Physicians’ Voices: What Skills and Supports Are Needed for Effective Practice in an Integrated Delivery System? A Case Study of Kaiser Permanente

Abstract

Payers are demanding that US health care become more accountable and integrated, posing new demands for physicians and the organizations that partner with them. We conducted focus groups with 30 physicians in a large integrated delivery system who had previous experience practicing in less integrated settings and asked about skills they need to succeed in this environment. Physicians identified 3 primary skills: orienting to teams and systems, engaging patients as individuals and as a panel, and integrating cost awareness into practice. Physicians also expressed a high level of trust that the system was designed to help them provide better care. This belief appeared to make the new demands and mental shifts tolerable, even welcome, standing in contrast to research showing widespread physician distrust of their institutional settings. Physicians’ new skills and the system features that promote trust are described in the article and should be a focus for systems transitioning to a more integrated, accountable model.

Introduction

Public and private payers are demanding that health care organizations in the United States become more accountable for quality and total costs of care. Many stakeholders have set goals around the proportion of health care payments that should be “value-based” in the near future.^1-3 Nearly all types of value-based payment programs—including capitation, bundling, shared savings, pay-for-performance, and so on—encourage greater collaboration among providers than generally exists today. At the same time, physicians are increasingly choosing employment in larger, more organized settings, as opposed to solo or small-group practices. This is due in part to hospitals purchasing physician practices, ⁴ and in part to physicians choosing the stability and work-life balance that larger employers can provide.^5-7

If these trends continue, health care organizations will increasingly resemble integrated delivery systems, defined as “network[s] of organizations that provide . . . a coordinated continuum of services to a defined population and [that are] willing to be held clinically and fiscally accountable for the outcomes and the health status of the population served” ⁸ (see also Suter et al ⁹ for a more detailed definition of the properties of integrated care and Valentijn et al ¹⁰ for a proposed taxonomy of levels of integration). Such a shift will likely pose new demands for physicians and for organizations that educate, certify, employ, or partner with them. ¹¹

We can learn from existing integrated delivery systems—such as Kaiser Permanente, the Mayo Clinic Health System, and Geisinger Health System—about how physicians adapt to working in these settings. In this study, we asked physicians about their transitions from a less integrated to a more integrated setting. Physicians’ accounts of such transitions may reveal specific skills or attitudes they need to be effective and satisfied in the latter environment. This is potentially important for several reasons. First, studies of change management in health care often identify physician buy-in as a key factor in determining whether organizational initiatives succeed.^12-14 Second, many studies have described physicians’ dysfunctional relationships with their care settings, including accounts of systemic distrust between physicians, administrators, and other providers^15,16; physician burnout ¹⁷ ; physicians’ perceptions that they must be hypervigilant to protect patients against unsafe hospital conditions^18,19; and the “hidden curriculum” embedded in physician education that teaches them to distrust their care systems.^20,21

Currently, we have limited understanding of how physicians fare when they transition into more integrated delivery settings. Studies have focused on physicians’ attitudes about specific aspects of practice, such as care management processes, compensation, workload, use of electronic medical records, or performance measurement.^22-26 Studies of differences across practice settings tend to focus on comparing employed versus independent physicians^27,28 or looking broadly at physicians’ experiences working in managed care.^29,30

Experts have opined about how to support the movement toward integrated care,^31,32 but the field lacks a nuanced view of the day-to-day challenges and opportunities physicians face as their practice environments change. Leaders of integrated delivery systems commonly observe that the culture of these systems is simply different from other environments, with some referring to a culture of “group accountability” ³³ or of “continuous improvement.” ³⁴ To our knowledge, however, no study has identified front-line physicians who have worked in integrated and non-integrated settings and asked them to articulate what those differences are.

Study Data and Methods

We conducted a qualitative descriptive study of how physicians experience the transition to working in an integrated, accountable system. ³⁵ We held focus groups in Fall 2015 with 30 physicians who came to work at one particular integrated delivery system—Kaiser Permanente—after having trained and practiced in other settings. Kaiser Permanente, headquartered in Oakland, California, is the nation’s largest not-for-profit, integrated delivery system, with 11.3 million patients and over 21,500 physicians in 8 states and the District of Columbia. ³⁶ Kaiser Permanente is a mutually exclusive partnership and contractual alliance between a not-for-profit insurer, Kaiser Foundation Health Plan; a not-for-profit hospital system, Kaiser Foundation Hospitals; and 8 physician professional organizations, the Permanente Medical Groups. ³⁷ In each of Kaiser Permanente’s geographic regions, the Health Plan offers coverage, and the associated Medical Group provides professional services for a negotiated per-member fee. Individual physicians are paid by salary with modest performance incentives. ³⁷ The focus of our study is the Southern California Permanente Medical Group (SCPMG). Together with their health plan and hospital partners, SCPMG’s nearly 7300 physicians serve over 4.2 million patients across Southern California.

