Month & 2016
1
The experience of Patient Aligned Care Team (PACT) members Amy C. Ladebue Christian D. Helfrich Zachary T. Gerdes Stephan D. Fihn Karin M. Nelson George G. Sayre Background: In April 2010, the Veterans Health Administration (VHA) launched the Patient Aligned Care Team (PACT) initiative to implement a patient-centered medical home (PCMH) model. Few evaluations have addressed the effects of PCMH on health care professionals’ experiences. Purposes: The aim of this study was to contribute to evaluation of the PACT initiative and the broader literature on PCMH by assessing respondents’ experiences of implementing a PCMH model and becoming a teamlet. Methodology/Approach: A retrospective qualitative analysis of open-text responses in a survey fielded to all VHA Primary Care personnel (VHA Primary Care physicians, nurse practitioners, physician assistants, nurse care managers, clinical associates, and administrative clerks) in May and June 2012 (approximately 2 years into the 5-year planned implementation of PACT) using deductive and inductive content analysis. The main measures were two open-response fields: ‘‘Is there anything else you would like us to relay to the VA leadership in Central Office?’’ and ‘‘Do you have any other comments or feedback on PACT?’’ The data consisted of free text responses of 3,868 survey participants who provided text for one or both of the open-response fields.
Key words: evaluation, patient-centered care, patient-centered medical home (PCMH), team-based care, teamlet Amy C. Ladebue, BA, is Research Assistant, Health Services Research and Development Center for Innovation for Veteran-Centered and Value-Driven Care, College of Arts and Sciences, Seattle University, Washington. Christian D. Helfrich, MPH, PhD, is Research Investigator, Health Services Research and Development Center for Innovation for Veteran-Centered and Value-Driven Care, and Research Assistant Professor, Department of Health Services, University of Washington, Seattle. Zachary T. Gerdes, BA, is Research Assistant, Health Services Research and Development Center for Innovation for Veteran-Centered and Value-Driven Care, and Doctoral Student, Department of Psychology, University of Akron, Ohio. Stephan D. Fihn, MD, MPH, is Director, Office of Analytics and Business Intelligence, U.S. Department of Veterans Affairs, Seattle, Washington, Professor and Head, Division of General Internal Medicine, Department of Medicine, University of Washington, Seattle, and Professor, Department of Health Services, University of Washington, Seattle. Karin M. Nelson, MD, MSHS, is Staff Physician, Health Services Research and Development Center for Innovation for Veteran-Centered and Value-Driven Care, and Associate Professor, Department of Medicine, University of Washington, Seattle. George G. Sayre, PsyD, is Health Science Researcher and Qualitative Resources Coordinator, Health Services Research and Development Center for Innovation for Veteran-Centered and Value-Driven Care, and Clinical Assistant Professor, Department of Health Services, University of Washington, Seattle. Authors are affiliated with the VA and this work was funded by the VA Office of Patient Care Services. The views expressed here do not necessarily reflect the position or policy of the Department of Veterans Affairs. The authors are employees or volunteers with the Department of Veterans Affairs, and this work was undertaken through the support of the VA’s Patient Aligned Care Team Demonstration Laboratory initiative. The authors have disclosed that they have no significant relationship with, or financial interest in, any commercial companies pertaining to this article. DOI: 10.1097/HMR.0000000000000048 Health Care Manage Rev, 2016, 00(0), 00Y00 Copyright B 2016 Wolters Kluwer Health | Lippincott Williams & Wilkins
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
2
Health Care Management Review
Month & 2016
Findings: Although respondents viewed PACT positively as a model and reported it improved relationships with patients and increased patient satisfaction, they described multiple barriers to achieving functioning teamlets and unintended consequences, including reduced time with patients, increased participant burnout, and decreased team efficacy because of low-performing team members. A central theme related to staffing being insufficient for the new model. Practice Implications: Insufficient staffing of PCMH teams is a critical barrier to realizing the benefits of the new model. Frontline staff have concrete recommendations for other problems, such as using back-up teams to cover during absences, but that will require providing more opportunities for feedback from staff to be heard.
Background The patient-centered medical home (PCMH) is a primary care model characterized by several features, including (a) team-based care, (b) enhanced access to care, (c) coordinated care across inpatient and outpatient settings or across primary and specialty care, and (d) comprehensive care, including preventive, acute, and chronic care (Jackson et al., 2013; Peikes et al., 2012). The PCMH model has been endorsed by four of the major professional societies representing primary care professionals (Patient-Centered Primary Care Collaborative, 2007). There have been dozens of demonstration projects, pilot studies, and quality improvement initiatives evaluating and testing various aspects of the PCMH model and its outcomes (Alexander & Bae, 2012; Jackson et al., 2013; Peikes et al., 2012) However, the current literature on PCMH implementation provides limited information on the experiences of the health care professionals implementing the model (Jackson et al., 2013; Peikes et al., 2012). Only three published studies have assessed staff experiences, and these were conducted with relatively few practitioners and clinicians (Jackson et al., 2013). An earlier Agency for Healthcare Research and Quality systematic review concluded that there was only one rigorous evaluation of the effects of PCMH on health care professional experiences, and it was inconclusive (Peikes et al., 2012). The transition to team-based care represents a significant transition for health care professionals. It may increase burnout and decrease job satisfaction. Increased burnout and decreased job satisfaction have previously been found to be associated with worse patient safety (Cimiotti, Aiken, Sloane, & Wu, 2012; Halbesleben, Wakefield, Wakefield, & Cooper, 2008; Shanafelt et al., 2010; Spence Laschinger & Leiter, 2006; West, Tan, Habermann, Sloan, & Shanafelt, 2009). Increased rates of burnout in clinicians also increase the likelihood of leaving one’s job (Hinami, Whelan, Miller, Wolosin, & Wetterneck, 2012; Landon, Reschovsky, Pham, & Blumenthal, 2006; Linzer et al., 2009; Shanafelt et al., 2012; Shanafelt, Sloan, Satele, & Balch, 2011; Van Bogaert, Clarke, Roelant, Meulemans, & Van de Heyning, 2010), taking sick leave, and experiencing relationship problems and depression (Parker & Kulik, 1995; Sargent, Sotile, Sotile, Rubash, & Barrack, 2009; Shanafelt, Bradley, Wipf, & Back, 2002). In April 2010, the Veterans Health Administration (VHA)
launched the Patient Aligned Care Team (PACT), a national initiative to implement a PCMH in primary care (Klein, 2011; Rosland et al., 2013). The goal was to improve clinical continuity, coordination, and patient-centeredness (Rosland et al., 2013). The central change has been team-based care. PACT ‘‘teamlets,’’ comprising a primary care provider (PCP), nurse care manager or registered nurse (RN), clinical associate (e.g., a licensed practical nurse [LPN] or medical assistant), and an administrative clerk who work with the PCP to provide more comprehensive and proactive care. For example, they might extend care beyond the clinic visit by contacting the patient before and after the visit, which may help address routine issues, such as preventive services that might otherwise take time away from addressing more critical issues (Bodenheimer & Laing, 2007; Nutting et al., 2010). PACT teamlets are expected to improve workflow by helping team members function at the top of their competencies (i.e., training and abilities; Rosland et al., 2013). In the present study, we sought to contribute to evaluation of the PACT initiative and to the broader findings of PCMH literature by assessing respondents’ experiences of implementing a PCMH model and becoming a teamlet. Particularly, we examined how the experiences of team members affected work satisfaction and PACT teamlets’ abilities to implement a model of patient-centered care.
Design We conducted a retrospective content analysis of open-text ‘‘general comment’’ responses to a survey fielded to VHA Primary Care personnel in May and June 2012, approximately 2 years into the 5-year planned implementation of PACT. Surveys were completed online and distributed by national leadership (the Office of the VHA Deputy UnderSecretary for Health for Operations and Management) via e-mailed links to leadership of the 21 regional VHA networks. Network leadership distributed the links via medical center directors and chiefs of staff to local primary care leadership and to frontline employees. Surveys were anonymous. In 19 of the 21 regional networks, all VA PCPs (physicians, nurse practitioners, and physician assistants), nurse care managers, clinical associates, administrative clerks, as well as primary care-based affiliated providers (e.g., social workers, clinical pharmacists, nutritionists) were eligible to participate.
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
3
Implications of Transitions to Teamlets
The two networks that did not participate were demonstration sites that fielded similar but more in-depth surveys not reported here. The two open-response fields were preceded by the prompts ‘‘Is there anything else you would like us to relay to the VA leadership in Central Office?’’ and ‘‘Do you have any other comments or feedback on PACT?’’ The survey sample and methodology have been previously described (Helfrich et al., 2014). Qualitative analyses were conducted on a de-identified dataset with respondent characteristics and other survey responses removed. Responses were analyzed using simultaneous deductive and inductive content analysis (Elo & Kynga¨s, 2008). Inductive content analysis consists of open/unstructured coding and allows for the identification of emergent previously unidentified or unexpected themes. Deductive content analysis is more structured and consists of identifying ‘‘meaning units’’ (discrete phrases, sentences, or series of sentences that convey one idea or one related set of perceptions) that fit within preidentified a priori categories (Elo & Kynga¨s, 2008). A priori codes included barriers and facilitators to PACT implementation and burnout. Coding was conducted by the first author (AL) utilizing AtlasTi qualitative data analysis software and reviewed throughout the analysis by the senior author (GS) for confirmability and groundedness (Sandelowski, 1986) including searching for negative instances (Morse, Barrett, Mayan, Olson, & Spiers, 2008). Subcodes were developed by identifying broad themes followed by subcoding schemes based on representative survey responses. Quotations that did not accurately fit existing subcodes were used to develop new subcodes iteratively. The results of analysis were reviewed by members (GS, CH, & KN) of the research team to assess thoroughness and comprehensiveness for analytical rigor. Clinical members of the research team (KN & SF) reviewed the findings to assure relevance and transferability.
Findings A total of 6,467 surveys were returned (a 25% estimated response rate), of which 3,868 (60%) included responses to one or both open-text questions. The responses varied across teamlet roles, showing that the responses were not limited to administrators or nurses but were from PCPs, nurses, administrators, and other roles within the teamlet. We began data analysis by line-by-line open coding using both a priori codes and simultaneously developing emergent codes as needed to describe salient aspects of the content. Initial categories were developed based on similar content characteristics and then grouped into higher-order categories containing related but meaningfully distinct subcategories. Open coding continued until saturation (the point at which subsequent data failed to produce new findings; Sandelowski, 1986); this occurred after analysis of 1,548 survey responses. After saturation was reached, 157 additional responses were coded in order to assure accuracy (1,705 observations coded). We identified seven higher order categories: holistic teamlet
experience, burnout, perceived effects on patients, level of competency within teams, the unheard voices of team members, unintended consequences of PACT, and respondents’ suggested improvements for the PACT model.
