Implementing Systemwide Physician Change Management in an Integrated Health Care Setting: Improving Physician Participation in an Advanced Care at Home Model

Abstract

Introduction

Advanced Care at Home is a clinical model that delivers hospital-level care in a patient’s home. This model of care has been studied for decades, but there have been difficulties scaling the model to a higher census because of poor physician participation. Kaiser Permanente at Home, an Advanced Care at Home model created by Kaiser Permanente Northwest, was able to quickly increase its patient census by using several different change management interventions. The aim of this study was to describe the specific physician change management interventions used and to determine their relative impacts on physician participation with Kaiser Permanente at Home.

Methods

This study used a retrospective qualitative approach. Hospitalist and emergency department (ED) physicians completed an online survey in December 2021. This was followed by focused, one-on-one interviews that were held in February 2022. Content analysis was performed using a general inductive approach to identify core themes.

Results

Of 78 ED and 79 hospitalist physicians recruited, 35% submitted responses. Of these respondents, 16 (29%) were ED physicians, and 39 (61%) were hospitalist physicians. Of these respondents, 90% rated Kaiser Permanente at Home favorably over the course of a year. More than 90% of respondents rated a combination of multiple approaches as impactful, but respondents overwhelmingly noted that physician-to-physician engagement was the most important (51%).

Conclusion

In the development of the Kaiser Permanente at Home, physicians highlighted that a multifactorial change management approach centered on peer-to-peer engagement had the most substantial effect on their participation, a process that could extend up to a year.

Introduction

Advanced Care at Home (ACaH) is a clinical model that duplicates hospital-level care in a patient’s home. This model of care has been investigated for decades,^1–4 and multiple studies demonstrate improved quality and utilization metrics such as lower readmission rates, lower costs of care, and fewer services used.^5–8 Despite these advantages, ACaH has not been implemented at scale (meaning with a daily census of patients increased to a higher capacity than those studied in pilots). There are many barriers associated with the development and scaling of this model of care, including physician hesitancy to refer patients to an ACaH program.^9–12 Other studies have found that physicians are hesitant to agree to refer patients to this model of care for several reasons; for example, Chua et al noted several themes associated with physician hesitancy, such as a time-consuming referral process and difficulty identifying and screening patients.¹²

In April 2020, Kaiser Permanente Northwest, an integrated delivery system, in the Portland, Oregon, and Vancouver, Washington, metropolitan area initiated its version of ACaH care called Kaiser Permanente at Home. Kaiser Permanente at Home provides hospital-level care at home as a substitute for traditional acute hospital care. To date, Kaiser Permanente at Home has involved the largest number of patients studied in an ACaH program and is the first study to demonstrate that ACaH can be quickly launched and scaled.¹³

Change management theory has been demonstrated to be effective in implementing physician change,^14–18 and thus, many change management strategies were required to address physician hesitancy at referring patients for admission to Kaiser Permanente at Home.

The aim of this study was to describe the specific physician change management interventions used and to determine their relative impacts on physician participation with Kaiser Permanente at Home.

Methods

Setting and context

Kaiser Permanente is an integrated delivery care system. Kaiser Permanente Northwest is 1 of 8 regions within the Kaiser Permanente umbrella. Kaiser Permanente Northwest provides care for approximately 636,854 members in northwest Oregon and southwest Washington.

Kaiser Permanente Northwest patients are managed at 1 of 5 regional hospitals, 2 of which are core locations reserved solely for Kaiser-insured members that are staffed primarily by Northwest Permanente physicians. From initial implementation in 2020 until the end of 2022, Kaiser Permanente at Home admitted patients only from the 2 core hospitals and emergency rooms in the Portland metropolitan area.

Kaiser Permanente at Home and the clinical model

Kaiser Permanente at Home involved the largest number of patients studied in an ACaH program and was the first ACaH program that was able to be quickly launched and scaled. Since its inception in April 2020, the Kaiser Permanente at Home program has admitted more than 2700 patients (with hospital-level acuity). This program was able to achieve scale with a daily peak of 22 patients per day and a monthly average daily census high of 16, which represented 9% of the hospitalist census. Care was found to be safe, with high quality and a comparable patient care experience, and to be associated with lower 30-day readmission rates.¹³

The Kaiser Permanente at Home virtual hospital is a business collaboration between Kaiser Permanente and Medically Home Group, Inc, a technology-enabled health care services company that provides the platform and associated technology that enables acute hospital-level integrated care at home.

