Healthcare 2 (2014) 245–250
Contents lists available at ScienceDirect
Healthcare journal homepage: www.elsevier.com/locate/hjdsi
Implementing a patient centered medical home in the Veterans health administration: Perspectives of primary care providers Samantha L. Solimeo a,b,n, Kenda R. Stewart a,b,c, Gregory L. Stewart a,b,d, Gary Rosenthal a,b,e,f a
US Department of Veterans Affairs, VISN 23 Patient Aligned Care Team Demonstration Lab, Iowa City VA Health Care System, Iowa City, IA 52246, United States US Department of Veterans Affairs, Center for Comprehensive Access & Delivery Research and Evaluation, Iowa City VA Health Care System, 152, Iowa City, IA 52246, United States c US Department of Veterans Affairs, Veteran Rural health Resource Center Central Region, Iowa City VA Health Care System, Iowa City, IA 52246, United States d Tippie College of Business, University of Iowa, Iowa City, IA 52242, United States e Division of General Internal Medicine, University of Iowa Carver College of Medicine, Iowa City, IA 52242, United States f Department of Health Management and Policy, University of Iowa College of Public Health, Iowa City, IA 52242, United States b
art ic l e i nf o
a b s t r a c t
Article history: Received 22 October 2013 Received in revised form 26 June 2014 Accepted 15 July 2014 Available online 13 August 2014
Implementation of a patient centered medical home challenges primary care providers to change their scheduling practices to enhance patient access to care as well as to learn how to use performance metrics as part of a self-reflective practice redesign culture. As medical homes become more commonplace, health care administrators and primary care providers alike are eager to identify barriers to implementation. The objective of this study was to identify non-technological barriers to medical home implementation from the perspective of primary care providers. We conducted qualitative interviews with providers implementing the medical home model in Department of Veterans Affairs clinics—the most comprehensive rollout to date. Primary care providers reported favorable attitudes towards the model but discussed the importance of data infrastructure for practice redesign and panel management. Respondents emphasized the need for administrative leadership to support practice redesign by facilitating time for panel management and recognizing providers who utilize non-face-to-face ways of delivering clinical care. Health care systems considering adoption of the medical home model should ensure that they support both technological capacities and vertically aligned expectations for provider performance. Published by Elsevier Inc.
Keywords: Patient-centered care Primary health care Leadership Qualitative research Veterans health Health care quality, access, and evaluation Organizational culture Telemedicine Empanelment Patient aligned care team
1. Introduction The Patient Centered Medical Home model is designed to improve patient outcomes and primary care provider satisfaction through practice redesign that challenges primary care providers in multiple ways.1,2 To be successful, primary care providers practicing in medical homes must simultaneously develop strategies to provide patient-centered, preventative, acute, and chronic disease care in an accessible way.3–6 In 2010, the Veterans Health Administration (VHA) began implementing the medical home model throughout its primary care clinics.2,7 VHA's initiative represents the nation's largest effort to fundamentally redesign the delivery of primary care. In VHA, medical homes are organized around “Patient Aligned Care Teams” also known as “PACTs” that are comprised of a primary care provider who leads the team, and a clerical associate, registered nurse, and
n Corresponding author at: US Department of Veterans Affairs, VISN 23 Patient Aligned Care Team Demonstration Lab, Iowa City VA Health Care System, Iowa City, IA 52246, United States. Tel.: þ 1 319 338 0581x7637. E-mail address: [email protected] (S.L. Solimeo).
http://dx.doi.org/10.1016/j.hjdsi.2014.07.004 2213-0764/Published by Elsevier Inc.
licensed practical nurse. PACTs function independently, but multiple PACTs have shared relationships with specialist physicians and other health care practitioners, including social workers and pharmacists. Within the PACT model, primary care providers are accountable for patient outcomes and for identifying ways to improve access and other quality performance metrics. When administrative leadership enables teams to shift time from direct patient care, primary care providers are assumed to be able to create processes to improve quality of care and patient satisfaction, provide greater patient access, and reduce costs.8,9 Ideally, PACTs acquire time to complete this work by utilizing team members to the top of their skill set through: (1) delegation, and (2) by utilizing “encounterless encounters”10 (e.g. telephone visits, secure messaging) to provide patient care in lieu of traditional face-to-face visits. Nonetheless, in many clinics primary care provider workdays remain structured around face-to-face encounters, and facility variation in the interpretation of nursing scope of practice can limit primary care provider within-team delegation.3,8,11–13 As part of a larger project studying team function we conducted qualitative interviews with members of 22 teams implementing PACT as part of a pilot launch. In the current manuscript we report
246
S.L. Solimeo et al. / Healthcare 2 (2014) 245–250
on data collected from primary care providers in these teams, whose role as team leaders is accompanied by responsibility for PACT implementation. As VHA primary care providers practice in settings already equipped with an integrated pharmacy, secure messaging, multiple telehealth platforms, and an electronic medical record, our objective was to identify non-technological challenges to providers' transition to the medical home care model.
technological barriers to implementation were encompassed by two domains: (1) primary care utilization of virtual encounters; and (2) the integration of performance metrics into team operations. Coded data for these two domains were then reviewed independently, refined, and exemplar passages selected for comparison. After comparison and further refinement, the anthropologists reviewed the primary care provider interviews at one-year follow up to assess change and sustainment.