Holistic Teamlet Experience PACT was implemented to provide employees with a teambased approach to care, coordinate their work, and define specific duties for each member. Several survey respondents expressed overall satisfaction with teamlets in terms of the impacts they have had for their teams and patients. One respondent stated: ‘‘Because of our [team] huddles and constant communication with our PACT team, we have taken initiatives to improve patient care.’’ Another identified the importance of the roles within team-based care: ‘‘Having an assigned RN, LPN, and clerk has improved medical delivery considerably.’’ Despite beliefs that PACT has benefits for patients and may be effective as a model for patient care, the majority of respondents articulated dissatisfaction with the implementation of PACT teamlets because of barriers such as understaffing and limited resources: ‘‘PACT model is awesome, but in order to make it work, each teamlet needs to be fully staffed. We are lucky that our team is staffed, but it’s not the same in every team and it really affects their measures and productivity.’’ Respondents described inconsistent implementation across PACT teamlets within the same clinic. One respondent who was on two different PACT teams described how the teams, ‘‘Ifunctioned significantly differently. One is very efficient and organized and the other, not so much. We all had the same initial education. It is interesting to see the difference.’’ Respondents identified a number of factors that limited PACT: (a) lack of appropriate staffing ratios and resources; (b) lack of sufficient training; (c) scheduling issues for teamlets and patients; (d) the distractions of telephones and computers; and (e) lack of satisfaction, clarity, and expectations in roles and responsibilities of all team members. Appropriate staffing ratios and resources. Respondents described the model as an ‘‘excellent philosophy’’ that could be ‘‘a great thing,’’ but the model cannot work without necessary resources. One participant wrote: ‘‘Ideally, PACT would be a good system; however, not realistic due to lack of staff and not enough hours in the day to complete the work that we are supposed to do.’’ If one staff member was absent, the rest of the team suffered because there were no back-up team members. Several respondents suggested that other roles be added to the teamlet, such as social workers, psychologists, and pharmacists. They described a need for increased staffing for these roles, particularly pharmacists: ‘‘[The] current ratio of CPS [clinical pharmacy specialists] to Teamlets [at seven or eight to one] is not sufficient for optimal pharmacotherapy management for our veterans, especially when a good number of our veterans are on 10-plus medications.’’
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
4
Health Care Management Review
Many respondents reported a lack of other resources, including space, time, and equipment in their clinics. As one respondent wrote: ‘‘Space has been the most negative factor in implementing the PACT modelIthere is not enough equipment, such as faxes [and] printers, available to send and get patient informationI.’’ Without these supports, some respondents felt unable to properly complete their duties and provide for their patients: ‘‘Currently we are limping along putting out fires and becoming burnt out in the process. We were never given the tools to succeed.’’ Lack of sufficient training. The majority of respondents reported a lack of thorough and consistent training for all team members prior to implementation in order to prepare them for the PACT model: ‘‘It would be helpful if the training could take place before changes are mandated. It’s a little like being dropped in the middle of a lake and told to swim when you’ve never done it.’’ Some expressed a need for better quality of training, describing the training to be a ‘‘waste of time and money.’’ Scheduling issues for teamlets and patients. Several respondents reported that appointment slots were overbooked and that administrative clerks felt overwhelmed with the number of calls, appointments, and walk-ins: ‘‘We have to become serious about not seeing walk-ins or being a ‘by appointment only’ clinic. The walk-ins are overwhelming and they have not been reduced.’’ There were also issues with lack of flexibility with scheduling visits of different durations: ‘‘No ability for the staff to adjust appointment times/slots to accommodate patients that need more/less time with the PCP.’’ Other concerns with shorter appointments included inadequate time for patients who require extensive care because of chronic pain and narcotic and other medication management. Several felt they lacked sufficient control over the clinic, including walk-ins, appointments, and their own schedules: ‘‘The clinic is chaos. Running at crisis level virtually all day every day.’’ Distractions of telephones and computers. Many participants reported unintended consequences of using a call center and computer-based reminders to improve patient care. Some felt that these constant reminders were a ‘‘waste of time’’ and ‘‘useless distractions’’ to their routine or to more important priorities. One reported that the ‘‘volume of clinical reminders and the volume of telephone care calls is a major burden.’’ This appeared to be a particular barrier for nurse care managers: ‘‘Currently phone calls are going into nurses’ officeVall phone calls should be diverted to central line so that the phone is not ringing all day in the nurses’ officeVclerical staff should be answering the phone calls.’’ Responding to telephone calls was not perceived as patient care: ‘‘Care managers should not be spending so much time creating reports for facility leadership and manning a nurse call line but instead, providing care management to patients.’’ Others supported the notion that more duties mean less patient care: ‘‘Since [the] PACT change, I have
Month & 2016
less interaction with the patients and feel ‘pushed’ all the time. I miss being with the patient.’’ Lack of satisfaction, clarity, and expectations in roles and responsibilities of all team members. Most respondents expressed a need for clear expectations for all team members, including leadership as well as accountability for leadership to do its part: ‘‘I think there should be accountability of management for the fact that our staffing is such that PACT cannot be implemented. Our veterans are hurt by their lack of attention and leadership.’’
Burnout The majority of respondents linked the barriers previously described to team member burnout. One respondent reported that ‘‘PACT has added many new duties without promised resources and has offered no new pay increases or incentives. It has brought on many inconsistencies from team to team and decreased job satisfaction.’’ Another respondent expressed worry over the burnout rates: ‘‘Burnout of my colleagues is a huge problem; I am worried we will lose some of our best providers.’’ RNs, LPNs, and administrative clerks in particular were reported to be ‘‘burning out rapidly.’’ Interestingly, this observation was made by respondents who identified as roles other than RNs, LPNs, and administrative clerks: ‘‘The biggest issue is if a Team Member is gone continuity of care becomes compromised and the workload may double or triple for at least one member of the teamVusually the RN or LPN.’’
Perceived Effect on Patients Respondents’ views on PACT’s impact on patient care were divided. Many reported that PACT resulted in positive improvements to patient care because patients are actively engaged in their care and feel heard and valued by their providers. Other respondents reported that patients may lose responsibility in their own care or may not fully understand or benefit from the model. Positive effects on patients. Many respondents expressed that patients ‘‘love’’ PACT and typically feel ‘‘more involved in their own care.’’ Some reported that PACT allowed for patients to be seen more quickly and that the teamlet was more accessible to the patient than with previous care models: ‘‘From the beginning of this PACT implementation, many veterans have really benefitted from it and they love the program [I] they can reach their teamlet, either through phone or secure messaging.’’ Those involved with PACT feel that the model allows for a better relationship with their patients: ‘‘It really builds relationships with Veterans. And you get to really know them as you case manage them, and help them get their needs met sooner rather than later.’’ The positive effects on patient care and negative effects on team members were not mutually exclusive: ‘‘PACT has
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
5
Implications of Transitions to Teamlets
been very helpful toward patient care; however, it has overburdened the staff.’’ Even some who expressed struggles with PACT and teamlets as being a ‘‘work in progress’’ acknowledged that, ‘‘Most importantly, the veterans are very satisfied with the change.’’ Negative effects on patients. Many respondents also expressed concern for the quality of patient care within the PACT model. They reported that, with the team-care model, patients lost responsibility for their own care because they either (a) did not fully understand the model or (b) believed that providers and nurses were supposed to take full responsibility for their health. One respondent particularly emphasized issue with medications: ‘‘Patients don’t understand PACT and how to be more responsible for something as simple as refilling medication. Many of the older vets don’t like using the telephone for reorders and prefer to have the PCP or Pharmacist reorder for them.’’ These respondents also expressed concern that the efforts to improve access had not necessarily succeeded. Patients still had to wait long periods of time or travel very far for care. Respondents seemed to feel that if a patient ‘‘has to wait 2 months to see [the doctor],’’ the model is not working as intended.
Level of Competency Within Teams Contrary to working to the top of their competency within the PACT, some respondents reported a growing burden of clerical work, particularly nurses: ‘‘I am an LPN but I feel like a glorified clerk most of the time. I do more clerical work than nursing.’’ Some employees felt they were required to work below their licensure, being caught under a large amount of paperwork that distracted from their training as a care provider. This appeared to lower job satisfaction and may cause tension within the clinic and between team members. In contrast, other respondents reported that the model allowed them to work to their full potential despite feeling overwhelmed and undertrained for the PACT model: ‘‘[I] will feel better once I have attended the PACT training. Teamwork has significantly improved with the PACT model and everyone is working to their highest training level.’’
The Unheard Voices of Team Members Quite a few employees felt undervalued for their work and felt caught in a system without being heard. One respondent wrote: ‘‘Listen to the concerns of the people in the trenches.’’ Others described feeling silenced when they offered suggestions to their leaders/managers in order to improve their situations: ‘‘When we have ideas to improve, leadership does not listen or work with us’’ and ‘‘[I] do not feel that upper management really understands what is going on in the clinics.’’ One respondent stated that ‘‘directives are always being changed and revised to the point that following them is cumbersome to providing actual care.’’ This appears to contribute to a sense of powerlessness:
There is a commanding and condescending tone when our leaders communicate with us. I feel as though we are not considered part of the process. One day we are simply told this is how we are doing such and such. We want more autonomy within our clinics.