The Kaiser Permanente at Home staffing structure uses a virtual physician and nurse assigned to each patient. The physician and nurse teams leverage both virtual care technology and a supply chain of in-home vendors to care for their patients (Supplemental Appendix A).

Physician care is provided virtually by hospitalist physicians from the Northwest Permanente, PC, physician group. These physicians perform rounds at both of the core hospitals and for Kaiser Permanente at Home. Nursing and care coordination is provided by specialized virtual care nurses who use telehealth technology, supply chain coordination, and scheduling.

Admission to Kaiser Permanente at Home

The clinical path for admitting patients to Kaiser Permanente at Home is depicted in Supplemental Appendix B.

Patients are identified for admission by the Kaiser Permanente at Home virtual nurses, who screen the charts of patients who are in the emergency departments (EDs) of the 2 Portland-metro-area core hospitals or have already been admitted from the medical floor. The initial patient eligibility criteria for Kaiser Permanente at Home include that the patient requires hospital admission for acute medical illness and that the patient lives within the program’s geographic catchment area.

If a patient is deemed eligible, a Kaiser Permanente at Home physician reaches out to the attending physician (hospitalist or ED physician) caring for the patient to agree upon the clinical appropriateness for Kaiser Permanente at Home care. If the clinical screening is deemed appropriate, the attending physician asks the patient whether they would be willing to continue their hospital admission with Kaiser Permanente at Home. If the patient consents to care with Kaiser Permanente at Home, a Kaiser Permanente at Home nurse contacts the patient for a social appropriateness screening. Once the patient is considered appropriate from both clinical and social standpoints and gives consent for admission, they are transported home and admitted to the Kaiser Permanente at Home program.

Determining the clinical appropriateness of a patient for Kaiser Permanente at Home care and performing the administrative work required to transfer a patient from the hospital to home for Kaiser Permanente at Home care are considered difficult for most ED and hospitalist physicians to learn.

Physician change management strategy

It was important to develop an integrated and multifaceted change management strategy. This project used several different theories of change management, including the Theoretical Domains Framework¹⁹ and operant learning theory,^14–18 along with other behavior change research, to develop a list of interventions.

Interventions

The interventions used for physician change management and their associated theoretical domains are listed in the Table.

Table:

Interventions used and their corresponding theoretical domains

Intervention	Theoretical Domain
Peer-to-peer communication: Kaiser Permanente at Home physician communication with attending ED and hospital medicine physicians occurs in the moment to aid in education and efficiency of admission	Peer influence
Education: Frequent presentations are made at department meetings (both ED and hospital medicine physicians)	Academic detailing Educational outreach CME
Secure messaging: Texting apps are used to communicate in the moment with attending physicians	e-Technologya
EMR clarity: Easy access to patient information in the EMR is ensured for referring physicians	Audit and feedback
Several attempts were made internally to improve name and program recognition among physicians including posters, stickers, and pens with the “Think KP at Home” insignia	Name and program recognition

^ae-Technology here refers to texting and communication apps.

CME, continuing medical education; ED, emergency department; EMR, electronic medical record.

Academic detailing, educational outreach, and continuing medical education

To educate physicians about a model with which they had not interacted in the past, several presentations were integrated into the department meetings of several specialties involved with the typical care of a hospitalized patient, including the hospitalist, emergency, nephrology, neurology, gastroenterology, general surgery, and cardiology departments. The content of these meetings included proof of quality of the model, transparency regarding metrics of the program, and descriptions of the patient experience. These educational presentations continued at department meetings with the emergency and hospitalist departments monthly.

Peer influence and e-technology

As described earlier, the admission process relies heavily on work by the attending physician. To make this process easier and more efficient for the attending physician, the Kaiser Permanente at Home physician used e-technology to reach out to the attending physician in order to educate them on clinical appropriateness and aid in the coordination of Kaiser Permanente at Home admission. In addition, the Kaiser Permanente at Home physician assisted the hospital attending physician with discharge paperwork to lessen the administrative burden of a Kaiser Permanente at Home admission. This collaboration between the hospital attending physician and the Kaiser Permanente at Home physician not only supported the education of the hospitalist physician regarding Kaiser Permanente at Home as a model of care but also helped reduce the attending physician’s workload.

