all college or postgraduate), race/ethnicity, and full-time employment at the VA. They also appeared to be quite stable based on the length of their team members’ tenure. All team members were professionals, unlike teams in other long-term care settings such as PACE or nursing homes where many members are paraprofessionals such as certified nurse assistants (CNAs). In serving the HBPC patients, CNAs are hired, as needed, by home health agencies on contract to VA, but they are not members of HBPC teams and are not directly managed or supervised by the program. This team homogeneity may, in part, explain why we observed relatively little variation in the overall team performance across different HBPC programs, and why individual-level characteristics, except for the respondents’ profession, were not statistically significant predictors of team effectiveness. Our analysis to identify predictors of overall team effectiveness highlights several potentially modifiable team and program level factors. Teams with a primary care model in which NPs served as PCPs were perceived to be more effective, compared to other models (hypothesis 3). To some extent, this may be because nurses, who are the single largest group in HBPC teams, may be expressing higher confidence in their profession as team leaders. Nurses may find NPs more accessible as leaders compared, for example, to physicians, and may feel more comfortable interacting with them. This may also be a reflection on the preparation and team-specific competencies that are part of the advance practice nurses’ educational model, which may uniquely prepare them to be effective in such team leadership positions (American Association of Colleges of Nursing [AACN], 2011). Furthermore, finding that teams led by NPs were perceived to be more effective is intriguing because, if effective from quality and safety perspectives, this may prove also important for improving access to high-value primary care in other care models in populations with complex medical conditions. However, many HBPC teams delivered care using more than one primary care model. It is not clear what specific roles NPs and physicians played in caring for an HBPC team’s panel of patients. These roles, and how they are allocated, may be important considerations in overall team performance. Further research is needed to better understand the relationship of NP versus MD-led models on assuring high-quality and cost-effective care for Veterans. Our findings also demonstrated that when teams were assessed as having group/developmental versus rational/hierarchical culture, respondents were more likely to assess their teams as highly effective (hypothesis 2). Changing organizational culture is a daunting task (Scott, Mannion, Davies, & Marshall, 2003). However, as each program is a blend of culture types, perhaps making small but committed and persistent transformations may be sufficient to move the culture type needle in the desired direction. Prior literature has suggested a number of factors that may encourage a group/developmental culture. For example, leaders who inspire cognitive change are known to regularly restate values they wish to inculcate in team members, praise specific actions to reinforce specific behaviors, and own their mistakes and create remedies to prevent future ones, thus keeping the values and norms of the organizational culture visible to team members on a daily basis (Parmelli et al., 2011). Although the regional location of HBPC programs was not a predictor of team effectiveness, access to adequate resources and staffing, as well as workplace stress, were found to be associated with team effectiveness (hypothesis 1). The findings emphasize that as measures of teamwork are multidimensional, the efforts to improve teamwork also need to be multifaceted. Study Limitations Several caveats should be noted. First, the way in which our survey captured the PCP model was constrained by the choices offered to the respondents, possibly obscuring important information about how these models operate. For example, identifying the NP as the main model for primary care did not detail the role that physicians may have had on that HBPC team. This is particularly true for teams where physicians might have played a wide range of roles ranging from case preceptor at interdisciplinary case conferences to a home-visiting PCP with a smaller, but high complexity panel of patients. Thus, additional investigation may be warranted to understand the team roles of NPs and physicians in HBPC teams. Second, although we captured team size, we were unable to assess the proportion of time providers spent working in their HBPC teams. It may be important to understand if presence of part-timers influences team members’ perceived team effectiveness. Third, we did not collect information on turnover, which may also be an important contributor to team effectiveness. But the information on staff team tenure, which we did include, allowed us to use it as a proxy variable suggesting than turnover may not be a critical issue on HBPC teams. As is true in all surveys, it may be possible that the HBPC program managers who