Unintended Consequences of PACT Implementation Within team conflict and the effect of ‘‘weakest link.’’ Several respondents described problems within team functioning: ‘‘This [teamlet model] will not work for a slacker.’’ This suggests that if one team member did not fully participate, the entire team may suffer and/or lose efficacy. The team-based model is contingent upon all members working to their highest potential and fulfilling their responsibilities for their team and patients. A respondent wrote about the need for the teams to work together: ‘‘If one person isn’t doing what needs to be done it fails our whole team, and we need to be able to rely on each other!’’ Another stated, ‘‘The success of the teamlet is limited by the weakest link.’’ Some participants reported that the PACT model has encountered resistance from some employees who felt that the model of care they had before PACT worked more effectively. These employees may have been resistant to the changes of PACT, which may have resulted in some of the difficulties of implementing the model. One respondent noted some of their colleagues as ‘‘overworked, low-functioning nursing staff’’ in the clinic who do not ‘‘buy into PACT’’ and have therefore made implementing PACT difficult. Not only does the model require individuals to buy into the team and do their best, but each member should be held consistently to appropriate expectations. Another respondent expressed: ‘‘We need a full team, full time. We need enforcement and consistency of each staff level expectations. Clear expectations of all members across the board.’’ ‘‘Atmosphere of anxiety.’’ Numerous respondents described a perceived lack of consideration to ‘‘day-to-day variability’’ in the clinic and found it difficult to compensate for variability throughout the day. There was a concern that leadership seemed to care more about the number of patients seen rather than the quality of care patients receive: ‘‘All [leadership] cares about is the number of patients to be seen in the clinic, not the quality of care.’’ Similarly, some felt that PACT placed too much emphasis on performance metrics that did not reflect true quality of care: ‘‘PACT seems to be too number-driven. [It] creates an atmosphere of anxiety. Teams [are] trying to improve numbers at the risk of good patient care and individual patient evaluation.’’ Others reported fears of losing their job if they did not fully adhere to the PACT program, even if parts of the program did not work for their specific duties or the clinic: ‘‘We have been told that if we think we are working too hard that there are people waiting in line for our jobs.’’
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
6
Health Care Management Review
Respondents’ Suggested Improvements for PACT Model Increased training. A vast majority of respondents called for improved training including training for managing the PACT model and the respective roles within it. ‘‘The whole program should have been worked out entirely instead of a learn-as-you-go thing.’’ Many noted that training should provide all employees with the same expectations and instruction, which could reduce burnout, stress, and dissatisfaction. Care plans and specific PACTs for chronic pain patients. A few respondents suggested specific improvements to PACT related to chronic pain management. Particularly, they asked for standardized care plans for patients with chronic pain, an increased need for narcotic awareness, and generally more care for patients than a teamlet can offer. Some suggested a ‘‘comprehensive pain program and living with chronic pain with all the specialists available and other resources not to be managed by primary care’’ and ‘‘a PACT team just to manage patients with chronic pain issues.’’ Back-up teams. Back-up teams could provide teams with an ‘‘extra hand’’ or a back-up in cases of call-ins, vacations, and sick days. Respondents felt that, by creating back-up teams, the clinic could continue to run more smoothly, even when some employees are absent. This could avoid circumstances where teams might need help from other teams (because of missing members), especially if the second team is also overwhelmed, understaffed, and therefore unable to fully support the other team: ‘‘There needs to be a sub system for call offs/vacations so that teams are not pulled apart to other staff areas’’ and ‘‘at least two float nurses to help all our teamlets out in times of vacations, sickness, and just for an extra hand so we can meet the needs of our patients.’’
Discussion These findings suggest that, in the first 2 years of implementation, PACT has created burdens for PCPs and staff. At the same time, many participants believe PACT provides positive benefits for patients. We find remarkably similar findings to recent analyses from five VA Demonstration Laboratories, which have conducted in-depth, prospective evaluations of the PACT model finding that primary care clinicians and staff were enthusiastic about the concept of PACT but felt necessary resources were not available (Tuepker et al., 2014); that inadequate staffing resources and leadership engagement were frequently barriers to implementation and created major burdens on primary care personnel (True et al., 2012); and that training was not sufficient, particularly regarding team roles. However, in spite of all of these challenges, VA clinicians and staff felt that PACT had improved the care experience for the veteran (Rodriguez et al., 2014). Our findings offer corroboration from a national sample of clinics and argue that the experience of demonstration laboratory sites is generalizable even though the
Month & 2016
laboratories were selected competitively and received funding for testing and evaluating PACT. Similar findings have also been reported outside the VA. A study of facilitators and barriers to implementing a primary care team redesign to improve patient-centered care found that physician engagement and local leadership support were key factors that distinguished clinics that succeeded and those that struggled (Grace, Rich, Chin, & Rodriguez, 2014). Early on, the CMS Medical Home Demonstration noted that change fatigue was a serious threat from implementing so many transformational changes at the same time and that it affects even highly capable, highly motivated teams (Nutting et al., 2009). In addition, this paper highlights two new findings. First, PCMH team-based care may have unintended consequences of increasing stress and burnout. Notably, team members may feel dependent on the ‘‘weakest link’’ in their team, and efforts to improve access through greater direct telephone access may increase stress. Alternative methods of contact, such as telephone, and enhanced computer support are key components of PCMH models, but these findings suggest that, without the necessary staffing and scheduled time to address them, they can add to distraction and stress. Prior quantitative analyses of the PACT survey data found that specific elements of PACT team-based care, such as having a fully staff teamlet and participatory decision-making, were associated with significantly lower burnout (Helfrich et al., 2014); the present findings provide an important qualifier suggesting that a team-based PCMH model reduces burnout when effectively implemented, but that effective implementation is the exception rather than the rule. Second, frontline employees were eager to offer feedback and recommendations for improving PACT implementation. Even though the open-text survey field is a brief, impersonal mechanism for providing feedback and recommendations, respondents took it with alacrity (60% of respondents answered one or both questions). They made concrete recommendations: creation of back-up teams (instead of cannibalizing other PACT teamlets) to cover for absent teamlet members and development of PACT teamlets specifically for chronic care patients. We know of no other surveys of this scope to collect and analyze open-text data. This may represent an important method for future research. It may also be helpful to view these findings in the context of broader conceptual work on innovation implementation in health care and the factors that drive it. One of the most widely cited, comprehensive implementation models is the Consolidated Framework for Implementation Research (CFIR), a conceptual model, derived from literature synthesis, of the implementation process and the factors that drive it (Damschroder et al., 2009). The CFIR is organized in five broad categories: (a) the innovation, (b) inner organizational setting, (c) outer setting, (d) the individuals involved, and (e) the process by which implementation is accomplished. Most of our findings can be interpreted as falling within the
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
7
Implications of Transitions to Teamlets
inner organizational setting, which is the structural, political, and cultural contexts through which the implementation occurs and includes available resources (e.g., issues with PACT staffing ratios), access to knowledge and information necessary to implement the new practice (e.g., need for further training support for the PACT teamlets), and learning climate (which arose in the theme of unheard voices of the team and the feeling that leadership does not listen). These are important for two reasons. First, the inner context represents factors that are largely within the control of the organization and not inherent issues with the PCMH model. Issues with the PCMH model would be related to the inherent clinical effectiveness or compatibility of the medical home model in primary care (i.e., the CFIR construct of the innovation), the broader primary care environment (i.e., the CFIR construct of the outer setting), or the primary care personnel (i.e., the CFIR construct of the individuals). Second, our findings also suggest a potential modification to CFIR. Among our key themes were unintended consequences, such as the team model resulting in team members being dependent on the ‘‘weakest link.’’ This is not an outcome of poor implementation, such as the increased stress from lack of adequate staffing, rather a consequence of a team-based model. Unintended consequences are a familiar element of systems theory applied to health care (Scott, Mannion, Davies, & Marshall, 2003) but are neither part of CFIR or other widely cited implementation and diffusion models, such as Greenhalgh’s model based on a review of implementation and diffusion in service organizations (Greenhalgh, Robert, Macfarlane, Bate, & Kyriakidou, 2004), nor Roger’s Diffusion of Innovations theory (Rogers, 2003). Although CFIR depicts the individuals involved in implementation as fundamentally unpredictable, unintended consequences from the innovation itself are not expressly modeled in the CFIR. The strength of our study includes the large sample size of over 3,800 respondents representing more than 600 primary care clinics across the United States while retaining the advantage of qualitative research that uses the respondent’s own words describing the relationships among conditions, such as the barriers and facilitators of PACT implementation or how team-based care has contributed to a better patient experience for the veteran. Our sample is exclusively from VA primary care, which differs in many important structural respects from other systems that have implemented PCMH models. However, many of the core changes introduced by PACT, and commented on by respondents, relate to factors such as team-based care, logistical constraints, and resource demands, which should be widely applicable to other primary care settings.
Practice Implications Implementation of team-based care may, by its nature, introduce interdependencies that make team members vulnerable
to underperforming or undertrained teammates and makes the PCMH model particularly dependent on adequate staffing and training. Respondents suggested use of back-up teams to care for patients when PACT teamlet members were absent (rather than cannibalizing other teamlets). They also recommended formation of chronic disease-specific PACT teamlets for specific subpopulations, notably chronic pain patients. Although not feasible in all settings, these and other solutions might improve PACT implementation and be applicable to other PCMH settings.
Limitations There are three potential limitations. First, this study relied on secondary survey data in which the text fields followed a variety of structured questions, such as their level of burnout, which might have influenced the participant’s responses. Open-text fields offered limited space for responses, and the descriptions may be less detailed than they would be if implementation barriers and facilitators and burnout were the primary data collection topic. Another limitation was that the data were de-identified, and consequently subanalyses regarding respondent position were not possible. Finally, the survey was voluntary, and because of the way the survey was distributed, we have no way of knowing for certain which employees had an opportunity to respond or how representative they are. We do have comparative demographic data from a separate, general employee survey fielded the month preceding the PACT survey and have compared respondents in the four main PACT occupations (i.e., providers, nurse care managers, clinical associates, and administrative clerks). This comparison, previously published, found that the PACT survey sample had a slightly higher proportion of supervisors and lower proportion of African American respondents but was virtually identical on other characteristics, including tenure, age, and gender (Helfrich et al., 2014). Despite these limitations, our secondary analysis was appropriate because there was a good ‘‘fit’’ between our research question and the primary data (Hinds, Vogel, & Clarke-Steffen, 1997; open-ended narratives focused on PACT implementation), and the large sample size allowed us to obtain very saturated data. The sample is also exceptionally large in terms of clinics (n = 626). Previous PCMH often comes from small numbers of early adopters or demonstration sites that are likely not representative of the broader population of clinics (Jackson et al., 2013). Our sample represented over two thirds of VA primary care clinics across the United States. For comparison, in the Agency for Healthcare Research and Quality review, among the PCMH interventions that they determined had been rigorously reviewed, the number of clinics studied ranged from 6 to 18 in total (Peikes et al., 2012). Nevertheless, we do not know to what extent the observed teamlet member experiences are generalizable to other
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
8
Health Care Management Review
providers and staff implementing a PCMH model. Moreover, this analysis was not intended to quantify the categories. The large number of respondents and sites gives us added confidence that these categories reflect the breadth of PACT team member experience in the early implementation of this model; however, this analysis was not designed to compare the relative prevalence or importance of one category or subcategories over another. Finally, the data and conclusions only reflect the attitude of participants who responded to the survey questions and cannot be used to make conclusions regarding patient quality of care or outcomes. Future research should seek to explore the experience of PCMH teamlet members in other settings and with differing PCMH models and their effect on patient outcomes.