2. Materials and methods 3. Results 2.1. Sample Twenty-two primary care providers participating in a PACT Learning Collaborative organized by the VA Midwest Health Care Network (VISN 23) located in the upper Midwestern United States, were eligible for the study. Primary care providers were recruited to participate by e-mail in a voluntary, confidential, one-on-one semi-structured interview six months after the initial Learning Collaborative. Twelve primary care providers agreed to participate in the initial interview; nine of these agreed to participate in a second interview one year later. Detailed description of the larger evaluation14 and the Learning Collaborative are reported elsewhere.12 2.2. Data collection In order to balance our desire for respondents to speak about issues that they felt were important with our need to have standardization across interviews, we used a semi-structured interview format.15 Interview topics were informed by: (1) review of published literature on medical home implementation, which identified a number of key concepts related to medical home implementation (e.g., enhanced access, care coordination, continuity of care, role of practice facilitators); (2) review of PACT training materials, which emphasize team function, delegation, and practice redesign; and (3) the research team's interest in understanding the within-team processes driving team function, such as role negotiation, delegation, and job satisfaction. Draft interview questions were developed, reviewed for face validity by the VISN 23 PACT Demonstration Lab leadership, and then revised for clarity and feasibility. See Fig. 1. All interviews were audio-recorded and designed to be completed in 30–60 min. 2.3. Analysis The interviews were transcribed by experienced transcriptionists, audited for accuracy, and imported to a qualitative data software platform (MAXQDA16) to facilitate analysis. Pairs of trained qualitative analysts coded transcripts to an 80% agreement benchmark, facilitated by the software to enhance coding reliability. The initial codebook was written using deductive parameters derived from the literature and interview guide. The research team developed additional codes through inductive review of the interview transcripts. The lead investigator (SLS) adjudicated disagreements. To understand primary care provider perceptions of PACT implementation demands, two anthropologists (SLS and KRS) analyzed interview data coded for primary care provider role, enhanced access practices, panel management, chronic disease management, data quality, and implementation barriers. Each anthropologist reviewed these coded data and independently developed a list of the effects of implementation on primary care provider practice and perceived implementation barriers. After the initial list development, the anthropologists met and identified broad themes to characterize primary care providers' implementation experience. A majority of respondents' discussion of non-
Table 1 summarizes participant characteristics. A majority (n ¼7) of participating primary care providers had worked at their current practice site for less than five years. Interviews ranged in length from 30–75 min. Despite being scheduled during providers' administrative time, interviews were frequently rescheduled to accommodate patient appointments or otherwise interrupted, underscoring the ubiquity of time pressures. 3.1. Primary care providers generally positive towards PACT Overall, respondents described enthusiasm for the model and expressed their buy-in as a result of being allowed to provide care in a way that they had long desired: “I think that because I'm used to this and this honestly is how I've always wanted to work, but the one glitch was always: ‘How do I clear a schedule when, if you do that somebody's going to think you don't want to see the patient?’ So now with this PACT I'm where I can delegate. My LPNs call back when I need call backs. They all give me information” (Primary Care Provider 1). Despite an expressed enthusiasm, a majority of respondents related concerns regarding the challenge of bringing clinical practice into alignment with the ideals of the PACT model. Of particular note were primary care provider observations that VHA's emphasis on bottom-up, team driven implementation (as opposed to centrally directed mandates) was a positive feature. However, some participants explained that being directed to innovate was intrinsically at odds with the scale of implementation and led them to question the ability of VHA to sustain PACT implementation efforts over time. Some participants also discussed the importance of having VHA leadership provide clear communication of PACT implementation goals and endpoints. In the context of implementation, primary care providers reported a sense of increasing expectations to provide more services to each patient, while assuming responsibility for a larger patient panel. Discussion of increasing panel size was connected to respondents' belief that implementation increased administrative leadership's scrutiny of their performance. Providers reported concerns with the ability of existing metrics to accurately reflect primary care provider performance and workload. These concerns constituted two primary themes common to the primary care provider interviews: (1) “mixed messages” primary care providers received regarding encounterless encounters; and (2) using performance metrics to redesign practice. A brief description of the overall trends within each theme as well as subtle differences in primary care provider perceptions between the initial and follow-up interviews follows. 3.2. Providers cannot embrace enhanced access strategies if rewarded for a full schedule Once PACTs had been formed, their first practice redesign efforts focused on improving access. While some providers
S.L. Solimeo et al. / Healthcare 2 (2014) 245–250
247
Fig. 1. Interview guide.
Table 1 Characteristics of primary care providers participating in the initial and follow-up interviews. Characteristic
Initial Followup
Lost to FollowUp
Participants, No. Medical Training MD PA APRN/ NP Number of female participants Mean number of years at practice site Practice setting Facility type Health Care System (e.g., hospital) Community Based Outpatient Clinic Rural Classification Urban Rural Mean number of PACT training events attendeda
12
9
3
7 2 3 7 7.0
6 2 1 5 6.9
1 0 2 2 7.3
6 6
6 3
0 3
9 3 2.1
8 1 2.0
1 2 2.3
a
Range for number of PACT training events attended¼ 0–5 of 5 possible.
emphasized work-flow and scheduling strategies (e.g., pre-visit calling to reduce in-clinic time), primary care leadership and PACT training emphasized the important role of encounterlesss encounters in providing patient-centered care through enhanced access. Primary care provider attitudes toward non face-to-face mechanisms for providing care were generally
positive, and as this passage exemplifies, teams integrated several strategies: “We used to have just any old sort of LPN—whoever was available to grab the patient—and now we have dedicated primary care LPNs on our team. … And they make calls to patients the day before and do as many reminders [as they can]—remind them of their appointment, remind them to take their blood pressure medicine—the day before. … We're starting to try telephone clinics. We've been using secure messaging” (Primary Care Provider 2). However, primary care providers also reported that their efforts to use encounterlesss encounters were at times limited by “mixed messages” from leadership. For example, primary care providers who created open access by utilizing encounterless encounters found themselves scheduled for face-to-face visits with patients of other providers whose schedules were too full to accommodate more visits. Providers described this as a lack of “protected” schedules and as a disconnect between organizational encouragement of non-face-to-face encounters and penalties for having unused appointments. For example: “Some of our performance pay money depends on the percentage of open slots we have available… Part of the issue there is I think… not having the administrative goals aligned with the provider or clinic goals” (Primary Care Provider 4). “I think the biggest change is that I have the ability now to get some people off the schedule that didn't need to be there… Again the problem is—because I'm able to do that and some of the others
248
S.L. Solimeo et al. / Healthcare 2 (2014) 245–250
aren't—my schedule gets filled up with other providers' patients which ruins the continuity of care. What I hear [from administrative leadership] is if there isn't a room in anybody else's [schedule] of course we have to see the patient” (Primary Care Provider 1). The conflicting messages from leadership that primary care providers perceived were also evident in the second major theme of primary care provider perspectives: challenges of using performance metrics to inform practice redesign.
“At [our facility], we're relatively focused on, “Well, Doctor A sees twelve patients a day and Doctor B only sees six patients. Well, Doctor B's got to work harder and see twelve patients a day.” Well, maybe not. Maybe Doctor B is actually doing the right thing and Doctor A needs to lighten up and not bring them back every time… Maybe he needs to utilize his RNs, LPNs or other support people a little bit more” (Primary Care Provider 8).