An interdisciplinary research team with expertise in cultural anthropology and health policy conducted a total of four 90-minute focus groups, each at a different site within the Kaiser Permanente Southern California region. Two groups had 6 physician participants each; 2 had 9 participants each. Participating physicians had 3 to 13 years of experience with SCPMG and came to the group after having previously practiced outside of Kaiser Permanente for at least 1 year. Most came from non-integrated care settings.

The sessions focused on the differences physicians observed between SCPMG and their previous workplaces and were designed to elicit answers to 2 primary questions: What new skills, behaviors, and attitudes (if any) did physicians have to develop in this new context, and how should training and certification organizations prepare and support physicians to thrive in this type of setting? (See the appendix for the full interview guide.) All 4 groups also addressed spontaneously 2 additional questions, without prompts from the researchers: What organizational features of the system itself enabled physicians to practice successfully in this environment, and what were the tradeoffs inherent in those organizational features?

We chose 3 years as the minimum tenure because that is the point at which physicians become full partners in the practice, indicating (to us) that participants had made an affirmative decision to remain at SCPMG, at least for the near future. Across the groups, participants appeared diverse in terms of age, race/ethnicity, and gender (11 women, 19 men). There were 16 primary care physicians and 14 specialists. Nineteen participants had spent between 3 and 6 years at SCPMG, and 11 had spent between 7 and 13 years there.

Using human resource records, we identified all physicians associated with each site who met the study criteria and emailed them an invitation to participate. We also placed a recruitment announcement in a weekly SCPMG electronic newsletter. We offered participants an honorarium of $250.

The same 2 members of our research team (BC and LT), neither of whom is a Kaiser Permanente employee, moderated all 4 focus groups. The moderators asked all questions in the interview guide of all participants, in every group, following the natural flow of conversation to ask questions in combination or in different sequence, where appropriate. We took detailed notes during the sessions and transcribed audio recordings.

A larger team of 4 (BC, LT, JL, SD) analyzed all notes and transcripts using the constant comparative method, ³⁵ beginning with a set of a priori codes based on the interview questions and constructs, and then identifying new codes and themes in an inductive manner as they emerged through reading and comparison. We used Excel to organize and code all data. Our conceptual framework for analyzing the data was one of iterative, reflexive sifting for meaning making. ³⁸ We searched the data both for the answers to our pre-existing questions and for the new questions and ideas present in our participants’ narratives, with a particular focus on how our participants themselves interpreted the experiences they described. Finally, we compared team members’ individual analyses with each other and with the original interview data, in an iterative manner, to refine and verify codes and themes.

All focus group participants gave written informed consent. The SCPMG Institutional Review Board approved the methods for the study (approval no 10796).

Results

We divide the results into 4 categories. The first 2 are related to questions we asked the participants explicitly: What new skills, behaviors, and attitudes did physicians need to develop in the new context, and how should training and certification organizations prepare and support them to thrive in this environment? The second 2 categories of results are related to issues that all 4 focus groups brought out spontaneously: What organizational features enable physicians to practice successfully in this environment, and what are some tradeoffs inherent in those organizational features?

New Physician Knowledge, Skills, and Abilities

Physicians identified 3 primary areas of knowledge, skills, and abilities they needed to acquire when they transitioned to this medical group: orienting to the team and system, engaging patients as individuals and as a panel (ie, as a group under the care of a single clinician or team), and integrating cost awareness into practice.

Orienting to team and system

Nearly all physicians described the need to be oriented to the larger team and the system in ways that were different from their experiences in other work settings. For example, physicians said they needed to work with support staff over whom they could not assert direct authority. Several were initially dismayed about this. As one physician said, “On day one . . . I went to ask a medical assistant something and they said to me, ‘Oh, ask so-and-so because I’m on lunch.’ And the concept of . . . ‘lunch’ and ‘breaks’ was astounding to me” (Group 1, Participant 5).

Physicians also described the need to accept input from administrators—input that had been absent or easy to dismiss in other settings. One said, “Before, I didn’t have a boss. If [hospital administrators] tried to be too controlling, we’d take our business elsewhere. Here, the administration sets goals. They’re not outrageous goals though” (Group 4, Participant 2).