Conclusion We found significant challenges posed by the implementation of PACT, including unintended consequences that impacted team member job satisfaction. In addition, staffing and training in the model were perceived as inadequate, and nurses, in particular, found themselves spending a greater amount of time on clerical work than before the PACT model. These findings reflect the early experience of an initiative expected to take up to 5 years to implement. At the same time, respondents expressed support for the PACT model, and some felt it was having a positive impact on the quality of patient care. Respondents provided key recommendations for how these challenges can be addressed, notably more systematic training, more training for chronic pain management, and the use of back-up staff to help alleviate interruptions in teams’ functioning.. Acknowledgments
Our thanks to Julie Kurutz and John Witzlib of the VA Healthcare Analysis and Information Group for fielding the survey, to Rachel Orlando for administrative support, and to members of the PACT Demonstration Laboratory Coordinating Center Organizational Function Working Group who contributed invaluable expertise and feedback in survey development. This work was undertaken as part of the VA’s PACT Demonstration Laboratory initiative, supporting and evaluating VA’s transition to a PCMH model. Funding for the PACT Demonstration Laboratory initiative is provided by the VA Office of Patient Care Services. 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.. References Alexander, J. A., & Bae, D. (2012). Does the patient-centred medical home work? A critical synthesis of research on patient-
Month & 2016
centred medical homes and patient-related outcomes. Health Services Management Research, 25(2), 51Y59. doi: 10.1258/hsmr .2012.012001 Bodenheimer, T., & Laing, B. Y. (2007). The teamlet model of primary care. Annals of Family Medicine, 5(5), 457Y461. doi: 10 .1370/afm.731 Cimiotti, J. P., Aiken, L. H., Sloane, D. M., & Wu, E. S. (2012). Nurse staffing, burnout, and health care-associated infection. American Journal of Infection Control, 40(6), 486Y490. doi: 10 .1016/j.ajic.2012.02.029 Damschroder, L. J., Aron, D. C., Keith, R. E., Kirsh, S. R., Alexander, J. A., & Lowery, J. C. (2009). Fostering implementation of health services research findings into practice: A consolidated framework for advancing implementation science. Implementation Science, 4(1), 50. Elo, S., & Kynga¨s, H. (2008). The qualitative content analysis process. Journal of Advanced Nursing, 62(1), 107Y115. doi: 10 .1111/j.1365-2648.2007.04569.x Grace, S. M., Rich, J., Chin, W., & Rodriguez, H. P. (2014). Flexible implementation and integration of new team members to support patient-centered care. Healthcare, 2(2), 145Y151. doi: 10.1016/j.hjdsi.2014.02.003 Greenhalgh, T., Robert, G., Macfarlane, F., Bate, P., & Kyriakidou, O. (2004). Diffusion of innovations in service organizations: Systematic review and recommendations. Milbank Quarterly, 82(4), 581Y629. Halbesleben, J. R., Wakefield, B. J., Wakefield, D. S., & Cooper, L. B. (2008). Nurse burnout and patient safety outcomes: Nurse safety perception versus reporting behavior. Western Journal of Nursing Research, 30(5), 560Y577. doi: 10.1177/0193945907311322 Helfrich, C. D., Dolan, E. D., Simonetti, J., Reid, R. J., Joos, S., Wakefield, B. J., I Nelson, K. (2014). Elements of team-based care in a patient-centered medical home are associated with lower burnout among va primary care employees. Journal of General Internal Medicine, 29(Suppl 2), 659Y666. doi: 10.1007/s11606-013-2702-z Hinami, K., Whelan, C. T., Miller, J. A., Wolosin, R. J., & Wetterneck, T. B. (2012). Job characteristics, satisfaction, and burnout across hospitalist practice models. Journal of Hospital Medicine, 7(5), 402Y410. Hinds, P. S., Vogel, R. J., & Clarke-Steffen, L. (1997). The possibilities and pitfalls of doing a secondary analysis of a qualitative data set. Qualitative Health Research, 7(3), 408Y424. Jackson, G. L., Powers, B. J., Chatterjee, R., Prvu Bettger, J., Kemper, A. R., Hasselblad, V., I Williams, J. J. W. (2013). The patientcentered medical home: A systematic review. Annals of Internal Medicine, 158(3), 169Y178. doi: 10.7326/0003-4819-158-3-20130 2050-00579 Klein, S. (2011). The Veterans Health Administration: Implementing patient-centered medical homes in the nation’s largest integrated delivery system. New York, NY: The Commonwealth Fund. Landon, B. E., Reschovsky, J. D., Pham, H. H., & Blumenthal, D. (2006). Leaving medicine: The consequences of physician dissatisfaction. Medical Care, 44(3), 234Y242. Linzer, M., Manwell, L. B., Williams, E. S., Bobula, J. A., Brown, R. L., Varkey, A. B., I Horner-Ibler, B. (2009). Working conditions in primary care: Physician reactions and care quality. Annals of Internal Medicine, 151(1), 28Y36. Morse, J. M., Barrett, M., Mayan, M., Olson, K., & Spiers, J. (2008). Verification strategies for establishing reliability and validity in qualitative research. International Journal of Qualitative Methods, 1(2), 13Y22. Nutting, P. A., Crabtree, B. F., Miller, W. L., Stewart, E. E., Stange, K. C., & Jaen, C. R. (2010). Journey to the patientcentered medical home: A qualitative analysis of the experiences
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
9
Implications of Transitions to Teamlets
of practices in the National Demonstration Project. Annals of Family Medicine, 8(Suppl 1), S45YS56; S92. doi: 10.1370/ afm.1075 Nutting, P. A., Miller, W. L., Crabtree, B. F., Jaen, C. R., Stewart, E. E., & Stange, K. C. (2009). Initial lessons from the first national demonstration project on practice transformation to a patient-centered medical home. Annals of Family Medicine, 7(3), 254Y260. doi: 10.1370/afm.1002 Parker, P. A., & Kulik, J. A. (1995). Burnout, self- and supervisorrated job performance, and absenteeism among nurses. Journal of Behavioral Medicine, 18(6), 581Y599. Patient-Centered Primary Care Collaborative. (2007). Joint principles of the patient-centered medical home. Washington, DC: Author. Peikes, D., Zutshi, A., Genevro, J., Smith, K., Parchman, M., & Meyers, D. (2012). Early evidence on the patient-centered medical home. Rockville, MD: Agency for Healthcare Research and Quality. Rodriguez, H., Giannitrapani, K., Stockdale, S., Hamilton, A., Yano, E., & Rubenstein, L. (2014). Teamlet structure and early experiences of medical home implementation for veterans. Journal of General Internal Medicine, 29(2), 623Y631. doi: 10 .1007/s11606-013-2680-1 Rogers, E. (2003). Diffusion of innovations (5th ed.). New York, NY: Free Press. Rosland, A. M., Nelson, K., Sun, H., Dolan, E. D., Maynard, C., Bryson, C., I Schectman, G. (2013). The patient-centered medical home in the Veterans Health Administration. The American Journal of Managed Care, 19(7), e263Ye272. Sandelowski, M. (1986). The problem of rigor in qualitative research [comparative study]. ANS. Advances in Nursing Science, 8(3), 27Y37. Sargent, M. C., Sotile, W., Sotile, M. O., Rubash, H., & Barrack, R. L. (2009). Quality of life during orthopaedic training and academic practice. Part 1: Orthopaedic surgery residents and faculty. The Journal of Bone and Joint Surgery. American Volume, 91(10), 2395Y2405. doi: 10.2106/jbjs.h.00665 Scott, T., Mannion, R., Davies, H. T. O., & Marshall, M. N. (2003). Implementing culture change in health care: Theory and practice. International Journal for Quality in Health Care, 15(2), 111Y118. doi: 10.1093/intqhc/mzg021
Shanafelt, T., Sloan, J., Satele, D., & Balch, C. (2011). Why do surgeons consider leaving practice? Journal of the American College of Surgeons, 212(3), 421Y422. doi: 10.1016/j.jamcollsurg .2010.11.006 Shanafelt, T. D., Balch, C. M., Bechamps, G., Russell, T., Dyrbye, L., Satele, D., I Freischlag, J. (2010). Burnout and medical errors among American surgeons. Annals of Surgery, 251(6), 995Y1000. doi: 10.1097/SLA.0b013e3181bfdab3 Shanafelt, T. D., Boone, S., Tan, L., Dyrbye, L. N., Sotile, W., Satele, D., I Oreskovich, M. R. (2012). Burnout and satisfaction with work-life balance among us physicians relative to the general us population. Archives of Internal Medicine, 172(18), 1377Y1385. doi: 10.1001/archinternmed.2012.3199 Shanafelt, T. D., Bradley, K. A., Wipf, J. E., & Back, A. L. (2002). Burnout and self-reported patient care in an internal medicine residency program. Annals of Internal Medicine, 136(5), 358Y367. Spence Laschinger, H. K., & Leiter, M. P. (2006). The impact of nursing work environments on patient safety outcomes: The mediating role of burnout/engagement. The Journal of Nursing Administration, 36(5), 259Y267. True, G., Butler, A. E., Lamparska, B. G., Lempa, M. L., Shea, J. A., Asch, D. A., & Werner, R. M. (2012). Open access in the patient-centered medical home: Lessons from the Veterans Health Administration. In Journal of General Internal Medicine. doi: 10.1007/s11606-012-2279-y Tuepker, A., Kansagara, D., Skaperdas, E., Nicolaidis, C., Joos, S., Alperin, M., & Hickam, D. (2014). ‘‘We’ve not gotten even close to what we want to do’’: A qualitative study of early patientcentered medical home implementation. Journal of General Internal Medicine, 29(2), 614Y622. doi: 10.1007/s11606-013-2690-z Van Bogaert, P., Clarke, S., Roelant, E., Meulemans, H., & Van de Heyning, P. (2010). Impacts of unit
1
The experience of Patient Aligned Care Team (PACT) members Amy C. Ladebue Christian D. Helfrich Zachary T. Gerdes Stephan D. Fihn Karin M. Nelson George G. Sayre Background: In April 2010, the Veterans Health Administration (VHA) launched the Patient Aligned Care Team (PACT) initiative to implement a patient-centered medical home (PCMH) model. Few evaluations have addressed the effects of PCMH on health care professionals’ experiences. Purposes: The aim of this study was to contribute to evaluation of the PACT initiative and the broader literature on PCMH by assessing respondents’ experiences of implementing a PCMH model and becoming a teamlet. Methodology/Approach: A retrospective qualitative analysis of open-text responses in a survey fielded to all VHA Primary Care personnel (VHA Primary Care physicians, nurse practitioners, physician assistants, nurse care managers, clinical associates, and administrative clerks) in May and June 2012 (approximately 2 years into the 5-year planned implementation of PACT) using deductive and inductive content analysis. The main measures were two open-response fields: ‘‘Is there anything else you would like us to relay to the VA leadership in Central Office?’’ and ‘‘Do you have any other comments or feedback on PACT?’’ The data consisted of free text responses of 3,868 survey participants who provided text for one or both of the open-response fields.