Audit and feedback

Feedback from referring physicians frequently noted that it was important to be able to easily access Kaiser Permanente at Home patient records. This allowed physicians to track their patients’ progress after a Kaiser Permanente at Home admission and also to cross-cover for after-hours patient issues.

Prior to the inception of the Kaiser Permanente at Home program, the electronic medical records (EMR) used by Kaiser Permanente Northwest did not have a build for this model of care. Therefore, when creating the EMR component of the Kaiser Permanente at Home program, care was taken to allow transparency of the records for physicians.

Study design

This study was a retrospective, mixed-methods qualitative cross-sectional survey of a defined population. It included both close- and open-ended questions with follow-up interviews. Physician participants were recruited from Kaiser Permanente Northwest to participate in an online survey regarding their perceived changes in attitude toward Kaiser Permanente at Home during different points of time after program initiation.

Online responses were collected from internal participants from December 2021 to January 2022. Every eligible full-time ED and hospitalist physician was sent an online recruitment email with a web link and instructions for how to complete the online study survey. Follow-up email reminders were sent 2 and 3 weeks later to physicians who had not yet completed the survey.

Responses were analyzed to assist in the construction of semistructured one-on-one interviews that were held in February 2022.

The Northwest Permanente Institutional Review Board (IRB) reviewed the details of this study and deemed it quality improvement and thus exempt from IRB review.

Survey procedures

The survey was administered using a Microsoft Forms survey platform. A link to the survey was distributed to Kaiser Permanente Northwest practitioners in relevant roles via email. The invitation and header of the survey described the purpose and intent to publish. The survey questions were designed to elicit the opinions of hospital physicians regarding the implementation of the Kaiser Permanente at Home program. The survey included quantitative questions and required approximately 5 to 10 minutes to complete.

Survey development and content

Survey questions were developed to create a face-valid survey that could be completed within a 5- to 10-minute timeframe. The final survey was a mixed-methods survey that included some forced-choice answers to questions, along with an opportunity to provide further comments in an open-text field. Questions included Likert scale questions to incorporate changes in attitudes over time. Respondents were then asked if they would consent to a one-on-one interview for further analysis. Respondents who consented were then contacted to set up a short 15-minute interview.

Qualitative study design and interviews

Data collection was performed through individual interviews with initial respondents who consented to a further semistructured individual interview. These interviews were performed in February 2022 during protected time with the respondent so that there were no time constraints. Each interview lasted about 15 to 20 minutes. In total, 6 ED physicians and 6 hospitalist physicians were interviewed. No respondent had worked in the past for Kaiser Permanente at Home. To enhance the consistency of data collection, the same semistructured interview protocol consisting of 3 open-ended questions was used to guide individual interviews. All interviews were completed by the same researcher who had been trained in qualitative data collection. All interviews were transcribed without any identifying participant details.

Data analysis of qualitative interviews

Thematic framework analysis was employed for data analysis. Each interview was analyzed in detail with close attention to themes and subthemes. The interpretation of quotes was facilitated by considering the frequency, context, and specificity of the data.

Results

Survey results

As depicted in Figure 1, of the 78 ED and 79 hospitalist physicians who were recruited to complete the survey, a total of 55 physicians submitted responses (35%). Of these respondents, 16 (29%) were ED physicians, and 39 (61%) were hospitalist physicians. Of the hospitalist physicians who submitted responses, 11% were also Kaiser Permanente at Home rounding physicians.

Figure 1:

Respondents’ specialties and involvement with Kaiser Permanente at Home (N = 55 respondents). Note: “EVE00” refers to a swing shift in which the attending hospitalist physician admits patients after-hours to the Kaiser Permanente at Home service. ED = emergency department; KP@H = Kaiser Permanente at Home.

To create a baseline for physician comfort level with change management, respondents were asked to describe their comfort levels with change both before and after Kaiser Permanente at Home was initiated. As indicated in Figure 2, there was minor variation in physicians’ overall perceived capacity for implementing change in clinical practice over time (which was during the COVID-19 pandemic).

Figure 2:

Physicians’ general comfort level with implementing change in daily practice before and after initiation of the Kaiser Permanente at Home program (N = 55 respondents).