Finally, physicians talked about needing to focus on organizational goals and practice guidelines, even if their value was not immediately apparent. As one said (and many others seconded),

There’s a thing known as “KP medicine,” which is value-driven. It’s got some constraints. [You ask,] “Who are they to tell me how to practice?” But over time, you get used to it and buy into it. . . . At first, you don’t understand the limits, because they’re different from what you [experienced] with outside insurance companies. You learn the evidence base behind it, and it’s cohesive. (Group 2, Participant 4)

Physicians also said they needed to accept limits on their autonomy, acknowledging they were part of a larger organization that holds collective responsibility for patients. However, they did not characterize this as an abrogation of individual “ownership” of their panel; rather, they needed to recognize that they were not the only ones responsible for their patients. One said,

This isn’t really your patient. This is Kaiser’s patient. And we’ve got to make sure that we’re doing everything efficiently and correctly. . . . Whereas, you know, in the private world [it’s], “I’m the doctor. I admitted the patient. I’m giving the hospital business. And I’m taking care of the patient the best I know how.” (Group 1, Participant 6)

They observed that this might not be equally easy for all physicians. As one said, “If the [physician] . . . has to be in charge of everything, and the only way things happen is their way, they don’t belong in a system like this” (Group 2, Participant 6). Another simply said, “If you want to be a ‘cowboy,’ this is not the place to be” (Group 2, Participant 2).

Engaging patients as individuals and as a panel

Physicians described several ways in which they had learned to work differently with patients. First, they paid closer attention to their patients’ experiences of care. Many health systems routinely survey patients; what was notable in this medical group was physicians’ perception that they would be held accountable for their patients’ experiences.

Second, they had to view their relationship with patients in the context of organizational goals, such as preventive care targets for an entire patient panel. One primary care physician recounted,

When I first started working here, I was amazed at the proactive nature of the “care gaps.” I would see a patient with a 103 [degree] fever, but I would still do a diabetic foot exam because it was due. Sometimes it’s uncomfortable, but sometimes it seems appropriate. (Group 2, Participant 1)

On the same topic, several talked about new panel-management skills and tasks they had to master. The same physician quoted above said, “You have to manage a panel. It means phone calls to them. It means letters. It means . . . encouraging them to go for all their routine health maintenance things” (Group 2, Participant 1).

Integrating cost awareness into practice

Many discussed needing to understand the costs associated with the care their patients receive. They said costs had been ignored in medical school and residency, and if they had learned about this issue in previous work settings, it was with an emphasis on maximizing volume and billing, rather than value. In this system, they said, they needed to be aware of their patients’ costs for treatments they prescribed, so that they could know when, for example, a patient would need to spend a significant amount over what insurance would cover.

Physicians also said they needed to understand the cost of care to the larger organization, so that they could act responsibly as stewards of the system’s resources. As one said, “When you work in a system like this, all of a sudden [cost] becomes very important. Looking at it more from the system’s point of view . . . it’s also important to you to know that we’re not paying for procedures that are not necessary or being duplicated or whatever” (Group 1, Participant 1).

Changes to Medical Education

We asked what medical schools and residency programs should do differently (if anything) to prepare physicians for integrated care. Participants’ answers reemphasized the above findings—system orientation, patient engagement, panel management, and especially what they glossed as “business fundamentals,” including integrating cost consciousness into practice. One observed, “What medical schools need to incorporate is more of a systems approach. . . . How [do] you work as one part of a team where, as the physician, you may not be the captain of the ship?” (Group 1, Participant 3). Regarding panel management, another commented, “Traditional [physician] education is based on treating patient[s’] symptoms, and they get better or they don’t. [Here], you’re responsible for a panel of patients, whether they come to see you or not. You are still responsible for them” (Group 2, Participant 1).

Physicians identified 2 reasons why these programs should focus on “business fundamentals.” First, knowledge about costs can help physicians understand and support an organization’s business decisions. One said, “Understanding the business aspects gives me more understanding and ownership of [the organization’s] policies and financial decisions” (Group 3, Participant 8). Second, learning about the business of medicine could make trainees aware of the different types of practice settings that exist in the United States, allowing them to make an informed choice about where to work. Participants felt they could have benefited from being aware of the possibility of working in an integrated system earlier in their career.