Key words: evaluation, patient-centered care, patient-centered medical home (PCMH), team-based care, teamlet Amy C. Ladebue, BA, is Research Assistant, Health Services Research and Development Center for Innovation for Veteran-Centered and Value-Driven Care, College of Arts and Sciences, Seattle University, Washington. Christian D. Helfrich, MPH, PhD, is Research Investigator, Health Services Research and Development Center for Innovation for Veteran-Centered and Value-Driven Care, and Research Assistant Professor, Department of Health Services, University of Washington, Seattle. Zachary T. Gerdes, BA, is Research Assistant, Health Services Research and Development Center for Innovation for Veteran-Centered and Value-Driven Care, and Doctoral Student, Department of Psychology, University of Akron, Ohio. Stephan D. Fihn, MD, MPH, is Director, Office of Analytics and Business Intelligence, U.S. Department of Veterans Affairs, Seattle, Washington, Professor and Head, Division of General Internal Medicine, Department of Medicine, University of Washington, Seattle, and Professor, Department of Health Services, University of Washington, Seattle. Karin M. Nelson, MD, MSHS, is Staff Physician, Health Services Research and Development Center for Innovation for Veteran-Centered and Value-Driven Care, and Associate Professor, Department of Medicine, University of Washington, Seattle. George G. Sayre, PsyD, is Health Science Researcher and Qualitative Resources Coordinator, Health Services Research and Development Center for Innovation for Veteran-Centered and Value-Driven Care, and Clinical Assistant Professor, Department of Health Services, University of Washington, Seattle. Authors are affiliated with the VA and this work was funded by the VA Office of Patient Care Services. The views expressed here do not necessarily reflect the position or policy of the Department of Veterans Affairs. The authors are employees or volunteers with the Department of Veterans Affairs, and this work was undertaken through the support of the VA’s Patient Aligned Care Team Demonstration Laboratory initiative. The authors have disclosed that they have no significant relationship with, or financial interest in, any commercial companies pertaining to this article. DOI: 10.1097/HMR.0000000000000048 Health Care Manage Rev, 2016, 00(0), 00Y00 Copyright B 2016 Wolters Kluwer Health | Lippincott Williams & Wilkins
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
2
Health Care Management Review
Month & 2016
Findings: Although respondents viewed PACT positively as a model and reported it improved relationships with patients and increased patient satisfaction, they described multiple barriers to achieving functioning teamlets and unintended consequences, including reduced time with patients, increased participant burnout, and decreased team efficacy because of low-performing team members. A central theme related to staffing being insufficient for the new model. Practice Implications: Insufficient staffing of PCMH teams is a critical barrier to realizing the benefits of the new model. Frontline staff have concrete recommendations for other problems, such as using back-up teams to cover during absences, but that will require providing more opportunities for feedback from staff to be heard.
Background The patient-centered medical home (PCMH) is a primary care model characterized by several features, including (a) team-based care, (b) enhanced access to care, (c) coordinated care across inpatient and outpatient settings or across primary and specialty care, and (d) comprehensive care, including preventive, acute, and chronic care (Jackson et al., 2013; Peikes et al., 2012). The PCMH model has been endorsed by four of the major professional societies representing primary care professionals (Patient-Centered Primary Care Collaborative, 2007). There have been dozens of demonstration projects, pilot studies, and quality improvement initiatives evaluating and testing various aspects of the PCMH model and its outcomes (Alexander & Bae, 2012; Jackson et al., 2013; Peikes et al., 2012) However, the current literature on PCMH implementation provides limited information on the experiences of the health care professionals implementing the model (Jackson et al., 2013; Peikes et al., 2012). Only three published studies have assessed staff experiences, and these were conducted with relatively few practitioners and clinicians (Jackson et al., 2013). An earlier Agency for Healthcare Research and Quality systematic review concluded that there was only one rigorous evaluation of the effects of PCMH on health care professional experiences, and it was inconclusive (Peikes et al., 2012). The transition to team-based care represents a significant transition for health care professionals. It may increase burnout and decrease job satisfaction. Increased burnout and decreased job satisfaction have previously been found to be associated with worse patient safety (Cimiotti, Aiken, Sloane, & Wu, 2012; Halbesleben, Wakefield, Wakefield, & Cooper, 2008; Shanafelt et al., 2010; Spence Laschinger & Leiter, 2006; West, Tan, Habermann, Sloan, & Shanafelt, 2009). Increased rates of burnout in clinicians also increase the likelihood of leaving one’s job (Hinami, Whelan, Miller, Wolosin, & Wetterneck, 2012; Landon, Reschovsky, Pham, & Blumenthal, 2006; Linzer et al., 2009; Shanafelt et al., 2012; Shanafelt, Sloan, Satele, & Balch, 2011; Van Bogaert, Clarke, Roelant, Meulemans, & Van de Heyning, 2010), taking sick leave, and experiencing relationship problems and depression (Parker & Kulik, 1995; Sargent, Sotile, Sotile, Rubash, & Barrack, 2009; Shanafelt, Bradley, Wipf, & Back, 2002). In April 2010, the Veterans Health Administration (VHA)
launched the Patient Aligned Care Team (PACT), a national initiative to implement a PCMH in primary care (Klein, 2011; Rosland et al., 2013). The goal was to improve clinical continuity, coordination, and patient-centeredness (Rosland et al., 2013). The central change has been team-based care. PACT ‘‘teamlets,’’ comprising a primary care provider (PCP), nurse care manager or registered nurse (RN), clinical associate (e.g., a licensed practical nurse [LPN] or medical assistant), and an administrative clerk who work with the PCP to provide more comprehensive and proactive care. For example, they might extend care beyond the clinic visit by contacting the patient before and after the visit, which may help address routine issues, such as preventive services that might otherwise take time away from addressing more critical issues (Bodenheimer & Laing, 2007; Nutting et al., 2010). PACT teamlets are expected to improve workflow by helping team members function at the top of their competencies (i.e., training and abilities; Rosland et al., 2013). In the present study, we sought to contribute to evaluation of the PACT initiative and to the broader findings of PCMH literature by assessing respondents’ experiences of implementing a PCMH model and becoming a teamlet. Particularly, we examined how the experiences of team members affected work satisfaction and PACT teamlets’ abilities to implement a model of patient-centered care.
Design We conducted a retrospective content analysis of open-text ‘‘general comment’’ responses to a survey fielded to VHA Primary Care personnel in May and June 2012, approximately 2 years into the 5-year planned implementation of PACT. Surveys were completed online and distributed by national leadership (the Office of the VHA Deputy UnderSecretary for Health for Operations and Management) via e-mailed links to leadership of the 21 regional VHA networks. Network leadership distributed the links via medical center directors and chiefs of staff to local primary care leadership and to frontline employees. Surveys were anonymous. In 19 of the 21 regional networks, all VA PCPs (physicians, nurse practitioners, and physician assistants), nurse care managers, clinical associates, administrative clerks, as well as primary care-based affiliated providers (e.g., social workers, clinical pharmacists, nutritionists) were eligible to participate.
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
3
Implications of Transitions to Teamlets
The two networks that did not participate were demonstration sites that fielded similar but more in-depth surveys not reported here. The two open-response fields were preceded by the prompts ‘‘Is there anything else you would like us to relay to the VA leadership in Central Office?’’ and ‘‘Do you have any other comments or feedback on PACT?’’ The survey sample and methodology have been previously described (Helfrich et al., 2014). Qualitative analyses were conducted on a de-identified dataset with respondent characteristics and other survey responses removed. Responses were analyzed using simultaneous deductive and inductive content analysis (Elo & Kynga¨s, 2008). Inductive content analysis consists of open/unstructured coding and allows for the identification of emergent previously unidentified or unexpected themes. Deductive content analysis is more structured and consists of identifying ‘‘meaning units’’ (discrete phrases, sentences, or series of sentences that convey one idea or one related set of perceptions) that fit within preidentified a priori categories (Elo & Kynga¨s, 2008). A priori codes included barriers and facilitators to PACT implementation and burnout. Coding was conducted by the first author (AL) utilizing AtlasTi qualitative data analysis software and reviewed throughout the analysis by the senior author (GS) for confirmability and groundedness (Sandelowski, 1986) including searching for negative instances (Morse, Barrett, Mayan, Olson, & Spiers, 2008). Subcodes were developed by identifying broad themes followed by subcoding schemes based on representative survey responses. Quotations that did not accurately fit existing subcodes were used to develop new subcodes iteratively. The results of analysis were reviewed by members (GS, CH, & KN) of the research team to assess thoroughness and comprehensiveness for analytical rigor. Clinical members of the research team (KN & SF) reviewed the findings to assure relevance and transferability.
Findings A total of 6,467 surveys were returned (a 25% estimated response rate), of which 3,868 (60%) included responses to one or both open-text questions. The responses varied across teamlet roles, showing that the responses were not limited to administrators or nurses but were from PCPs, nurses, administrators, and other roles within the teamlet. We began data analysis by line-by-line open coding using both a priori codes and simultaneously developing emergent codes as needed to describe salient aspects of the content. Initial categories were developed based on similar content characteristics and then grouped into higher-order categories containing related but meaningfully distinct subcategories. Open coding continued until saturation (the point at which subsequent data failed to produce new findings; Sandelowski, 1986); this occurred after analysis of 1,548 survey responses. After saturation was reached, 157 additional responses were coded in order to assure accuracy (1,705 observations coded). We identified seven higher order categories: holistic teamlet
experience, burnout, perceived effects on patients, level of competency within teams, the unheard voices of team members, unintended consequences of PACT, and respondents’ suggested improvements for the PACT model.