4. Discussion 3.3. Performance metrics must be sensitive to practice PACT training emphasizes the importance of proactive patient panel management to practice redesign and improving patient outcomes. Despite training, primary care providers reported instrumental challenges (e.g., locating specific metrics, administrative process involved in accessing data). Even those who recounted being able to readily access performance metrics identified other potential problems that limited use for practice redesign. For example, primary care providers noted that the time lag between practice redesign activities and data availability made it difficult to assess the impact of process changes. Other primary care providers remarked that some metrics were only available at a facility level, which prevented providers from isolating the effect of their efforts from those of other clinic providers. As this primary care provider remarked, “I mean the original information was facility level and with [many] providers, I have a very small input. I just throw [that metric] out because it doesn't reflect at all what we're doing” (Primary Care Provider 4). Primary care providers also believed that some metrics did not accurately reflect workload, whereas others were not actionable: “I think it would be nice to know exactly what data they're pulling on our teams and what are they looking at specifically…I would hate for us to be doing all this work, and then for them not to capture our work. That's one kind of fear that I have. … I guess that's just probably one of my bigger concerns because we are trying hard and we're trying to work as our PACT. I just don't want that to go unnoticed for something stupid like we didn't set [a clinic] up correctly” (Primary Care Provider 5). “I have influence over whether patients are seen in my clinic only during business hours and during regular days of the week—when our clinic is open. So if you include patients that come in [during] off-business hours or on weekends or holidays, that's part of the overall [emergency room] utilization rate, but I have no control over that, you know? I have no way to see patients in the evenings. I have no way to see patients on weekends. If they come in at that time, it's their choice and I don't have any control over that” (Primary Care Provider 4). 3.4. Change and sustainment over time Comparison of primary care provider responses between the initial interview and the follow-up interviews conducted after primary care providers had additional experience practicing in the PACT model revealed little difference in their attitudes. At 18 months post-implementation providers reported the same “mixed messages” and performance metrics barriers to PACT implementation as they did at the 6 month baseline interviews. However, there was an overarching, but subtle shift from referring to PACT as a new project to references such as the following which signaled adoption of PACT:
The current analysis was conducted to identify primary care providers' perceived non-technological barriers to PACT implementation. Medical home implementation is a comprehensive process which compels multi-level organizational change. Although participants discussed incorporation of technologies such as administrative data, scheduling systems, and encounterless encounters, the barriers to utilizing such tools were primarily organizational and not technological in nature. Qualitative research with VHA primary care providers implementing the PACT model in a large Midwestern region demonstrates the importance of multi-level alignment of implementation ideals with organizational practices.14 The primary care providers we interviewed were generally enthusiastic about the potential of the PACT model to improve care, and a majority readily adopted PACT strategies such as encounterless encounters. However, providers who were more successful at creating open access often experienced negative reinforcement from having to see patients of other primary care providers and from performance metrics that rewarded providers for full appointment calendars regardless of whether those appointments were in line with PACT ideals. In a related vein, primary care providers' efforts to use performance metrics in redesigning practice were limited not by the technological challenges that private sector practices contend with, but by lack of computer savvy required to find the data or by higher level data quality issues, such as concerns about data specificity and sensitivity. Another challenge was the belief that performance metrics were not equally actionable (e.g. trying to reduce emergency department use while not being able to expand clinic hours). These reported challenges were experienced in a context of both perceived greater provider accountability for patient outcomes and a bottom-up implementation approach, which likely contributed to primary care providers' report of need for enhanced and clarified communication from leadership as to PACT goals and expectations. Recent publications reporting on VHA PACT implementation emphasize the importance of supportive leadership on PACT team function,1,12,14 particularly with regard to sustaining same day access gains3 and improving panel management. As Murray and Tantau reported in 2000, primary care providers' efforts to provide continuity and same-day access must be organizationally valued and protected.17 Relationships between the time pressures experienced by providers and delegation are also important. A time allocation study conducted in VHA prior to PACT implementation determined that primary care providers averaged 8.6 h per week on patient care work outside of office visits.18 In a context of limited time, providers struggle to manage the additional workload19 and to prioritize direct patient care work over panel management activities.20 Underutilized intra-team delegation is a related component of time limitations,21 but a sense of obligation may reduce primary are providers' willingness and ability to share work with teammates.22 As Saba and colleagues have noted, “In our profession's efforts to create [high functioning health care teams] we have yet to articulate a myth that takes us from the lone physician to the team-based physician.”23 The lack of a “myth” to
S.L. Solimeo et al. / Healthcare 2 (2014) 245–250
foster PACT function may be in part due to a lag between the speed of individuals to adopt change and the speed of the larger organization to align leadership goals at multiple levels. While the PACT model compels multi-level organizational change, implementation emphasized bottom-up innovation. Administrative leadership at all levels should be integrated into medical home implementation efforts. The current study recruited PACT primary care providers involved in the first wave of implementation, facilitated by an intensive learning collaborative modeled after the Institute for Healthcare Improvement framework,12,24 making the data potentially biased towards the experiences of better trained providers. Thus, challenges reported among our sample may be more severe among providers who did not receive this focused training. We also were not able to link participants' perceptions to their actual performance metrics, which may have yielded a more balanced understanding of the challenges they reported.
249
Acknowledgments The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the United States government. This work could not have been done without the support of the VISN 23 Patient Aligned Care Team Demonstration Laboratory, supported by the Office of Patient Care Services, Department of Veterans Affairs. Additional support comes from the Center for Comprehensive Access & Delivery Research and Evaluation (CADRE) (Grant no. CIN 13-412), Department of Veterans Affairs, Iowa City VA Health Care System, Iowa City, IA. We wish to acknowledge technical assistance from the VA Midwest Health Care Network, Primary Care and Specialty Medicine Service Line and the VISN 23 PACT Collaborative teams.
References 5. Conclusions As an integrated medical system with a sophisticated electronic medical record, clinical reminder infrastructure, and pharmacy, VHA has tremendous potential to leverage the cost and quality improvements promised by the medical home model. Yet these apparent advantages which may lead casual observers to regard the PACT implementation as unique should be weighed together with the scope of the VHA project: PACT is being implemented in all VHA primary care settings among an extremely diverse primary care workforce rather than in highly selected clinics. What we learn from PACT has broad applicability to private sector settings and reveals the true challenges that underlie the surface dressing of technological infrastructure. Regardless of VHA's infrastructural and fiscal advantages over traditional fee for service health care organizations, providers implementing PACT may be challenged by an initial misalignment between PACT ideals and certain larger organizational policies and performance measures. That providers in this study reported the same barriers to PACT implementation at 6 and 18 months after launch is indicative of the profound role of organizational culture in everyday practice. Misalignment and other challenges may be surmounted over time as the organization more tightly aligns metrics of PACT success with the overarching goals of the PACT model and with the realities of clinical practice. The current study indicates the importance of such efforts in supporting primary care provider who are making the difficult transition to medical home adoption.