System Features That Support Acquisition of New Skills

As noted, many participants offered unsolicited descriptions of system features that engendered their trust and made them willing (possibly even eager) to acquire the new skills. These included structures supporting collaboration and learning among colleagues, structures supporting evidence-based medicine and panel management, and compensation based on salary and capitation instead of volume and fee-for-service.

Collaboration and learning

Physicians described working in an information-rich environment relative to their previous work settings. The health system’s well-developed electronic health record (EHR) allowed them to access and share information about their patients’ histories and interactions with other providers. One said, “I just love how everything’s all in one system. . . . I’m not having to make phone calls to talk to [a patient’s] outside primary care doc. I just look it up” (Group 4, Participant 4).

Physicians enumerated other mechanisms for collaboration. Rather than being incentivized to refer patients to specialists, physicians said they were encouraged to consult with specialists themselves. The system offered several tools for this, including a mechanism embedded in the EHR that allows doctors to query specialists in near real-time, tools for sharing images with dermatologists via the EHR, and a specialized hotline for primary care physicians to speak to an on-call orthopedist. Physicians said they valued the ready availability of specialist advice and noted the contrast with what they had experienced elsewhere. One primary care physician said, “I . . . got [a response from a specialist] within ten minutes one time. By the time I was finished charting, it was there” (Group 4, Participant 3).

Physicians also described receiving meaningful clinical feedback from other physicians and administrators. Surprisingly to the research team, they said this feedback felt appropriate and helpful, not intrusive, as illustrated by this exchange:

Physician:

When you go to put in the referral [to a specialist], a screen will pop up. . . . It’ll actually give you some information of common treatments or common labs to order, or scans, or what the patient might need.

Discussion leader:

And do you find that helpful?

Physician:

Oh, yes!. . . . And then, you know, you’ll get a note sometimes back in the chart [from the specialist, saying], “Hey, saw this patient. . . . This is what we did. . . .” So you can reflect on, hey, was there something I could’ve done a little bit more? . . . And sometimes you might even get a staff message . . . from the specialist who’s reviewing your referral [saying] “Hey, these are some things you can do for the patient right now.” . . . [In other settings], all you’ll hear is “Thank you for the interesting consult.” (Group 4, Participant 6)

Supports for evidence-based medicine and panel management

Physicians described many ways that the system supported them in managing patient panels and practicing evidence-based medicine (although most did not use the latter term), including a department that summarizes evidence and develops recommendations for new medications and technologies. One said,

Here we have a whole department that just analyzes [new laboratory tests and technologies . . . They] review all the literature, and they have epidemiologists, and PhDs, and MDs. . . . It’s a great resource. (Group 1, Participant 3)

Physicians also said they appreciated the numerous goal-tracking and panel-management tools in the EHR to help them know whether they are meeting care goals for individual patients and for their panels.

Salary and capitation

Physicians described benefits of working in a system of capitated payments and salaried physicians: primarily, that the system’s financial incentives were aligned with providing effective care, not maximizing volume of services. One said, “The common goal is ‘take good care of the patient.’ Financial motivation is out of the picture” (Group 4, Participant 7). Another said, “Previously . . . I was encouraged to see as many [patients] or code as highly as I possibly could, to make money for the group. [H]ere, we’re actually encouraged to give quality care” (Group 4, Participant 4). They said this enhanced their sense of providing the best care possible to their patients.

Tradeoffs

At the same time, physicians described tradeoffs inherent in all these system features. For example, the focus on prevention and population-based care largely aligned with physicians’ own goals, but it could also sometimes feel intrusive in its requirements to “push” certain care. One noted,

There’s no way to put [on a patient’s chart] “forget about the mammogram and leave her alone about the pap smear.” Even if you were to write it in: “doesn’t want [a mammogram],” it doesn’t matter because they’re still going to be caught in the computer data mining that [says], “this person hasn’t had a mammogram.” (Group 2, Participant 1)

The large size of the organization and some of its bureaucratic features, including reliance on unionized workforces, provided a sense of stability and freed physicians from hassles such as hiring and payroll. However, they acknowledged that change can feel slow in such a large organization. One said, “The responsiveness to change is significantly slower here. . . . There are multiple layers and multiple levels that have to check off on change here. For better or for worse, [in] smaller groups, you would get a new idea and you’d just do it” (Group 3, Participant 8).

Regarding financial incentives, participants agreed that physician salary and group capitation greatly encouraged a focus on patients’ needs and the “right” amount of care. However, several noted that salary might lead some colleagues to be slow to respond to referrals.

Participants largely regarded these tradeoffs as “worth it,” with the downsides often described with humor, as something to tolerate as part of a bigger, more positive picture. For the research team, the fact that physicians acknowledged these inherent tradeoffs reinforced the impression that they trust their system and believe its interests align with their own.

Discussion

Systems That Earn Physicians’ Trust

Because our primary focus was on physicians’ knowledge, skills, and abilities, we expected to hear about issues such as learning new approaches to collaboration and panel management, or learning to accommodate system-set rules or guidance. We did not expect to gather data on how the system had earned physicians’ trust, making them amenable to developing these new competencies. This trust in the system contrasts with what we have seen in the literature and in our previous work in non-integrated settings, where physicians often view the “system” as something they must work around to provide good patient care and where attempts to influence what physicians do are rare and unwelcome when they occur. ¹⁸ Our participants described ways the system supported them to provide a different, better kind of care. This was not something most had consciously sought when coming to SCPMG. Rather, most said they had come to this group for reasons other than care quality—primarily, for work-life balance and financial stability. But having sought personal balance and stability, they discovered a system that resonated with their own values and priorities for patient care.

For some, this experience effected a change in those values and priorities. The new work setting demanded different skills and attitudes than had been required elsewhere, and putting those into practice in the context of an integrated system redefined their perspective on what being a physician entailed. They were able to accomplish things with their teams and for their patients that would not have been possible, or even conceivable, in other settings. A good example of this is the dynamic one physician described in which he appreciated receiving feedback from a specialist about his referral decision because it would help him make the most appropriate referrals in the future.

Participants’ new perspectives on the nature of the physician’s job may be particularly relevant to nascent Accountable Care Organizations (ACOs), and they appear to lend credence to common strategies ACOs have used to affect physician behavior. In particular, Phipps-Taylor and Shortell recently found that the most common change motivators ACOs used with their physicians were opportunities to have a greater impact on patients and to be a more effective physician. ³⁹

The experience of transitioning to work in this integrated system carried with it a change in what anthropological theory terms worldview and ethos—in how physicians envision the world and what good conduct in that world should entail. ⁴⁰ This cultural reframing seems to arise from the gestalt of the system features described above, which align closely with the criteria for integrated care proposed by Shortell and Gillies, ⁸ Suter et al, ⁹ or Valentijn et al. ¹⁰ The system features mentioned by group participants function not just to support care providers but to enable this new cultural framework as well.

This is underscored by our participants’ references to the importance of physicians’ acceptance of teamwork and limits on traditionally defined professional autonomy. Seen from the perspective taught implicitly and explicitly to physicians in training,^20,21 limits on autonomy would normally be framed in terms of loss—privileges and affordances made unavailable by the constraints of this integrated care system. But unfettered autonomy can also be understood as an aspect of physician practice and culture best left behind, a symptom of deficient systems—a bug, not a feature.

That said, in analyzing data such as these, we must avoid idealizing integrated systems. As discussed below (see the Remaining Questions section), surely there are some physicians who have chosen to exit this system or others like it; understanding their reasons for doing so could prove illuminating. In addition, research on physicians in integrated systems has described their need to negotiate unhelpful information technology and bureaucracy, just like physicians in other settings. ⁴¹ Still, our participants’ accounts of what they had to accustom themselves to in this new setting paint a telling portrait that differs sharply from the default nature of health care practices they had experienced previously: reactive, information poor, uncoordinated, and when it comes to cost and efficiency, haphazard at best and driven by counterproductive incentives at worst.

Implications for Training and Standard Setting

Our study has implications for medical schools and for funders of Graduate Medical Education. Consistent with previous research, ⁴² physicians said their education had ignored topics such as system orientation, patient engagement, panel management, and business fundamentals. We do not suggest that these should be layered on top of existing curricula—this would merely add to the already-great volume of information that medical trainees must absorb. Rather, to the extent possible, physician training should incorporate these concepts organically, so that they are seen as fundamental to delivering care, as in the movement to include cost awareness in resident education.^43,44

Our findings also have implications for organizations that set standards for physicians and support their ongoing professional development. Medical specialty societies and other providers of continuing medical education should consider how they might emphasize several specific skills and attributes. The first is system-based practice, including the ability to work within limits and toward goals set by the larger organization. The second is inter-professional teamwork, encompassing collaboration with both face-to-face and virtual teams. The third is patient centeredness, meaning understanding and facilitating patients’ pathways through the care system, as well as understanding what care costs patients. Specialty societies and certification and licensing boards might wish to consider developing standards to assure that physicians have achieved and maintain competency in these areas.