Holistic Teamlet Experience PACT was implemented to provide employees with a teambased approach to care, coordinate their work, and define specific duties for each member. Several survey respondents expressed overall satisfaction with teamlets in terms of the impacts they have had for their teams and patients. One respondent stated: ‘‘Because of our [team] huddles and constant communication with our PACT team, we have taken initiatives to improve patient care.’’ Another identified the importance of the roles within team-based care: ‘‘Having an assigned RN, LPN, and clerk has improved medical delivery considerably.’’ Despite beliefs that PACT has benefits for patients and may be effective as a model for patient care, the majority of respondents articulated dissatisfaction with the implementation of PACT teamlets because of barriers such as understaffing and limited resources: ‘‘PACT model is awesome, but in order to make it work, each teamlet needs to be fully staffed. We are lucky that our team is staffed, but it’s not the same in every team and it really affects their measures and productivity.’’ Respondents described inconsistent implementation across PACT teamlets within the same clinic. One respondent who was on two different PACT teams described how the teams, ‘‘Ifunctioned significantly differently. One is very efficient and organized and the other, not so much. We all had the same initial education. It is interesting to see the difference.’’ Respondents identified a number of factors that limited PACT: (a) lack of appropriate staffing ratios and resources; (b) lack of sufficient training; (c) scheduling issues for teamlets and patients; (d) the distractions of telephones and computers; and (e) lack of satisfaction, clarity, and expectations in roles and responsibilities of all team members. Appropriate staffing ratios and resources. Respondents described the model as an ‘‘excellent philosophy’’ that could be ‘‘a great thing,’’ but the model cannot work without necessary resources. One participant wrote: ‘‘Ideally, PACT would be a good system; however, not realistic due to lack of staff and not enough hours in the day to complete the work that we are supposed to do.’’ If one staff member was absent, the rest of the team suffered because there were no back-up team members. Several respondents suggested that other roles be added to the teamlet, such as social workers, psychologists, and pharmacists. They described a need for increased staffing for these roles, particularly pharmacists: ‘‘[The] current ratio of CPS [clinical pharmacy specialists] to Teamlets [at seven or eight to one] is not sufficient for optimal pharmacotherapy management for our veterans, especially when a good number of our veterans are on 10-plus medications.’’
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
4
Health Care Management Review
Many respondents reported a lack of other resources, including space, time, and equipment in their clinics. As one respondent wrote: ‘‘Space has been the most negative factor in implementing the PACT modelIthere is not enough equipment, such as faxes [and] printers, available to send and get patient informationI.’’ Without these supports, some respondents felt unable to properly complete their duties and provide for their patients: ‘‘Currently we are limping along putting out fires and becoming burnt out in the process. We were never given the tools to succeed.’’ Lack of sufficient training. The majority of respondents reported a lack of thorough and consistent training for all team members prior to implementation in order to prepare them for the PACT model: ‘‘It would be helpful if the training could take place before changes are mandated. It’s a little like being dropped in the middle of a lake and told to swim when you’ve never done it.’’ Some expressed a need for better quality of training, describing the training to be a ‘‘waste of time and money.’’ Scheduling issues for teamlets and patients. Several respondents reported that appointment slots were overbooked and that administrative clerks felt overwhelmed with the number of calls, appointments, and walk-ins: ‘‘We have to become serious about not seeing walk-ins or being a ‘by appointment only’ clinic. The walk-ins are overwhelming and they have not been reduced.’’ There were also issues with lack of flexibility with scheduling visits of different durations: ‘‘No ability for the staff to adjust appointment times/slots to accommodate patients that need more/less time with the PCP.’’ Other concerns with shorter appointments included inadequate time for patients who require extensive care because of chronic pain and narcotic and other medication management. Several felt they lacked sufficient control over the clinic, including walk-ins, appointments, and their own schedules: ‘‘The clinic is chaos. Running at crisis level virtually all day every day.’’ Distractions of telephones and computers. Many participants reported unintended consequences of using a call center and computer-based reminders to improve patient care. Some felt that these constant reminders were a ‘‘waste of time’’ and ‘‘useless distractions’’ to their routine or to more important priorities. One reported that the ‘‘volume of clinical reminders and the volume of telephone care calls is a major burden.’’ This appeared to be a particular barrier for nurse care managers: ‘‘Currently phone calls are going into nurses’ officeVall phone calls should be diverted to central line so that the phone is not ringing all day in the nurses’ officeVclerical staff should be answering the phone calls.’’ Responding to telephone calls was not perceived as patient care: ‘‘Care managers should not be spending so much time creating reports for facility leadership and manning a nurse call line but instead, providing care management to patients.’’ Others supported the notion that more duties mean less patient care: ‘‘Since [the] PACT change, I have
Month & 2016
less interaction with the patients and feel ‘pushed’ all the time. I miss being with the patient.’’ Lack of satisfaction, clarity, and expectations in roles and responsibilities of all team members. Most respondents expressed a need for clear expectations for all team members, including leadership as well as accountability for leadership to do its part: ‘‘I think there should be accountability of management for the fact that our staffing is such that PACT cannot be implemented. Our veterans are hurt by their lack of attention and leadership.’’
Burnout The majority of respondents linked the barriers previously described to team member burnout. One respondent reported that ‘‘PACT has added many new duties without promised resources and has offered no new pay increases or incentives. It has brought on many inconsistencies from team to team and decreased job satisfaction.’’ Another respondent expressed worry over the burnout rates: ‘‘Burnout of my colleagues is a huge problem; I am worried we will lose some of our best providers.’’ RNs, LPNs, and administrative clerks in particular were reported to be ‘‘burning out rapidly.’’ Interestingly, this observation was made by respondents who identified as roles other than RNs, LPNs, and administrative clerks: ‘‘The biggest issue is if a Team Member is gone continuity of care becomes compromised and the workload may double or triple for at least one member of the teamVusually the RN or LPN.’’
Perceived Effect on Patients Respondents’ views on PACT’s impact on patient care were divided. Many reported that PACT resulted in positive improvements to patient care because patients are actively engaged in their care and feel heard and valued by their providers. Other respondents reported that patients may lose responsibility in their own care or may not fully understand or benefit from the model. Positive effects on patients. Many respondents expressed that patients ‘‘love’’ PACT and typically feel ‘‘more involved in their own care.’’ Some reported that PACT allowed for patients to be seen more quickly and that the teamlet was more accessible to the patient than with previous care models: ‘‘From the beginning of this PACT implementation, many veterans have really benefitted from it and they love the program [I] they can reach their teamlet, either through phone or secure messaging.’’ Those involved with PACT feel that the model allows for a better relationship with their patients: ‘‘It really builds relationships with Veterans. And you get to really know them as you case manage them, and help them get their needs met sooner rather than later.’’ The positive effects on patient care and negative effects on team members were not mutually exclusive: ‘‘PACT has
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
5
Implications of Transitions to Teamlets
been very helpful toward patient care; however, it has overburdened the staff.’’ Even some who expressed struggles with PACT and teamlets as being a ‘‘work in progress’’ acknowledged that, ‘‘Most importantly, the veterans are very satisfied with the change.’’ Negative effects on patients. Many respondents also expressed concern for the quality of patient care within the PACT model. They reported that, with the team-care model, patients lost responsibility for their own care because they either (a) did not fully understand the model or (b) believed that providers and nurses were supposed to take full responsibility for their health. One respondent particularly emphasized issue with medications: ‘‘Patients don’t understand PACT and how to be more responsible for something as simple as refilling medication. Many of the older vets don’t like using the telephone for reorders and prefer to have the PCP or Pharmacist reorder for them.’’ These respondents also expressed concern that the efforts to improve access had not necessarily succeeded. Patients still had to wait long periods of time or travel very far for care. Respondents seemed to feel that if a patient ‘‘has to wait 2 months to see [the doctor],’’ the model is not working as intended.
Level of Competency Within Teams Contrary to working to the top of their competency within the PACT, some respondents reported a growing burden of clerical work, particularly nurses: ‘‘I am an LPN but I feel like a glorified clerk most of the time. I do more clerical work than nursing.’’ Some employees felt they were required to work below their licensure, being caught under a large amount of paperwork that distracted from their training as a care provider. This appeared to lower job satisfaction and may cause tension within the clinic and between team members. In contrast, other respondents reported that the model allowed them to work to their full potential despite feeling overwhelmed and undertrained for the PACT model: ‘‘[I] will feel better once I have attended the PACT training. Teamwork has significantly improved with the PACT model and everyone is working to their highest training level.’’
The Unheard Voices of Team Members Quite a few employees felt undervalued for their work and felt caught in a system without being heard. One respondent wrote: ‘‘Listen to the concerns of the people in the trenches.’’ Others described feeling silenced when they offered suggestions to their leaders/managers in order to improve their situations: ‘‘When we have ideas to improve, leadership does not listen or work with us’’ and ‘‘[I] do not feel that upper management really understands what is going on in the clinics.’’ One respondent stated that ‘‘directives are always being changed and revised to the point that following them is cumbersome to providing actual care.’’ This appears to contribute to a sense of powerlessness:
There is a commanding and condescending tone when our leaders communicate with us. I feel as though we are not considered part of the process. One day we are simply told this is how we are doing such and such. We want more autonomy within our clinics.
Unintended Consequences of PACT Implementation Within team conflict and the effect of ‘‘weakest link.’’ Several respondents described problems within team functioning: ‘‘This [teamlet model] will not work for a slacker.’’ This suggests that if one team member did not fully participate, the entire team may suffer and/or lose efficacy. The team-based model is contingent upon all members working to their highest potential and fulfilling their responsibilities for their team and patients. A respondent wrote about the need for the teams to work together: ‘‘If one person isn’t doing what needs to be done it fails our whole team, and we need to be able to rely on each other!’’ Another stated, ‘‘The success of the teamlet is limited by the weakest link.’’ Some participants reported that the PACT model has encountered resistance from some employees who felt that the model of care they had before PACT worked more effectively. These employees may have been resistant to the changes of PACT, which may have resulted in some of the difficulties of implementing the model. One respondent noted some of their colleagues as ‘‘overworked, low-functioning nursing staff’’ in the clinic who do not ‘‘buy into PACT’’ and have therefore made implementing PACT difficult. Not only does the model require individuals to buy into the team and do their best, but each member should be held consistently to appropriate expectations. Another respondent expressed: ‘‘We need a full team, full time. We need enforcement and consistency of each staff level expectations. Clear expectations of all members across the board.’’ ‘‘Atmosphere of anxiety.’’ Numerous respondents described a perceived lack of consideration to ‘‘day-to-day variability’’ in the clinic and found it difficult to compensate for variability throughout the day. There was a concern that leadership seemed to care more about the number of patients seen rather than the quality of care patients receive: ‘‘All [leadership] cares about is the number of patients to be seen in the clinic, not the quality of care.’’ Similarly, some felt that PACT placed too much emphasis on performance metrics that did not reflect true quality of care: ‘‘PACT seems to be too number-driven. [It] creates an atmosphere of anxiety. Teams [are] trying to improve numbers at the risk of good patient care and individual patient evaluation.’’ Others reported fears of losing their job if they did not fully adhere to the PACT program, even if parts of the program did not work for their specific duties or the clinic: ‘‘We have been told that if we think we are working too hard that there are people waiting in line for our jobs.’’
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
6
Health Care Management Review
Respondents’ Suggested Improvements for PACT Model Increased training. A vast majority of respondents called for improved training including training for managing the PACT model and the respective roles within it. ‘‘The whole program should have been worked out entirely instead of a learn-as-you-go thing.’’ Many noted that training should provide all employees with the same expectations and instruction, which could reduce burnout, stress, and dissatisfaction. Care plans and specific PACTs for chronic pain patients. A few respondents suggested specific improvements to PACT related to chronic pain management. Particularly, they asked for standardized care plans for patients with chronic pain, an increased need for narcotic awareness, and generally more care for patients than a teamlet can offer. Some suggested a ‘‘comprehensive pain program and living with chronic pain with all the specialists available and other resources not to be managed by primary care’’ and ‘‘a PACT team just to manage patients with chronic pain issues.’’ Back-up teams. Back-up teams could provide teams with an ‘‘extra hand’’ or a back-up in cases of call-ins, vacations, and sick days. Respondents felt that, by creating back-up teams, the clinic could continue to run more smoothly, even when some employees are absent. This could avoid circumstances where teams might need help from other teams (because of missing members), especially if the second team is also overwhelmed, understaffed, and therefore unable to fully support the other team: ‘‘There needs to be a sub system for call offs/vacations so that teams are not pulled apart to other staff areas’’ and ‘‘at least two float nurses to help all our teamlets out in times of vacations, sickness, and just for an extra hand so we can meet the needs of our patients.’’
Discussion These findings suggest that, in the first 2 years of implementation, PACT has created burdens for PCPs and staff. At the same time, many participants believe PACT provides positive benefits for patients. We find remarkably similar findings to recent analyses from five VA Demonstration Laboratories, which have conducted in-depth, prospective evaluations of the PACT model finding that primary care clinicians and staff were enthusiastic about the concept of PACT but felt necessary resources were not available (Tuepker et al., 2014); that inadequate staffing resources and leadership engagement were frequently barriers to implementation and created major burdens on primary care personnel (True et al., 2012); and that training was not sufficient, particularly regarding team roles. However, in spite of all of these challenges, VA clinicians and staff felt that PACT had improved the care experience for the veteran (Rodriguez et al., 2014). Our findings offer corroboration from a national sample of clinics and argue that the experience of demonstration laboratory sites is generalizable even though the
Month & 2016
laboratories were selected competitively and received funding for testing and evaluating PACT. Similar findings have also been reported outside the VA. A study of facilitators and barriers to implementing a primary care team redesign to improve patient-centered care found that physician engagement and local leadership support were key factors that distinguished clinics that succeeded and those that struggled (Grace, Rich, Chin, & Rodriguez, 2014). Early on, the CMS Medical Home Demonstration noted that change fatigue was a serious threat from implementing so many transformational changes at the same time and that it affects even highly capable, highly motivated teams (Nutting et al., 2009). In addition, this paper highlights two new findings. First, PCMH team-based care may have unintended consequences of increasing stress and burnout. Notably, team members may feel dependent on the ‘‘weakest link’’ in their team, and efforts to improve access through greater direct telephone access may increase stress. Alternative methods of contact, such as telephone, and enhanced computer support are key components of PCMH models, but these findings suggest that, without the necessary staffing and scheduled time to address them, they can add to distraction and stress. Prior quantitative analyses of the PACT survey data found that specific elements of PACT team-based care, such as having a fully staff teamlet and participatory decision-making, were associated with significantly lower burnout (Helfrich et al., 2014); the present findings provide an important qualifier suggesting that a team-based PCMH model reduces burnout when effectively implemented, but that effective implementation is the exception rather than the rule. Second, frontline employees were eager to offer feedback and recommendations for improving PACT implementation. Even though the open-text survey field is a brief, impersonal mechanism for providing feedback and recommendations, respondents took it with alacrity (60% of respondents answered one or both questions). They made concrete recommendations: creation of back-up teams (instead of cannibalizing other PACT teamlets) to cover for absent teamlet members and development of PACT teamlets specifically for chronic care patients. We know of no other surveys of this scope to collect and analyze open-text data. This may represent an important method for future research. It may also be helpful to view these findings in the context of broader conceptual work on innovation implementation in health care and the factors that drive it. One of the most widely cited, comprehensive implementation models is the Consolidated Framework for Implementation Research (CFIR), a conceptual model, derived from literature synthesis, of the implementation process and the factors that drive it (Damschroder et al., 2009). The CFIR is organized in five broad categories: (a) the innovation, (b) inner organizational setting, (c) outer setting, (d) the individuals involved, and (e) the process by which implementation is accomplished. Most of our findings can be interpreted as falling within the
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
7
Implications of Transitions to Teamlets
inner organizational setting, which is the structural, political, and cultural contexts through which the implementation occurs and includes available resources (e.g., issues with PACT staffing ratios), access to knowledge and information necessary to implement the new practice (e.g., need for further training support for the PACT teamlets), and learning climate (which arose in the theme of unheard voices of the team and the feeling that leadership does not listen). These are important for two reasons. First, the inner context represents factors that are largely within the control of the organization and not inherent issues with the PCMH model. Issues with the PCMH model would be related to the inherent clinical effectiveness or compatibility of the medical home model in primary care (i.e., the CFIR construct of the innovation), the broader primary care environment (i.e., the CFIR construct of the outer setting), or the primary care personnel (i.e., the CFIR construct of the individuals). Second, our findings also suggest a potential modification to CFIR. Among our key themes were unintended consequences, such as the team model resulting in team members being dependent on the ‘‘weakest link.’’ This is not an outcome of poor implementation, such as the increased stress from lack of adequate staffing, rather a consequence of a team-based model. Unintended consequences are a familiar element of systems theory applied to health care (Scott, Mannion, Davies, & Marshall, 2003) but are neither part of CFIR or other widely cited implementation and diffusion models, such as Greenhalgh’s model based on a review of implementation and diffusion in service organizations (Greenhalgh, Robert, Macfarlane, Bate, & Kyriakidou, 2004), nor Roger’s Diffusion of Innovations theory (Rogers, 2003). Although CFIR depicts the individuals involved in implementation as fundamentally unpredictable, unintended consequences from the innovation itself are not expressly modeled in the CFIR. The strength of our study includes the large sample size of over 3,800 respondents representing more than 600 primary care clinics across the United States while retaining the advantage of qualitative research that uses the respondent’s own words describing the relationships among conditions, such as the barriers and facilitators of PACT implementation or how team-based care has contributed to a better patient experience for the veteran. Our sample is exclusively from VA primary care, which differs in many important structural respects from other systems that have implemented PCMH models. However, many of the core changes introduced by PACT, and commented on by respondents, relate to factors such as team-based care, logistical constraints, and resource demands, which should be widely applicable to other primary care settings.
Practice Implications Implementation of team-based care may, by its nature, introduce interdependencies that make team members vulnerable
to underperforming or undertrained teammates and makes the PCMH model particularly dependent on adequate staffing and training. Respondents suggested use of back-up teams to care for patients when PACT teamlet members were absent (rather than cannibalizing other teamlets). They also recommended formation of chronic disease-specific PACT teamlets for specific subpopulations, notably chronic pain patients. Although not feasible in all settings, these and other solutions might improve PACT implementation and be applicable to other PCMH settings.
Limitations There are three potential limitations. First, this study relied on secondary survey data in which the text fields followed a variety of structured questions, such as their level of burnout, which might have influenced the participant’s responses. Open-text fields offered limited space for responses, and the descriptions may be less detailed than they would be if implementation barriers and facilitators and burnout were the primary data collection topic. Another limitation was that the data were de-identified, and consequently subanalyses regarding respondent position were not possible. Finally, the survey was voluntary, and because of the way the survey was distributed, we have no way of knowing for certain which employees had an opportunity to respond or how representative they are. We do have comparative demographic data from a separate, general employee survey fielded the month preceding the PACT survey and have compared respondents in the four main PACT occupations (i.e., providers, nurse care managers, clinical associates, and administrative clerks). This comparison, previously published, found that the PACT survey sample had a slightly higher proportion of supervisors and lower proportion of African American respondents but was virtually identical on other characteristics, including tenure, age, and gender (Helfrich et al., 2014). Despite these limitations, our secondary analysis was appropriate because there was a good ‘‘fit’’ between our research question and the primary data (Hinds, Vogel, & Clarke-Steffen, 1997; open-ended narratives focused on PACT implementation), and the large sample size allowed us to obtain very saturated data. The sample is also exceptionally large in terms of clinics (n = 626). Previous PCMH often comes from small numbers of early adopters or demonstration sites that are likely not representative of the broader population of clinics (Jackson et al., 2013). Our sample represented over two thirds of VA primary care clinics across the United States. For comparison, in the Agency for Healthcare Research and Quality review, among the PCMH interventions that they determined had been rigorously reviewed, the number of clinics studied ranged from 6 to 18 in total (Peikes et al., 2012). Nevertheless, we do not know to what extent the observed teamlet member experiences are generalizable to other
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
8
Health Care Management Review
providers and staff implementing a PCMH model. Moreover, this analysis was not intended to quantify the categories. The large number of respondents and sites gives us added confidence that these categories reflect the breadth of PACT team member experience in the early implementation of this model; however, this analysis was not designed to compare the relative prevalence or importance of one category or subcategories over another. Finally, the data and conclusions only reflect the attitude of participants who responded to the survey questions and cannot be used to make conclusions regarding patient quality of care or outcomes. Future research should seek to explore the experience of PCMH teamlet members in other settings and with differing PCMH models and their effect on patient outcomes.
Conclusion We found significant challenges posed by the implementation of PACT, including unintended consequences that impacted team member job satisfaction. In addition, staffing and training in the model were perceived as inadequate, and nurses, in particular, found themselves spending a greater amount of time on clerical work than before the PACT model. These findings reflect the early experience of an initiative expected to take up to 5 years to implement. At the same time, respondents expressed support for the PACT model, and some felt it was having a positive impact on the quality of patient care. Respondents provided key recommendations for how these challenges can be addressed, notably more systematic training, more training for chronic pain management, and the use of back-up staff to help alleviate interruptions in teams’ functioning.. Acknowledgments
Our thanks to Julie Kurutz and John Witzlib of the VA Healthcare Analysis and Information Group for fielding the survey, to Rachel Orlando for administrative support, and to members of the PACT Demonstration Laboratory Coordinating Center Organizational Function Working Group who contributed invaluable expertise and feedback in survey development. This work was undertaken as part of the VA’s PACT Demonstration Laboratory initiative, supporting and evaluating VA’s transition to a PCMH model. Funding for the PACT Demonstration Laboratory initiative is provided by the VA Office of Patient Care Services. 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.. References Alexander, J. A., & Bae, D. (2012). Does the patient-centred medical home work? A critical synthesis of research on patient-
Month & 2016
centred medical homes and patient-related outcomes. Health Services Management Research, 25(2), 51Y59. doi: 10.1258/hsmr .2012.012001 Bodenheimer, T., & Laing, B. Y. (2007). The teamlet model of primary care. Annals of Family Medicine, 5(5), 457Y461. doi: 10 .1370/afm.731 Cimiotti, J. P., Aiken, L. H., Sloane, D. M., & Wu, E. S. (2012). Nurse staffing, burnout, and health care-associated infection. American Journal of Infection Control, 40(6), 486Y490. doi: 10 .1016/j.ajic.2012.02.029 Damschroder, L. J., Aron, D. C., Keith, R. E., Kirsh, S. R., Alexander, J. A., & Lowery, J. C. (2009). Fostering implementation of health services research findings into practice: A consolidated framework for advancing implementation science. Implementation Science, 4(1), 50. Elo, S., & Kynga¨s, H. (2008). The qualitative content analysis process. Journal of Advanced Nursing, 62(1), 107Y115. doi: 10 .1111/j.1365-2648.2007.04569.x Grace, S. M., Rich, J., Chin, W., & Rodriguez, H. P. (2014). Flexible implementation and integration of new team members to support patient-centered care. Healthcare, 2(2), 145Y151. doi: 10.1016/j.hjdsi.2014.02.003 Greenhalgh, T., Robert, G., Macfarlane, F., Bate, P., & Kyriakidou, O. (2004). Diffusion of innovations in service organizations: Systematic review and recommendations. Milbank Quarterly, 82(4), 581Y629. Halbesleben, J. R., Wakefield, B. J., Wakefield, D. S., & Cooper, L. B. (2008). Nurse burnout and patient safety outcomes: Nurse safety perception versus reporting behavior. Western Journal of Nursing Research, 30(5), 560Y577. doi: 10.1177/0193945907311322 Helfrich, C. D., Dolan, E. D., Simonetti, J., Reid, R. J., Joos, S., Wakefield, B. J., I Nelson, K. (2014). Elements of team-based care in a patient-centered medical home are associated with lower burnout among va primary care employees. Journal of General Internal Medicine, 29(Suppl 2), 659Y666. doi: 10.1007/s11606-013-2702-z Hinami, K., Whelan, C. T., Miller, J. A., Wolosin, R. J., & Wetterneck, T. B. (2012). Job characteristics, satisfaction, and burnout across hospitalist practice models. Journal of Hospital Medicine, 7(5), 402Y410. Hinds, P. S., Vogel, R. J., & Clarke-Steffen, L. (1997). The possibilities and pitfalls of doing a secondary analysis of a qualitative data set. Qualitative Health Research, 7(3), 408Y424. Jackson, G. L., Powers, B. J., Chatterjee, R., Prvu Bettger, J., Kemper, A. R., Hasselblad, V., I Williams, J. J. W. (2013). The patientcentered medical home: A systematic review. Annals of Internal Medicine, 158(3), 169Y178. doi: 10.7326/0003-4819-158-3-20130 2050-00579 Klein, S. (2011). The Veterans Health Administration: Implementing patient-centered medical homes in the nation’s largest integrated delivery system. New York, NY: The Commonwealth Fund. Landon, B. E., Reschovsky, J. D., Pham, H. H., & Blumenthal, D. (2006). Leaving medicine: The consequences of physician dissatisfaction. Medical Care, 44(3), 234Y242. Linzer, M., Manwell, L. B., Williams, E. S., Bobula, J. A., Brown, R. L., Varkey, A. B., I Horner-Ibler, B. (2009). Working conditions in primary care: Physician reactions and care quality. Annals of Internal Medicine, 151(1), 28Y36. Morse, J. M., Barrett, M., Mayan, M., Olson, K., & Spiers, J. (2008). Verification strategies for establishing reliability and validity in qualitative research. International Journal of Qualitative Methods, 1(2), 13Y22. Nutting, P. A., Crabtree, B. F., Miller, W. L., Stewart, E. E., Stange, K. C., & Jaen, C. R. (2010). Journey to the patientcentered medical home: A qualitative analysis of the experiences
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
9
Implications of Transitions to Teamlets
of practices in the National Demonstration Project. Annals of Family Medicine, 8(Suppl 1), S45YS56; S92. doi: 10.1370/ afm.1075 Nutting, P. A., Miller, W. L., Crabtree, B. F., Jaen, C. R., Stewart, E. E., & Stange, K. C. (2009). Initial lessons from the first national demonstration project on practice transformation to a patient-centered medical home. Annals of Family Medicine, 7(3), 254Y260. doi: 10.1370/afm.1002 Parker, P. A., & Kulik, J. A. (1995). Burnout, self- and supervisorrated job performance, and absenteeism among nurses. Journal of Behavioral Medicine, 18(6), 581Y599. Patient-Centered Primary Care Collaborative. (2007). Joint principles of the patient-centered medical home. Washington, DC: Author. Peikes, D., Zutshi, A., Genevro, J., Smith, K., Parchman, M., & Meyers, D. (2012). Early evidence on the patient-centered medical home. Rockville, MD: Agency for Healthcare Research and Quality. Rodriguez, H., Giannitrapani, K., Stockdale, S., Hamilton, A., Yano, E., & Rubenstein, L. (2014). Teamlet structure and early experiences of medical home implementation for veterans. Journal of General Internal Medicine, 29(2), 623Y631. doi: 10 .1007/s11606-013-2680-1 Rogers, E. (2003). Diffusion of innovations (5th ed.). New York, NY: Free Press. Rosland, A. M., Nelson, K., Sun, H., Dolan, E. D., Maynard, C., Bryson, C., I Schectman, G. (2013). The patient-centered medical home in the Veterans Health Administration. The American Journal of Managed Care, 19(7), e263Ye272. Sandelowski, M. (1986). The problem of rigor in qualitative research [comparative study]. ANS. Advances in Nursing Science, 8(3), 27Y37. Sargent, M. C., Sotile, W., Sotile, M. O., Rubash, H., & Barrack, R. L. (2009). Quality of life during orthopaedic training and academic practice. Part 1: Orthopaedic surgery residents and faculty. The Journal of Bone and Joint Surgery. American Volume, 91(10), 2395Y2405. doi: 10.2106/jbjs.h.00665 Scott, T., Mannion, R., Davies, H. T. O., & Marshall, M. N. (2003). Implementing culture change in health care: Theory and practice. International Journal for Quality in Health Care, 15(2), 111Y118. doi: 10.1093/intqhc/mzg021
Shanafelt, T., Sloan, J., Satele, D., & Balch, C. (2011). Why do surgeons consider leaving practice? Journal of the American College of Surgeons, 212(3), 421Y422. doi: 10.1016/j.jamcollsurg .2010.11.006 Shanafelt, T. D., Balch, C. M., Bechamps, G., Russell, T., Dyrbye, L., Satele, D., I Freischlag, J. (2010). Burnout and medical errors among American surgeons. Annals of Surgery, 251(6), 995Y1000. doi: 10.1097/SLA.0b013e3181bfdab3 Shanafelt, T. D., Boone, S., Tan, L., Dyrbye, L. N., Sotile, W., Satele, D., I Oreskovich, M. R. (2012). Burnout and satisfaction with work-life balance among us physicians relative to the general us population. Archives of Internal Medicine, 172(18), 1377Y1385. doi: 10.1001/archinternmed.2012.3199 Shanafelt, T. D., Bradley, K. A., Wipf, J. E., & Back, A. L. (2002). Burnout and self-reported patient care in an internal medicine residency program. Annals of Internal Medicine, 136(5), 358Y367. Spence Laschinger, H. K., & Leiter, M. P. (2006). The impact of nursing work environments on patient safety outcomes: The mediating role of burnout/engagement. The Journal of Nursing Administration, 36(5), 259Y267. True, G., Butler, A. E., Lamparska, B. G., Lempa, M. L., Shea, J. A., Asch, D. A., & Werner, R. M. (2012). Open access in the patient-centered medical home: Lessons from the Veterans Health Administration. In Journal of General Internal Medicine. doi: 10.1007/s11606-012-2279-y Tuepker, A., Kansagara, D., Skaperdas, E., Nicolaidis, C., Joos, S., Alperin, M., & Hickam, D. (2014). ‘‘We’ve not gotten even close to what we want to do’’: A qualitative study of early patientcentered medical home implementation. Journal of General Internal Medicine, 29(2), 614Y622. doi: 10.1007/s11606-013-2690-z Van Bogaert, P., Clarke, S., Roelant, E., Meulemans, H., & Van de Heyning, P. (2010). Impacts of unit