Respondents were next asked to describe their comfort level with Kaiser Permanente at Home before implementation, 6 months after implementation, and again 1 year after implementation. Figure 3 shows overall respondent answers, indicating that most physicians (> 91%) took many months to a year to feel more comfortable with this care model for their patients.

Figure 3:

Comfort level referring to Kaiser Permanente at Home: initially, then after 6 months and again 20 months after implementation of Kaiser Permanente at Home (N = 55 respondents). KP@H = Kaiser Permanente at Home.

As presented in Figures 4 and 5, respondents were asked to identify the specific change management interventions that seemed most effective in improving their comfort level with and understanding of the Kaiser Permanente at Home program. Overwhelmingly, respondents described a combination of these interventions as similarly important (choices rated as “some” or “high” impact). However, “[p]ersonal communication between Kaiser Permanente at Home physician and you” and other interventions that involved direct communication between a physician and their peer were more highly ranked.

Figure 4:

Impacts of change management techniques on the willingness of physicians to refer patients to Kaiser Permanente at Home (N = 55 respondents). EMR = electronic medical records; KP@H = Kaiser Permanente at Home.

Figure 5:

Relative importance of change management interventions to physician participation in the Kaiser Permanente at Home program (N = 55 respondents). EMR = electronic medical records; KP@H = Kaiser Permanente at Home.

Qualitative Results: Thematic Structure

Of the 55 survey respondents, 22 agreed to a one-on-one interview (40%). Because of time and scheduling constraints, 12 were randomly selected (6 ED physicians and 6 hospitalist physicians) for interviews. The 4 themes that emerged during interviews are presented in the following discussion, along with supporting verbatim excerpts from participant interview transcripts.

Concern for Kaiser Permanente at Home to Demonstrate Clinical Equivalence to Hospital-Level Care

Participants’ concerns for this model of care were typically focused on its ability to demonstrate clinical equivalence to hospital-level care. Their focus surrounded the well-being of their patients and their concept of typical hospital care as being the gold standard.

For instance, Dr A, a hospitalist physician, explained:

“Initially not having face-to-face contact and 24/7 support like in a hospital and a sense that I didn’t believe that they would get the same care as in a hospital. I didn’t see how it could have been the same as in the hospital. Do the paramedics really know how to assess patients as well as a nurse would in the hospital? Also, a virtual physical exam is not something trusted by me.”

This concern was echoed by ED physician respondents such as Dr B who explained:

“I was concerned that they wouldn’t be able to monitor my patients overnight as closely as we do in the hospital, and whether things needed to be escalated If they could be in a safe manner.”

Dr B also added,

“I was worried they wouldn’t be able to handle the acuity of my patients and my bounce-back rate would increase.”

Note that the terminology “bounce-back rate” refers to a patients’ return to an ED after ED discharge.

Concern for Administrative Issues

As noted in the preceding section, clinical concerns are not the only barrier to care with Kaiser Permanente at Home. Sometimes, the administrative burden placed on a physician also becomes a barrier. This was described by Dr C, who explained

“There were a lot of inefficiencies at first in the method for sending my patients to KP at Home. It took a lot longer initially for me to do that work than it would have for me to keep the patient in the hospital”

Dr D, an ED physician, echoed these concerns, explaining

“The administrative burden was a big deal for me. I think the whole “what am I supposed to do” felt overwhelming – in the sense of learning a new flow of admitting my patients to KP at Home.”

Things Kaiser Permanente at Home Did That Improved Willingness to Refer Patients

In response to questions regarding interventions that alleviated physician concerns and improved comfort level in Kaiser Permanente at Home, both ED and hospitalist physicians typically responded that the intervention that helped them improve their concept for Kaiser Permanente at Home the most was direct physician-to-physician communication. This feedback supports the earlier overall survey findings presented in previous sections.

This was exemplified by Dr E, who explained:

“When we have high census and you have a KPAH doc who reaches out and says, ‘I looked at this [patient] and ...,’ it really helps me focus my attention on what patient aspects are appropriate for KP at Home care. Over time, these conversations helped educate me much better than if you had just handed me a list of criteria.”