Support This work received financial support from the VISN 23 Patient Aligned Care Team Demonstration Lab, Iowa City VA Health Care System, Iowa City, IA, which is funded by the VA Office of Patient Care Services. Additional support comes from the Center for Comprehensive Access & Delivery Research and Evaluation (CADRE), Department of Veterans Affairs, Iowa City VA Health Care System, Iowa City, IA.
Conflicts of interest Samantha L. Solimeo, PhD, MPH: no conflicts to report. Kenda R. Stewart, PhD: no conflicts to report. Gregory L. Stewart, PhD: no conflicts to report. Gary Rosenthal, MD: no conflicts to report.
1. Reid R, Wagner E. The Veterans Health Administration Patient Aligned Care Teams: lessons in primary care transformation. J Gen Intern Med. 2014;29 (2):552–554, http://dx.doi.org/10.1007/s11606-014-2827-8. 2. Rosland A, Nelson K, Sun H, et al. The patient-centered medical home in the Veterans Health Administration. Am J Manag Care. 2013;19(7):e262–e272. 3. True G, Butler AE, Lamparska BG, et al. Open access in the patient-centered medical home: lessons from the Veterans Health Administration. J Gen Intern Med. 2013;28(4):539–545, http://dx.doi.org/10.1007/s11606-012-2279-y. 4. Jackson GL, Powers BJ, Chatterjee R, et al. Patient-centered medical home: a systematic review. Ann Intern Med. 2013;158(3):169–178, http://dx.doi.org/ 10.7326/0003-4819-158-3-201302050-00579. 5. Landon BE. Moving ahead with the PCMH: some progress, but more testing needed. J Gen Intern Med. 2013;28(6):753–755, http://dx.doi.org/10.1007/ s11606-013-2434-0. 6. Leasure EL, Jones RR, Meade LB, et al. There Is no I in teamwork in the patient-centered medical home: defining teamwork competencies for academic practice. Acad Med. 2013;88(5):585–592, http://dx.doi.org/10.1097/ACM. 0b013e31828b0289. 7. Schectman G, Stark R. Orchestrating large organizational change in primary care: the Veterans' Health Administration experience implementing a patientcentered medical home. J Gen Intern Med. 2014;29(2):550–551, http://dx.doi. org/10.1007/s11606-014-2828-7. 8. Meyer H. Group health's move to the medical home: for doctors, it's often a hard journey. Health Aff. 2010;29(5):844–851, http://dx.doi.org/10.1377/ hlthaff.2010.0345. 9. Parker LE, Kirchner JE, Bonner LM, et al. Creating a quality-improvement dialogue: utilizing knowledge from frontline staff, managers, and experts to foster health care quality improvement. Qual Health Res. 2009;19(2):229–242, http://dx.doi.org/10.1177/1049732308329481. 10. Fortney J, Burgess J, Bosworth H, Booth B, Kaboli P. A re-conceptualization of access for 21st century healthcare. J Gen Intern Med.. 2011;26(0):639–647 http://dx.doi.org/10.1007/s11606-011-1806-6. 11. Casalino LP. A martian's prescription for primary care: overhaul the physician's workday. Health Aff. 2010;29(5):785–790, http://dx.doi.org/10.1377/hlthaff. 2010.0133. 12. Solimeo SL, Hein M, Paez M, Ono S, Lampman M, Stewart GL. Medical homes require more than an EMR and aligned incentives. Am J Manag Care. 2013;19 (2):132–140. 13. Charles-Jones H, Latimer J, May C. Transforming general practice: the redistribution of medical work in primary care. Sociol Health Illn. 2003;25(1):71–92. 14. True G, Stewart G, Lampman M, Pelak M, Solimeo SL. Teamwork and delegation in medical homes: primary care staff perspectives in the Veterans Health Administration. J Gen Intern Med. 2014;29(2):632–639, http://dx.doi.org/ 10.1007/s11606-013-2666-z. 15. Britten N, Jones R, Murphy E, Stacy R. Qualitative research methods in general practice and primary care. Fam Pract. 1995;12(1):104–114, http://dx.doi.org/ 10.1093/fampra/12.1.104. 16. MAXQDA 10. Software for Qualitative Data Analysis. ([Computer Program]). Berlin, Germany: VERBI Software-Consult-Sozialforschung GmbH; 1998–2012. 17. Murray M, Tantau C. Same-day appointments: exploding the access paradigm. Fam Pract Manag. 2000;7(8):45–50. 18. Doerr E, Galpin K, Jones-Taylor C, et al. Between-visit workload in primary care. J Gen Intern Med. 2010;25(12):1289–1292, http://dx.doi.org/10.1007/s11606010-1470-2. 19.. Neuwirth EEB, Schmittdiel JA, Tallman K, Bellows J. Understanding panel management: a comparative study of an emerging approach to population care. Perm J. 2007;11(3):12–20. 20. Østbye T, Yarnall KS, Krause KM, Pollak KI, Gradison M, Michener JL. Is there time for management of patients with chronic diseases in primary care? Ann Fam Med. 2005;3:209–214.
250
S.L. Solimeo et al. / Healthcare 2 (2014) 245–250
21. Taylor EF, Machta RM, Meyers DS, Genevro J, Peikes DN. Enhancing the primary care team to provide redesigned care: the roles of practice facilitators and care managers. Ann Fam Med. 2013;11:80–83, http://dx.doi.org/10.1370/afm.1462. 22. Dingwall R. Problems of teamwork in primary care. In: Lonsdale S, Webb A, Briggs T, editors. Teamwork in the Personal Social Services and Health Care. London: Croom Helm; 1980. p. 111–137.
23. Saba GW, Villela TJ, Chen E, Hmmer H, Bodenheimer T. The myth of the lone physician: toward a collaborative alternative. Ann Fam Med. 2012;10:169–173, http://dx.doi.org/10.1370/afm.1353. 24. Institute for Healthcare Improvement. The Breakthrough Series: IHI's Collaborative Model for Achieving Breakthrough Improvement. Boston, MA; 2003.