recognized their teams as being low functioning did not provide us with team rosters and that members of teams that were experiencing difficulties were more likely not to respond. This concern is somewhat attenuated based on our comparison of responding teams to all HBPC teams showing they were largely no different in terms of the characteristics of Veterans they served. Teams in all VISNs participated in the survey. Finally, this study is based on a cross-sectional survey that limits causal interpretations. Conclusion In conclusion, perceived team effectiveness is an important indirect measure of how HBPC teams function and this study has identified some areas that may be modifiable and could be used in quality improvement efforts. Ultimately, however, what counts is the teams’ ability to bring about good patient care. The next step, therefore, needs to focus on identifying those team characteristics that are associated with improved quality and efficiency of care for Veterans. Funding This work was supported by a Veterans Affairs Health Science Research and Development grant (HSR5-011-13S), Project ID 1193882, VACO No. IIR12-152–3. This work was supported by a Merit Award (IIR 12–152) from the U.S. Department of Veterans Affairs Health Services Research and Development Program. The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the US government. Conflict of Interest None reported. References American Association of Colleges of Nursing (AACN) . ( 2011 ). The Essentials of Master’s Education in Nursing . Retrieved from http://www.aacnnursing.org/Portals/42/Publications/MastersEssentials11.pdf (accessed September 2, 2018). Anderson , R. A. , Issel , L. M. , & McDaniel , R. R. Jr . ( 2003 ). Nursing homes as complex adaptive systems: Relationship between management practice and resident outcomes . Nursing Research , 52 , 12 – 21 . doi: https://doi.org/10.1097/00006199-200301000-00003 Google Scholar Crossref Search ADS PubMed Beales , J. L. , & Edes , T . ( 2009 ). Veteran’s Affairs Home Based Primary Care . Clinics in Geriatric Medicine , 25 , 149 – 54, viii . doi: https://doi.org/10.1016/j.cger.2008.11.002 Google Scholar Crossref Search ADS PubMed Brazil , K. , Wakefield , D. B. , Cloutier , M. M. , Tennen , H. , & Hall , C. B . ( 2010 ). Organizational culture predicts job satisfaction and perceived clinical effectiveness in pediatric primary care practices . Health Care Management Review , 35 , 365 – 371 . doi: https://doi.org/10.1097/HMR.0b013e3181edd957 Google Scholar Crossref Search ADS PubMed Burtscher , M. J. , Kolbe , M. , Wacker , J. , & Manser , T . ( 2011 ). Interactions of team mental models and monitoring behaviors predict team performance in simulated anesthesia inductions . Journal of Experimental Psychology. Applied , 17 , 257 – 269 . doi: https://doi.org/10.1037/a0025148 Google Scholar Crossref Search ADS PubMed Edes , T. , Kinosian , B. , Vuckovic , N. H. , Nichols , L. O. , Becker , M. M. , & Hossain , M . ( 2014 ). Better access, quality, and cost for clinically complex veterans with home-based primary care . Journal of the American Geriatrics Society , 62 , 1954 – 1961 . doi: https://doi.org/10.1111/jgs.13030 Google Scholar Crossref Search ADS PubMed Edwards , S. T. , Prentice , J. C. , Simon , S. R. , & Pizer , S. D . ( 2014 ). Home-based primary care and the risk of ambulatory care-sensitive condition hospitalization among older veterans with diabetes mellitus . Journal of American Medical Association Internal Medicine , 74 , 1796 – 1803 . doi: https://doi.org/10.1001/jamainternmed.2014.4327 Flood , A. B . ( 1994 ). The impact of organizational and managerial factors on the quality of care in health care organizations . Medical Care Review , 51 , 381 – 428 . doi: https://doi.org/10.1177/107755879405100402 Google Scholar Crossref Search ADS PubMed Gittell , J. , Weinberg , D. , Pfefferle S. , & Bishop C . ( 2008 ). Impact of relational coordination on job satisfaction and quality outcomes: A study of nursing homes . Human Resources , 18 , 154 – 170 . doi: https://doi.org/10.1111/j.1748-8583.2007.00063.x Hughes , S. L. , Weaver , F. M. , Giobbie-Hurder , A. , Manheim , L. , Henderson , W. , Kubal , J. D. , … Cummings , J .; Department of Veterans Affairs Cooperative Study Group on Home-Based Primary Care . ( 2000 ). Effectiveness of team-managed home-based primary care: A randomized multicenter trial . JAMA , 284 , 2877 – 2885 . doi:10.1001/jama.284.22.2877 Google Scholar Crossref Search ADS PubMed Karuza , J. , Gillespie , S. M. , Olsan , T. , Cai , X. , Dang , S. , Intrator , O., . . . Edes , T . ( 2017 ). National Structural Survey of Veterans Affairs Home-Based Primary Care Programs . Journal of the American Geriatrics Society , 65 , 2697 – 2701 . doi: https://doi.org/10.1111/jgs.15126 Google Scholar Crossref Search ADS PubMed Kinosian , B. , Edes , T. , Becker , P. , & Hossain , M . ( 2014 ). Expanding VA’s home based primary care (HBPC) while improving performance: 2006–2012 . Journal of the American Geriatrics Society , 62 , S136 . doi:ezpminer.urmc.rochester.edu/10.1111/jgs.12870 Google Scholar Crossref Search ADS Körner , M. , Wirtz , M. A. , Bengel , J. , & Göritz , A. S . ( 2015 ). Relationship of organizational culture, teamwork and job satisfaction in interprofessional teams . BMC Health Services Research , 15 , 243 . doi: https://doi.org/10.1186/s12913-015-0888-y Google Scholar Crossref Search ADS PubMed Lemieux-Charles , L. , & McGuire , W. L . ( 2006 ). What do we know about health care team effectiveness? A review of the literature . Medical Care Research and Review: MCRR , 63 , 263 – 300 . doi: https://doi.org/10.1177/1077558706287003 Google Scholar Crossref Search ADS PubMed Lim , B. C. , & Klein , K. J . ( 2006 ). Team mental models and team performance: A field study of the effects of team mental model similarity and accuracy . Journal of Organizational Behaviour , 27 , 403 – 418 . doi:10.1002/job.387 Google Scholar Crossref Search ADS Marchand , A. , Haines , V. Y. III , & Dextras-Gauthier , J . ( 2013 ). Quantitative analysis of organizational culture in occupational health research: A theory-based validation in 30 workplaces of the organizational culture profile instrument . BMC Public Health , 13 , 443 . doi: https://doi.org/10.1186/1471-2458-13-443 Google Scholar Crossref Search ADS PubMed Mukamel , D. B. , Temkin-Greener , H. , Delavan , R. , Peterson , D. R. , Gross , D. , Kunitz , S. , & Williams , T. F . ( 2006 ). Team performance and risk-adjusted health outcomes in the Program of All-Inclusive Care for the Elderly (PACE) . The Gerontologist , 46 , 227 – 237 . doi:10.1093/geront/46.2.227 Google Scholar Crossref Search ADS PubMed Parmelli , E. , Flodgren , G. , Beyer , F. , Baillie , N. , Schaafsma , M. E. , & Eccles , M. P . ( 2011 ). The effectiveness of strategies to change organisational culture to improve healthcare performance: A systematic review . Implementation Science: IS , 6 , 33 . doi: https://doi.org/10.1186/1748-5908-6-33 Google Scholar Crossref Search ADS PubMed Percy , A. , Gilmore , J. , & Goldberg , M . ( 2007 ). Health System for Veterans: An Interim Report . Washington, DC . https://www.cbo.gov/sites/default/files/cbofiles/ftpdocs/88xx/doc8892/12-21-va_healthcare.pdf. Accessed September 25, 2018 . Scott , T. , Mannion , R. , Davies , H. T. O. , & Marshall , M. N . ( 2003 ). Policy roundtable implementing culture change in health care: Theory and practice . International Journal for Quality in Health Care , 15 , 111 – 118 . doi:10.1093/intqhc/mzg021 Google Scholar Crossref Search ADS PubMed Shortell , S. M. , Rousseau , D. M. , Gillies , R. R. , Devers , K. J. , & Simons , T. L . ( 1991 ). Organizational assessment in intensive care units (ICUs): Construct development, reliability, and validity of the ICU nurse-physician questionnaire . Medical Care , 29 , 709 – 726 . doi:10.1097/00005650-199108000-00004 Google Scholar Crossref Search ADS PubMed Strasser , D. C. , Smits , S. J. , Falconer , J. A. , Herrin , J. S. , & Bowen , S. E . ( 2002 ). The influence of hospital culture on rehabilitation team functioning in VA hospitals . Journal of Rehabilitation Research and Development , 39 , 115 – 125 . Google Scholar PubMed Temkin-Greener , H. , Cai , S. , Zheng , N. T. , Zhao , H. , & Mukamel , D. B . ( 2012 ). Nursing home work environment and the risk of pressure ulcers and incontinence . Health Services Research , 47 , 1179 – 1200 . doi: https://doi.org/10.1111/j.1475-6773.2011.01353.x Google Scholar Crossref Search ADS PubMed Temkin-Greener , H. P. , Gross , D. P. , Kunitz , S. J. , & Mukamel , D. B . ( 2004 ). Measuring interdisciplinary team performance in a long-term care setting . Medical Care , 42 , 472 – 481 . doi: https://doi.org/10.1097/01.mlr.0000124306.28397.e2 Google Scholar Crossref Search ADS PubMed Temkin-Greener , H. , Zheng , N. T. , Cai , S. , Zhao , H. , & Mukamel , D. B . ( 2010 ). Nursing home environment and organizational performance: Association with deficiency citations . Medical Care , 48 , 357 – 364 . doi: https://doi.org/10.1097/MLR.0b013e3181ca3d70 Google Scholar Crossref Search ADS PubMed Temkin-Greener , H. , Zheng , N. , Katz , P. , Zhao , H. , & Mukamel , D. B . ( 2009 ). Measuring work environment and performance in nursing homes . Medical Care , 47 , 482 – 491 . doi:10.1097/MLR.0b013e318190cfd3 Google Scholar Crossref Search ADS PubMed VHA . ( 2018 ). Patient Aligned Care Team (PACT) Handbook . Washington, DC . https://www.google.com/search?q=Patient+Aligned+Care+Team+(PACT)+Handbook.&oq=Patient+Aligned+Care+Team+(PACT)+Handbook.&aqs=chrome..69i57j0.1355j0j7&sourceid=chrome&ie=UTF-8. Accessed September 25, 2018 . Zammuto , R. , & Krakower , J . ( 1991 ). Quantitative and qualitative studies of organizational culture . In R. Woodman & W. Pasmore (Eds.), Research in Organizational Change and Development . Greenwich, CT : JAI Press . Zazzali , J. L. , Alexander , J. A. , Shortell , S. M. , & Burns , L. R . ( 2007 ). Organizational culture and physician satisfaction with dimensions of group practice . Health Services Research , 42 , 1150 – 1176 . doi: https://doi.org/10.1111/j.1475-6773.2006.00648.x Google Scholar Crossref Search ADS PubMed © The Author(s) 2019. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: [email protected]. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)