Remaining Questions

We found an unusually high degree of trust among these physicians that their system is designed to help them practice good medicine. This trust seemed rooted in physicians’ sense that the organization’s goals aligned with their own goals for patient care. While we cannot comment on causation, we do know that this medical group’s leadership expends significant effort to ensure physicians understand and buy into organizational goals. One way they promote this is through a multi-day program for new hires that focuses on what the organization expects from physicians and what they can expect in return. Program leaders report they intend to lay the groundwork for physicians to develop the trust in the system our participants described.

Another feature of this medical group that may contribute to the level of trust we observed is its representative governance. Each partner may vote for a member of the Board of Directors from his or her geographic area. Board candidates are self-nominated and must win the support of their colleagues.

This group’s unique payment structures (group-level capitation and individual physician salary) may also have contributed to the trust we observed. As noted, several participants mentioned that the system’s financial incentives were aligned with their own preferences for doing what is best for the patient, rather than simply maximizing the volume of services. Some experts argue that ACOs and similar entities will inevitably fail if they do not take bolder steps toward capitation. ⁴⁵

The role played by this system’s unique features—including onboarding programs, representative governance, and capitation/salary—merits further study. It will also be important to explore whether physicians in other established integrated delivery systems express similar views, as well as whether the observed level of physician trust in this system is evident in other, similar systems.

Further research could also compare our findings with those from a group of SCPMG physicians with no previous experience outside of the system. Is this group’s apparent level of satisfaction related to our participants’ having worked in other settings that they chose to leave? Would we see similar results with physicians who started practicing in a highly integrated setting straight out of residency? Answers to these questions could have implications for the hiring of physicians by nascent integrated delivery systems. In addition, research with physicians who have left, or were asked to leave, Kaiser Permanente or integrated systems to work in another type of setting could illuminate other aspects of physicians’ experiences in these settings and what leads physicians to feel such a setting is “right” for them as a professional.

Finally, further research could explore the perspectives on similar questions held by non-physician providers, managers, administrators, and patients.

Limitations

This study includes only a limited number of physicians from a specific region of a single, and very unique, health system. Hypotheses about how our findings could apply more broadly are therefore speculative. We sought to mitigate this limitation by focusing on topics that are relevant for every integrated practice setting (eg, supports for evidence-based medicine), rather than specific to this particular system (eg, full vertical integration among physicians, hospitals and health plan, or having a 70-year history of organizational stability).

Focus groups have inherent limitations: We gathered qualitative data from a limited number of participants, covering topics they felt comfortable broaching in a semi-public setting. Our data consist of these particular physicians’ recollections and perceptions of their previous work settings and their adaptation to their current one; these may not be comprehensive and accurate. We sought to address these limitations by systematically comparing participants’ perceptions for consistency. Research with a broader, representative sample (eg, a survey) would be a useful supplement to our findings; however, such a survey was beyond the scope of the project. Furthermore, we sought a more in-depth interaction with participants than would have been possible with a survey and chose our methodology accordingly, knowingly giving up some amount of generalizability for a richer analysis of our particular participants’ experiences.

In addition, the study included only physicians who had worked or trained elsewhere, then chose to work at SCPMG, remained there for 3 years, became partners, and volunteered to participate in our study. Other physicians—for example, those who left SCPMG or were asked to leave before 3 years, or who came there straight from residency—may have different perspectives, particularly on the system features that are needed to support effective practice in an integrated system.

Finally, several members of the study team are Kaiser Permanente employees, potentially introducing bias. However, the interview questions were designed to elicit neutral perspectives, and the interviews were conducted and attended only by team members (including the lead researcher) who are not Kaiser Permanente employees—a point that was made clear to participants. The participant information sheet also outlined the study’s procedures for keeping information confidential from SCPMG leadership.

Conclusion

According to physicians with experience both in and outside of an integrated health care delivery system, practicing in an integrated system requires a distinct set of competencies and attitudes, including a team orientation, a willingness to embrace larger organizational goals, and the ability to participate in proactively managing a patient panel. In the case of SCPMG, these new competencies and attitudes were accompanied by feeling meaningfully supported by a trusted network of peers, teammates, and a robust information and resource infrastructure. Physicians described how this system had earned their trust and enabled them to practice a different, better kind of medicine than what had been possible elsewhere. This has implications for physicians who practice in integrated delivery systems, for organizations that train and certify physicians and other providers, and for delivery systems striving to provide more accountable, integrated care.