Contents lists available at ScienceDirect
Healthcare journal homepage: www.elsevier.com/locate/hjdsi
Implementing a patient centered medical home in the Veterans health administration: Perspectives of primary care providers Samantha L. Solimeo a,b,n, Kenda R. Stewart a,b,c, Gregory L. Stewart a,b,d, Gary Rosenthal a,b,e,f a
US Department of Veterans Affairs, VISN 23 Patient Aligned Care Team Demonstration Lab, Iowa City VA Health Care System, Iowa City, IA 52246, United States US Department of Veterans Affairs, Center for Comprehensive Access & Delivery Research and Evaluation, Iowa City VA Health Care System, 152, Iowa City, IA 52246, United States c US Department of Veterans Affairs, Veteran Rural health Resource Center Central Region, Iowa City VA Health Care System, Iowa City, IA 52246, United States d Tippie College of Business, University of Iowa, Iowa City, IA 52242, United States e Division of General Internal Medicine, University of Iowa Carver College of Medicine, Iowa City, IA 52242, United States f Department of Health Management and Policy, University of Iowa College of Public Health, Iowa City, IA 52242, United States b
art ic l e i nf o
a b s t r a c t
Article history: Received 22 October 2013 Received in revised form 26 June 2014 Accepted 15 July 2014 Available online 13 August 2014
Implementation of a patient centered medical home challenges primary care providers to change their scheduling practices to enhance patient access to care as well as to learn how to use performance metrics as part of a self-reflective practice redesign culture. As medical homes become more commonplace, health care administrators and primary care providers alike are eager to identify barriers to implementation. The objective of this study was to identify non-technological barriers to medical home implementation from the perspective of primary care providers. We conducted qualitative interviews with providers implementing the medical home model in Department of Veterans Affairs clinics—the most comprehensive rollout to date. Primary care providers reported favorable attitudes towards the model but discussed the importance of data infrastructure for practice redesign and panel management. Respondents emphasized the need for administrative leadership to support practice redesign by facilitating time for panel management and recognizing providers who utilize non-face-to-face ways of delivering clinical care. Health care systems considering adoption of the medical home model should ensure that they support both technological capacities and vertically aligned expectations for provider performance. Published by Elsevier Inc.
Keywords: Patient-centered care Primary health care Leadership Qualitative research Veterans health Health care quality, access, and evaluation Organizational culture Telemedicine Empanelment Patient aligned care team
1. Introduction The Patient Centered Medical Home model is designed to improve patient outcomes and primary care provider satisfaction through practice redesign that challenges primary care providers in multiple ways.1,2 To be successful, primary care providers practicing in medical homes must simultaneously develop strategies to provide patient-centered, preventative, acute, and chronic disease care in an accessible way.3–6 In 2010, the Veterans Health Administration (VHA) began implementing the medical home model throughout its primary care clinics.2,7 VHA's initiative represents the nation's largest effort to fundamentally redesign the delivery of primary care. In VHA, medical homes are organized around “Patient Aligned Care Teams” also known as “PACTs” that are comprised of a primary care provider who leads the team, and a clerical associate, registered nurse, and
n Corresponding author at: US Department of Veterans Affairs, VISN 23 Patient Aligned Care Team Demonstration Lab, Iowa City VA Health Care System, Iowa City, IA 52246, United States. Tel.: þ 1 319 338 0581x7637. E-mail address: [email protected] (S.L. Solimeo).
http://dx.doi.org/10.1016/j.hjdsi.2014.07.004 2213-0764/Published by Elsevier Inc.
licensed practical nurse. PACTs function independently, but multiple PACTs have shared relationships with specialist physicians and other health care practitioners, including social workers and pharmacists. Within the PACT model, primary care providers are accountable for patient outcomes and for identifying ways to improve access and other quality performance metrics. When administrative leadership enables teams to shift time from direct patient care, primary care providers are assumed to be able to create processes to improve quality of care and patient satisfaction, provide greater patient access, and reduce costs.8,9 Ideally, PACTs acquire time to complete this work by utilizing team members to the top of their skill set through: (1) delegation, and (2) by utilizing “encounterless encounters”10 (e.g. telephone visits, secure messaging) to provide patient care in lieu of traditional face-to-face visits. Nonetheless, in many clinics primary care provider workdays remain structured around face-to-face encounters, and facility variation in the interpretation of nursing scope of practice can limit primary care provider within-team delegation.3,8,11–13 As part of a larger project studying team function we conducted qualitative interviews with members of 22 teams implementing PACT as part of a pilot launch. In the current manuscript we report
246
S.L. Solimeo et al. / Healthcare 2 (2014) 245–250
on data collected from primary care providers in these teams, whose role as team leaders is accompanied by responsibility for PACT implementation. As VHA primary care providers practice in settings already equipped with an integrated pharmacy, secure messaging, multiple telehealth platforms, and an electronic medical record, our objective was to identify non-technological challenges to providers' transition to the medical home care model.
technological barriers to implementation were encompassed by two domains: (1) primary care utilization of virtual encounters; and (2) the integration of performance metrics into team operations. Coded data for these two domains were then reviewed independently, refined, and exemplar passages selected for comparison. After comparison and further refinement, the anthropologists reviewed the primary care provider interviews at one-year follow up to assess change and sustainment.
2. Materials and methods 3. Results 2.1. Sample Twenty-two primary care providers participating in a PACT Learning Collaborative organized by the VA Midwest Health Care Network (VISN 23) located in the upper Midwestern United States, were eligible for the study. Primary care providers were recruited to participate by e-mail in a voluntary, confidential, one-on-one semi-structured interview six months after the initial Learning Collaborative. Twelve primary care providers agreed to participate in the initial interview; nine of these agreed to participate in a second interview one year later. Detailed description of the larger evaluation14 and the Learning Collaborative are reported elsewhere.12 2.2. Data collection In order to balance our desire for respondents to speak about issues that they felt were important with our need to have standardization across interviews, we used a semi-structured interview format.15 Interview topics were informed by: (1) review of published literature on medical home implementation, which identified a number of key concepts related to medical home implementation (e.g., enhanced access, care coordination, continuity of care, role of practice facilitators); (2) review of PACT training materials, which emphasize team function, delegation, and practice redesign; and (3) the research team's interest in understanding the within-team processes driving team function, such as role negotiation, delegation, and job satisfaction. Draft interview questions were developed, reviewed for face validity by the VISN 23 PACT Demonstration Lab leadership, and then revised for clarity and feasibility. See Fig. 1. All interviews were audio-recorded and designed to be completed in 30–60 min. 2.3. Analysis The interviews were transcribed by experienced transcriptionists, audited for accuracy, and imported to a qualitative data software platform (MAXQDA16) to facilitate analysis. Pairs of trained qualitative analysts coded transcripts to an 80% agreement benchmark, facilitated by the software to enhance coding reliability. The initial codebook was written using deductive parameters derived from the literature and interview guide. The research team developed additional codes through inductive review of the interview transcripts. The lead investigator (SLS) adjudicated disagreements. To understand primary care provider perceptions of PACT implementation demands, two anthropologists (SLS and KRS) analyzed interview data coded for primary care provider role, enhanced access practices, panel management, chronic disease management, data quality, and implementation barriers. Each anthropologist reviewed these coded data and independently developed a list of the effects of implementation on primary care provider practice and perceived implementation barriers. After the initial list development, the anthropologists met and identified broad themes to characterize primary care providers' implementation experience. A majority of respondents' discussion of non-
Table 1 summarizes participant characteristics. A majority (n ¼7) of participating primary care providers had worked at their current practice site for less than five years. Interviews ranged in length from 30–75 min. Despite being scheduled during providers' administrative time, interviews were frequently rescheduled to accommodate patient appointments or otherwise interrupted, underscoring the ubiquity of time pressures. 3.1. Primary care providers generally positive towards PACT Overall, respondents described enthusiasm for the model and expressed their buy-in as a result of being allowed to provide care in a way that they had long desired: “I think that because I'm used to this and this honestly is how I've always wanted to work, but the one glitch was always: ‘How do I clear a schedule when, if you do that somebody's going to think you don't want to see the patient?’ So now with this PACT I'm where I can delegate. My LPNs call back when I need call backs. They all give me information” (Primary Care Provider 1). Despite an expressed enthusiasm, a majority of respondents related concerns regarding the challenge of bringing clinical practice into alignment with the ideals of the PACT model. Of particular note were primary care provider observations that VHA's emphasis on bottom-up, team driven implementation (as opposed to centrally directed mandates) was a positive feature. However, some participants explained that being directed to innovate was intrinsically at odds with the scale of implementation and led them to question the ability of VHA to sustain PACT implementation efforts over time. Some participants also discussed the importance of having VHA leadership provide clear communication of PACT implementation goals and endpoints. In the context of implementation, primary care providers reported a sense of increasing expectations to provide more services to each patient, while assuming responsibility for a larger patient panel. Discussion of increasing panel size was connected to respondents' belief that implementation increased administrative leadership's scrutiny of their performance. Providers reported concerns with the ability of existing metrics to accurately reflect primary care provider performance and workload. These concerns constituted two primary themes common to the primary care provider interviews: (1) “mixed messages” primary care providers received regarding encounterless encounters; and (2) using performance metrics to redesign practice. A brief description of the overall trends within each theme as well as subtle differences in primary care provider perceptions between the initial and follow-up interviews follows. 3.2. Providers cannot embrace enhanced access strategies if rewarded for a full schedule Once PACTs had been formed, their first practice redesign efforts focused on improving access. While some providers
S.L. Solimeo et al. / Healthcare 2 (2014) 245–250
247
Fig. 1. Interview guide.
Table 1 Characteristics of primary care providers participating in the initial and follow-up interviews. Characteristic
Initial Followup
Lost to FollowUp
Participants, No. Medical Training MD PA APRN/ NP Number of female participants Mean number of years at practice site Practice setting Facility type Health Care System (e.g., hospital) Community Based Outpatient Clinic Rural Classification Urban Rural Mean number of PACT training events attendeda
12
9
3
7 2 3 7 7.0
6 2 1 5 6.9
1 0 2 2 7.3
6 6
6 3
0 3
9 3 2.1
8 1 2.0
1 2 2.3
a
Range for number of PACT training events attended¼ 0–5 of 5 possible.
emphasized work-flow and scheduling strategies (e.g., pre-visit calling to reduce in-clinic time), primary care leadership and PACT training emphasized the important role of encounterlesss encounters in providing patient-centered care through enhanced access. Primary care provider attitudes toward non face-to-face mechanisms for providing care were generally
positive, and as this passage exemplifies, teams integrated several strategies: “We used to have just any old sort of LPN—whoever was available to grab the patient—and now we have dedicated primary care LPNs on our team. … And they make calls to patients the day before and do as many reminders [as they can]—remind them of their appointment, remind them to take their blood pressure medicine—the day before. … We're starting to try telephone clinics. We've been using secure messaging” (Primary Care Provider 2). However, primary care providers also reported that their efforts to use encounterlesss encounters were at times limited by “mixed messages” from leadership. For example, primary care providers who created open access by utilizing encounterless encounters found themselves scheduled for face-to-face visits with patients of other providers whose schedules were too full to accommodate more visits. Providers described this as a lack of “protected” schedules and as a disconnect between organizational encouragement of non-face-to-face encounters and penalties for having unused appointments. For example: “Some of our performance pay money depends on the percentage of open slots we have available… Part of the issue there is I think… not having the administrative goals aligned with the provider or clinic goals” (Primary Care Provider 4). “I think the biggest change is that I have the ability now to get some people off the schedule that didn't need to be there… Again the problem is—because I'm able to do that and some of the others
248
S.L. Solimeo et al. / Healthcare 2 (2014) 245–250
aren't—my schedule gets filled up with other providers' patients which ruins the continuity of care. What I hear [from administrative leadership] is if there isn't a room in anybody else's [schedule] of course we have to see the patient” (Primary Care Provider 1). The conflicting messages from leadership that primary care providers perceived were also evident in the second major theme of primary care provider perspectives: challenges of using performance metrics to inform practice redesign.
“At [our facility], we're relatively focused on, “Well, Doctor A sees twelve patients a day and Doctor B only sees six patients. Well, Doctor B's got to work harder and see twelve patients a day.” Well, maybe not. Maybe Doctor B is actually doing the right thing and Doctor A needs to lighten up and not bring them back every time… Maybe he needs to utilize his RNs, LPNs or other support people a little bit more” (Primary Care Provider 8).
4. Discussion 3.3. Performance metrics must be sensitive to practice PACT training emphasizes the importance of proactive patient panel management to practice redesign and improving patient outcomes. Despite training, primary care providers reported instrumental challenges (e.g., locating specific metrics, administrative process involved in accessing data). Even those who recounted being able to readily access performance metrics identified other potential problems that limited use for practice redesign. For example, primary care providers noted that the time lag between practice redesign activities and data availability made it difficult to assess the impact of process changes. Other primary care providers remarked that some metrics were only available at a facility level, which prevented providers from isolating the effect of their efforts from those of other clinic providers. As this primary care provider remarked, “I mean the original information was facility level and with [many] providers, I have a very small input. I just throw [that metric] out because it doesn't reflect at all what we're doing” (Primary Care Provider 4). Primary care providers also believed that some metrics did not accurately reflect workload, whereas others were not actionable: “I think it would be nice to know exactly what data they're pulling on our teams and what are they looking at specifically…I would hate for us to be doing all this work, and then for them not to capture our work. That's one kind of fear that I have. … I guess that's just probably one of my bigger concerns because we are trying hard and we're trying to work as our PACT. I just don't want that to go unnoticed for something stupid like we didn't set [a clinic] up correctly” (Primary Care Provider 5). “I have influence over whether patients are seen in my clinic only during business hours and during regular days of the week—when our clinic is open. So if you include patients that come in [during] off-business hours or on weekends or holidays, that's part of the overall [emergency room] utilization rate, but I have no control over that, you know? I have no way to see patients in the evenings. I have no way to see patients on weekends. If they come in at that time, it's their choice and I don't have any control over that” (Primary Care Provider 4). 3.4. Change and sustainment over time Comparison of primary care provider responses between the initial interview and the follow-up interviews conducted after primary care providers had additional experience practicing in the PACT model revealed little difference in their attitudes. At 18 months post-implementation providers reported the same “mixed messages” and performance metrics barriers to PACT implementation as they did at the 6 month baseline interviews. However, there was an overarching, but subtle shift from referring to PACT as a new project to references such as the following which signaled adoption of PACT:
The current analysis was conducted to identify primary care providers' perceived non-technological barriers to PACT implementation. Medical home implementation is a comprehensive process which compels multi-level organizational change. Although participants discussed incorporation of technologies such as administrative data, scheduling systems, and encounterless encounters, the barriers to utilizing such tools were primarily organizational and not technological in nature. Qualitative research with VHA primary care providers implementing the PACT model in a large Midwestern region demonstrates the importance of multi-level alignment of implementation ideals with organizational practices.14 The primary care providers we interviewed were generally enthusiastic about the potential of the PACT model to improve care, and a majority readily adopted PACT strategies such as encounterless encounters. However, providers who were more successful at creating open access often experienced negative reinforcement from having to see patients of other primary care providers and from performance metrics that rewarded providers for full appointment calendars regardless of whether those appointments were in line with PACT ideals. In a related vein, primary care providers' efforts to use performance metrics in redesigning practice were limited not by the technological challenges that private sector practices contend with, but by lack of computer savvy required to find the data or by higher level data quality issues, such as concerns about data specificity and sensitivity. Another challenge was the belief that performance metrics were not equally actionable (e.g. trying to reduce emergency department use while not being able to expand clinic hours). These reported challenges were experienced in a context of both perceived greater provider accountability for patient outcomes and a bottom-up implementation approach, which likely contributed to primary care providers' report of need for enhanced and clarified communication from leadership as to PACT goals and expectations. Recent publications reporting on VHA PACT implementation emphasize the importance of supportive leadership on PACT team function,1,12,14 particularly with regard to sustaining same day access gains3 and improving panel management. As Murray and Tantau reported in 2000, primary care providers' efforts to provide continuity and same-day access must be organizationally valued and protected.17 Relationships between the time pressures experienced by providers and delegation are also important. A time allocation study conducted in VHA prior to PACT implementation determined that primary care providers averaged 8.6 h per week on patient care work outside of office visits.18 In a context of limited time, providers struggle to manage the additional workload19 and to prioritize direct patient care work over panel management activities.20 Underutilized intra-team delegation is a related component of time limitations,21 but a sense of obligation may reduce primary are providers' willingness and ability to share work with teammates.22 As Saba and colleagues have noted, “In our profession's efforts to create [high functioning health care teams] we have yet to articulate a myth that takes us from the lone physician to the team-based physician.”23 The lack of a “myth” to
S.L. Solimeo et al. / Healthcare 2 (2014) 245–250
foster PACT function may be in part due to a lag between the speed of individuals to adopt change and the speed of the larger organization to align leadership goals at multiple levels. While the PACT model compels multi-level organizational change, implementation emphasized bottom-up innovation. Administrative leadership at all levels should be integrated into medical home implementation efforts. The current study recruited PACT primary care providers involved in the first wave of implementation, facilitated by an intensive learning collaborative modeled after the Institute for Healthcare Improvement framework,12,24 making the data potentially biased towards the experiences of better trained providers. Thus, challenges reported among our sample may be more severe among providers who did not receive this focused training. We also were not able to link participants' perceptions to their actual performance metrics, which may have yielded a more balanced understanding of the challenges they reported.
249
Acknowledgments The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the United States government. This work could not have been done without the support of the VISN 23 Patient Aligned Care Team Demonstration Laboratory, supported by the Office of Patient Care Services, Department of Veterans Affairs. Additional support comes from the Center for Comprehensive Access & Delivery Research and Evaluation (CADRE) (Grant no. CIN 13-412), Department of Veterans Affairs, Iowa City VA Health Care System, Iowa City, IA. We wish to acknowledge technical assistance from the VA Midwest Health Care Network, Primary Care and Specialty Medicine Service Line and the VISN 23 PACT Collaborative teams.
References 5. Conclusions As an integrated medical system with a sophisticated electronic medical record, clinical reminder infrastructure, and pharmacy, VHA has tremendous potential to leverage the cost and quality improvements promised by the medical home model. Yet these apparent advantages which may lead casual observers to regard the PACT implementation as unique should be weighed together with the scope of the VHA project: PACT is being implemented in all VHA primary care settings among an extremely diverse primary care workforce rather than in highly selected clinics. What we learn from PACT has broad applicability to private sector settings and reveals the true challenges that underlie the surface dressing of technological infrastructure. Regardless of VHA's infrastructural and fiscal advantages over traditional fee for service health care organizations, providers implementing PACT may be challenged by an initial misalignment between PACT ideals and certain larger organizational policies and performance measures. That providers in this study reported the same barriers to PACT implementation at 6 and 18 months after launch is indicative of the profound role of organizational culture in everyday practice. Misalignment and other challenges may be surmounted over time as the organization more tightly aligns metrics of PACT success with the overarching goals of the PACT model and with the realities of clinical practice. The current study indicates the importance of such efforts in supporting primary care provider who are making the difficult transition to medical home adoption.
Support This work received financial support from the VISN 23 Patient Aligned Care Team Demonstration Lab, Iowa City VA Health Care System, Iowa City, IA, which is funded by the VA Office of Patient Care Services. Additional support comes from the Center for Comprehensive Access & Delivery Research and Evaluation (CADRE), Department of Veterans Affairs, Iowa City VA Health Care System, Iowa City, IA.
Conflicts of interest Samantha L. Solimeo, PhD, MPH: no conflicts to report. Kenda R. Stewart, PhD: no conflicts to report. Gregory L. Stewart, PhD: no conflicts to report. Gary Rosenthal, MD: no conflicts to report.
1. Reid R, Wagner E. The Veterans Health Administration Patient Aligned Care Teams: lessons in primary care transformation. J Gen Intern Med. 2014;29 (2):552–554, http://dx.doi.org/10.1007/s11606-014-2827-8. 2. Rosland A, Nelson K, Sun H, et al. The patient-centered medical home in the Veterans Health Administration. Am J Manag Care. 2013;19(7):e262–e272. 3. True G, Butler AE, Lamparska BG, et al. Open access in the patient-centered medical home: lessons from the Veterans Health Administration. J Gen Intern Med. 2013;28(4):539–545, http://dx.doi.org/10.1007/s11606-012-2279-y. 4. Jackson GL, Powers BJ, Chatterjee R, et al. Patient-centered medical home: a systematic review. Ann Intern Med. 2013;158(3):169–178, http://dx.doi.org/ 10.7326/0003-4819-158-3-201302050-00579. 5. Landon BE. Moving ahead with the PCMH: some progress, but more testing needed. J Gen Intern Med. 2013;28(6):753–755, http://dx.doi.org/10.1007/ s11606-013-2434-0. 6. Leasure EL, Jones RR, Meade LB, et al. There Is no I in teamwork in the patient-centered medical home: defining teamwork competencies for academic practice. Acad Med. 2013;88(5):585–592, http://dx.doi.org/10.1097/ACM. 0b013e31828b0289. 7. Schectman G, Stark R. Orchestrating large organizational change in primary care: the Veterans' Health Administration experience implementing a patientcentered medical home. J Gen Intern Med. 2014;29(2):550–551, http://dx.doi. org/10.1007/s11606-014-2828-7. 8. Meyer H. Group health's move to the medical home: for doctors, it's often a hard journey. Health Aff. 2010;29(5):844–851, http://dx.doi.org/10.1377/ hlthaff.2010.0345. 9. Parker LE, Kirchner JE, Bonner LM, et al. Creating a quality-improvement dialogue: utilizing knowledge from frontline staff, managers, and experts to foster health care quality improvement. Qual Health Res. 2009;19(2):229–242, http://dx.doi.org/10.1177/1049732308329481. 10. Fortney J, Burgess J, Bosworth H, Booth B, Kaboli P. A re-conceptualization of access for 21st century healthcare. J Gen Intern Med.. 2011;26(0):639–647 http://dx.doi.org/10.1007/s11606-011-1806-6. 11. Casalino LP. A martian's prescription for primary care: overhaul the physician's workday. Health Aff. 2010;29(5):785–790, http://dx.doi.org/10.1377/hlthaff. 2010.0133. 12. Solimeo SL, Hein M, Paez M, Ono S, Lampman M, Stewart GL. Medical homes require more than an EMR and aligned incentives. Am J Manag Care. 2013;19 (2):132–140. 13. Charles-Jones H, Latimer J, May C. Transforming general practice: the redistribution of medical work in primary care. Sociol Health Illn. 2003;25(1):71–92. 14. True G, Stewart G, Lampman M, Pelak M, Solimeo SL. Teamwork and delegation in medical homes: primary care staff perspectives in the Veterans Health Administration. J Gen Intern Med. 2014;29(2):632–639, http://dx.doi.org/ 10.1007/s11606-013-2666-z. 15. Britten N, Jones R, Murphy E, Stacy R. Qualitative research methods in general practice and primary care. Fam Pract. 1995;12(1):104–114, http://dx.doi.org/ 10.1093/fampra/12.1.104. 16. MAXQDA 10. Software for Qualitative Data Analysis. ([Computer Program]). Berlin, Germany: VERBI Software-Consult-Sozialforschung GmbH; 1998–2012. 17. Murray M, Tantau C. Same-day appointments: exploding the access paradigm. Fam Pract Manag. 2000;7(8):45–50. 18. Doerr E, Galpin K, Jones-Taylor C, et al. Between-visit workload in primary care. J Gen Intern Med. 2010;25(12):1289–1292, http://dx.doi.org/10.1007/s11606010-1470-2. 19.. Neuwirth EEB, Schmittdiel JA, Tallman K, Bellows J. Understanding panel management: a comparative study of an emerging approach to population care. Perm J. 2007;11(3):12–20. 20. Østbye T, Yarnall KS, Krause KM, Pollak KI, Gradison M, Michener JL. Is there time for management of patients with chronic diseases in primary care? Ann Fam Med. 2005;3:209–214.
250
S.L. Solimeo et al. / Healthcare 2 (2014) 245–250
21. Taylor EF, Machta RM, Meyers DS, Genevro J, Peikes DN. Enhancing the primary care team to provide redesigned care: the roles of practice facilitators and care managers. Ann Fam Med. 2013;11:80–83, http://dx.doi.org/10.1370/afm.1462. 22. Dingwall R. Problems of teamwork in primary care. In: Lonsdale S, Webb A, Briggs T, editors. Teamwork in the Personal Social Services and Health Care. London: Croom Helm; 1980. p. 111–137.
23. Saba GW, Villela TJ, Chen E, Hmmer H, Bodenheimer T. The myth of the lone physician: toward a collaborative alternative. Ann Fam Med. 2012;10:169–173, http://dx.doi.org/10.1370/afm.1353. 24. Institute for Healthcare Improvement. The Breakthrough Series: IHI's Collaborative Model for Achieving Breakthrough Improvement. Boston, MA; 2003.