In addition, the ability to identify appropriate patients is not second nature to most physicians. Dr D, an emergency room physician, explained:

“When I’m taking care of ED patients, it’s hard for me to focus on which specialist to contact when. I really appreciate it when the KP at Home team reaches out to me to help me think about KP at Home.”

Direct physician-to-physician communication also seemed to alleviate administrative concerns as described by physician G, a hospitalist, who explained:

“I remember I was thinking, ‘it would be so much easier for me to just keep my patient here than going through all the work of sending them to KP at Home.’ When I explained this to the KP at Home physician, they walked me through the steps, and it made it so much easier. I was much more willing to work with KP at Home after that.”

Communication and Education About the Program

Finally, the effectiveness of other interventions was also noted as important. This was described by Dr F, an ED physician who said:

“Continuing to educate and partner with the ED at our meetings creates so much good will because of the spirit of cooperation, it helps me feel more comfortable using KP@H.”

Discussion

Spurred by the COVID-19 pandemic, use of the ACaH model of care has increased substantially. However, to scale a successful program to a higher patient census, a key aspect is addressing physician acceptance. According to the results of this study, physicians described the use of a combination of interventions as impactful, with the use of personal, peer-to-peer communication being highly impactful. The follow-up interviews echoed these results. Furthermore, the results showed that it took up to a year before the majority of respondents felt more comfortable with this model, while their openness to change in general remained steady.

The interventions used in the Kaiser Permanente at Home program targeted several underlying reasons for physician hesitancy toward this care model. These included education regarding the clinical model and assistance with the referral and admission process. This study revealed that, over the course of these interventions, physician favorability toward Kaiser Permanente at Home exhibited a positive trend. This was corroborated by the increasing average daily census of the Kaiser Permanente at Home over the same period.¹³

Echoing findings from other studies focused on physician change management,^14–18 this study emphasized a multifaceted strategy based on change management theory to gain meaningful acceptance by physician stakeholders. In addition, these change management interventions are known to take time. On average, it took up to a year before the majority of respondents felt more comfortable with this model. This finding is consistent with other change management studies describing adaptation to change management requiring a realistic timeline that could be on the order of years.¹⁹ In addition, peer-to-peer interactions were found to hold particular impact. This result is consistent with studies describing the importance of one-on-one interactions toward physician change management.^16–18

Although the majority of respondents reported that the inclusion of peer-to-peer interactions was instrumental in their increased comfort with the Kaiser Permanente at Home program, the time-consuming nature of dedicating a physician to assisting their colleagues in administrative and educational tasks could be considered a substantial drawback. However, because of its substantial impact on physician acceptance, although initially time-intensive for the Kaiser Permanente at Home physician, this physician’s time dedicated to peer-to-peer interactions may decrease as physician acceptance increases.

The limited respondent count for the online survey used in this study and possible responder bias may have reduced its internal validity and generalizability. In addition, the absence of respondent demographics brings up potential sampling bias concerns, further impacting result generalizability. Moreover, the low response rate could compromise data reliability and validity. The low response rate could have been due to limited methods of recruiting responders. Future studies might address this issue by expanding the methods used for recruiting respondents, for example, recruiting responders with different methods other than email or even offering incentives for participation. Although the one-to-one interview sample size was small, such a number is not uncommon in qualitative research.²⁰ Finally, the study’s scope, restricted to a single health system, Kaiser Permanente, within a distinct geographic region among physicians from Northwest Permanente, PC, might also limit the applicability of its findings.

Conclusion

In the pursuit of an increased average daily patient census for the Kaiser Permanente at Home program, the critical role of physician acceptance has been underscored. Physician feedback within this study highlighted the significance of adopting a multifactorial change management approach in ensuring acceptance toward the Kaiser Permanente at Home program. Particularly notable was the emphasis on peer-to-peer engagement, which emerged as a predominant influence driving physician participation. ED and hospitalist physicians valued one-on-one conversations to navigate the administrative burden of referring patients to Kaiser Permanente at Home. In effect, the Kaiser Permanente at Home physicians became “navigators” for their hospital colleagues. Finally, it is important to note that effective change management, as evidenced by this study, can span as much as a year or possibly more. Although initially time-consuming, the change management interventions associated with this study may decrease over time but will always remain necessary to maintain consistency and sustain referrals and admissions.

Supplementary Material

Supplementary Material 1: