How to Have a High-Performing Employed Medical Group Without a Hospital Subsidy

Article · November 22, 2017

Hospital-employed physician groups notoriously hemorrhage money. This issue is one of the most frustrating challenges faced by health care system leadership teams across the country. It also represents a source of endless ridicule from the independent physicians who get along just fine without hospital support. In 2014, the authors were hired by Michael Young, the CEO of PinnacleHealth System, to create a high-performing physician group, improve patient care, and reduce the hospital financial subsidy.

Industry History

It is important to understand the origins of physician employment by health care systems. Two factors conspired to change hospital economics in the late 1970s and early 1980s. Specifically, the development of managed care and the establishment of diagnosis-related groups resulted in average length-of-stay and bed occupancy being cut in half. As a result, maximizing hospital occupancy became the sole focus of hospital management.

Because the main drivers of hospital admissions were physicians, the challenge for hospital management teams was to figure out how to get physicians to admit patients to their hospitals as opposed to their competitors’ hospitals. The “solution” was to buy the physician practices and pay physicians a premium salary that was higher than what they had been earning as independent practitioners, in the hope of capturing all their inpatient work.

The results were utterly disastrous. Hospital management teams never understood these small businesses. Cost outlays related to hiring hospital employees (e.g., paying hospital-level compensation and providing benefits to the staff), federal regulations that were forced upon the practices, and new facility requirements that were previously required only of hospitals further increased expenses.

Most importantly, the providers were now completely disengaged from the decision-making process. Physicians who were trained to be self-reliant and to question authority were now being governed by mid-level administrators with little or no experience running physician practices.

These small businesses previously had been very lean, with the physicians working 60-hour work weeks when self-employed. The new hospital employment agreements called for a mere 36 hours per week in the clinic, with fixed salaries. The contracts were further padded with 4 weeks of paid vacation, 1 to 2 weeks for continuing medical education, and 10 to 12 paid holidays per year. So, in effect, all physicians were now working barely half as much as they had been — and at a higher salary.

As a result, practice overhead skyrocketed and productivity plummeted. Regulatory and compliance issues further burdened these practices. Buyer’s remorse soon set in among hospital management teams. Doctors were called lazy, and the practices generated huge losses. The relationship between employed physicians and hospital management deteriorated, never to recover. Make no mistake: for hospitals, physician employment was a self-inflicted wound.

PinnacleHealth Medical Group

In 2014, PinnacleHealth Medical Group (PHMG) included 100 providers: 70 physicians and 30 advanced clinical practitioners (ACPs), mostly nurse practitioners (NPs) along with some physician assistants (PAs). Those providers occupied 25 clinical sites in the three counties of Pinnacle’s service area. The group had annual revenues of approximately $36 million and expenses of $45 million, resulting in a loss of $93,000 per full-time-equivalent (FTE) provider.

After many attempts to rectify the losses at PHMG, the authors were tasked with three objectives that had eluded previous leaders of the group:

  1. Reduce the loss/subsidy/investment per provider to under $50,000/year.
  2. Significantly improve the quality of patient care.
  3. Grow the number of patients cared for by the group.

Upon review, it became clear that several issues needed to be addressed.

Contracts and Access

The provider employment contract was not being enforced. Although the contract called for 36 hours of clinical time, providers actually saw patients for only 59% of the contracted time. Office hours of operation were strictly limited to 9 to 5, with no nights, extended hours, or weekend availability.

As a result, patient access was severely limited. Over the years, providers had introduced >1100 “rules” to determine when various kinds of patient visits could be put into their schedules. We conducted several Secret Shopper surveys and discovered that appointments were inexplicably denied to our fictitious patients, despite there being numerous openings on the schedule.

Our first action was to take control of the schedule away from the providers and to transfer the responsibility of booking patients to office managers who answered to the Director of Operations. Our second action was to increase the hours of operation by pushing the opening time to 7 a.m. and adding extended hours for most weekdays as well as Saturdays. We also opened the schedule to same-day requests and walk-in appointments and eventually allowed patients to self-schedule through our patient portal. As a result of these steps, the proportion of time that providers spent on direct patient care increased from 59% to 95% of their contracted hours.

Simultaneously, we developed two retail clinics at local supermarkets and began to add Express Care locations at our office locations and even in our emergency room at the hospital to increase access, resulting in a drop in ED utilization by >50%.

We continued the Secret Shopper calls until we were satisfied that the access issue had been solved.


Patient quality scores were abysmal. For example, only around 20% of our patients were receiving screenings for colon, breast, and cervical cancer. Few, if any, patients were being contacted or seen after inpatient stays to coordinate their care. The physicians had their heads down and only seemed to deal with the problem at hand during the visit, with no focus on the long-term risk management that is essential for better outcomes. As a result, payments for quality were minimal.

Several tactics were used to improve the physicians’ quality performance. We identified gaps in patient care by reviewing the electronic medical records (EMRs) and made sure that patients came in to have the care gaps closed. We tasked medical assistants, nurses, and ACPs to assist the physicians with adult wellness visits and any identified gaps in care. These steps allowed everyone to work at the top level of their licensure and freed physicians to focus on more complex tasks.

In one office, an NP took it upon herself to preview the charts of all patients being seen by providers in the office that day and to flag all items that were required so that the providers would order them during the visit. That office moved from the worst-quality performer to the best-quality performer within 6 months.

Finally, when the hospital installed a new EMR system, we created a process that required providers to address any outstanding quality gaps prior to completion of their visit notes.

These efforts paid off quickly: group quality payments increased to $6 million in 2015 and to >$10 million 2016.


A new compensation plan was needed to ensure that we were paying market-rate compensation for market-rate productivity. The providers, including both the physicians and the ACPs, agreed to simplify the plan by linking the Medical Group Management Association national survey data for compensation to the corresponding level of productivity. Under the model, productivity was calculated based on the average of three components: patient encounters, work relative value units, and collections. The provider was then paid at the corresponding percentile of that average.

We also withheld 20% of each provider’s biweekly salary draw and tied it to achieving agreed-upon patient quality metrics such as screening levels for breast, colon, and cervical cancers, control of diabetes, and compliance with standardized protocols for the management of congestive heart failure and chronic obstructive lung disease.

Operations and Governance

We streamlined our corporate operations and replaced both the Medical Director and the Director of Operations. In addition, we disbanded the PHMG Board of Trustees and all of the combined lay physician committees, which we deemed burdensome, bureaucratic, and unnecessary. We retained only the compensation committee, as it was integral to our core business. We also appointed a new Medical Director, recruited a Chief Quality Officer (a family physician with extensive quality improvement experience), and added a Doctor of Pharmacy to our team.

We then appointed one provider from each site to serve as the Site Leader. The Site Leaders, which included both physicians and ACPs, were responsible for bidirectional communication between management and providers at their respective sites. The idea was to decentralize management to recreate the private practice ethos that had been lost. The Site Leaders met every other week to discuss ways to improve our performance and to implement the transformation of our care model.

Consolidation and Centralization

Each of the 25 PHMG offices employed staff to perform such tasks as scanning, handling managed care referrals, answering phone calls, scheduling appointments, filling prescriptions, etc. One by one, we moved these FTEs out of the physician offices and into the largely unoccupied corporate office, thereby centralizing these functions to achieve significant economies of scale and allowing us to reduce the workforce from 535 to 460. With the reduced staffing at the offices, we were now able to consolidate many of these 25 small, inefficient, and tired spaces to 15 larger Class-A offices. The final step was to create a centralized call center with a nurse triage group that used standardized protocols to route our patients to the appropriate level of care at the most convenient location.

Behavioral and HR Issues

During the first few months that our new leadership team was in place, there was some resistance to the changes we were implementing. These challenges were all dealt with swiftly and definitively. In the first year, close to 20% of providers separated from the organization either by their own volition or through termination because of noncompliance with the new process and policy changes or the compensation plan. Some saw it as a kind of brutal frontier justice, while others applauded it as long overdue.

Patient Retention

Unlike in mature managed care markets, much of the specialty care in Central Pennsylvania is “self-referred;” that is, the patients decide what care they need and who will provide that care, with no financial consequences for either the patients or the providers. In 2014, >80% of the specialty care that was required by our patients was provided by specialists outside of our system at competing hospitals. As a result, care was highly fragmented and completely uncoordinated.

In addition, the loss of revenue to the system was substantial. To address the problem, we first educated PCPs on the importance of care coordination as well as on the negative financial impact of out-of-system referrals. We then provided them with the rationale and scripting for their patient conversations, emphasizing the importance of continued PCP involvement in patient care.

Finally, we established a process whereby any out-of-system referrals had to be requested in writing. PCPs were required to explain why the care could not be provided inside the system by one of our specialists. This task was annoying and burdensome, but the tactic proved to be highly effective. During the first month, we received an average of 10 requests for out-of-system referrals per day. Over time, as the PCPs observed the improvements in care (and especially care coordination) that were possible when patients remained inside the system, the requests for outside referrals decreased to 10 per month and the leakage rate plummeted to <10%. Retaining specialty care and the associated revenue generated at our hospitals helped to offset the financial impact of the drop in medical utilization that our care transformation model was having on the system.

Clinical Innovations

Once the top-down model governance and management structure was replaced with the new Site Leaders model, the providers responded enthusiastically and put forth several new and innovative ideas, including the following:

  • A comprehensive approach to the overprescribing of opioids, resulting in a 90% drop in opioid prescriptions.
  • Worksite clinics in collaboration with large local employers.
  • A diabetes management initiative, which produced a 1.5% drop in the hemoglobin A1C levels of our patients.
  • Reengineering and standardization of the office policies and practices to eliminate waste and optimize the level of licensure function.
  • New standing orders that allowed medical assistants and nurses to be involved in the care process directly with patients.
  • Delineation, standardization, and expansion of the role of the ACPs.
  • Development of the Team Care Model, with the creation of teams of physicians, ACPs, and medical assistants, to provide patients with access to more team members and thereby facilitate the delivery of all aspects of care provided at the medical home, such as care navigation, health coaches, diabetic educators, and pharmacists.
  • Improved CGCAHPS (Clinician and Group Consumer Assessment of Healthcare Providers and Systems) scores and improvement of the patient experience. A Patient Engagement specialist was hired not only to assist our providers in direct patient involvement, but also to coach and train them in their day-to-day patient interactions.
  • Continued transformation of the care model from traditional fee-for-service care to value-based care that included colocation of behavioral health, palliative care, and addiction services into our primary care offices.


News of our success led to several positive developments:

  • One large primary care group approached us and was acquired in 2014, and several more small practices joined us in 2015 and 2016.
  • Recruitment of graduates from top medical schools and residents from training programs such as Johns Hopkins, Yale, and the University of Pennsylvania had been quite rare for PHMG. That all changed once we began to transform the culture and reenergize providers. We recruited 30 new primary care physicians between 2015 and 2016.
  • Our reputation for treating ACPs with equal status to the physicians also helped us attract 45 top-level ACPs.
  • Our group grew to >150 providers, and our overall number of patients increased by >50% (from 120,000 to >200,000).
  • Despite the drop in ED and hospital utilization that our new care model produced, growth in our patient panels, increased patient retention, and the increase in quality payments that we received allowed the hospital to achieve operating margins that averaged between 4% and 6% from 2014 to 2106.

Summary of Results

  • From Q1 2014 to Q4 2016, the subsidy of $93,000 per provider was eliminated.
  • Quality metrics improved from the 20th to 30th percentile to the 85th to 95th percentile according to national data.
  • Quality payments went from approximately $300,000 per year to over $10 million.
  • Utilization rates in the ED dropped by 50%, and inpatient admissions dropped by 35%.
  • The number of staff FTEs decreased by 14% (from 535 to 460).
  • Annual rent decreased by 25%.
  • Improved recruitment processes led to a 50% increase in the number of PCPs and a 100% in the number of ACPs.
  • The number of patients in the group grew by >50% (from 120,000 to 205,000).
  • The group developed the resources and staffing required to successfully navigate the transition to value-based care while living in a predominantly fee-for-service environment.

The Path to Profitability

Those outside the health care industry will undoubtedly find many of the changes that were instituted to be self-evident. However, the need for a perpetual hospital financial subsidy for employed primary care groups has plagued hospital administrators for decades. We believe we have demonstrated that implementing these common sense tactics could drastically reduce and even eliminate the need for hospitals to expend precious resources to support employed PCPs.

Call for submissions:

Now inviting expert articles, longform articles, and case studies for peer review


A weekly email newsletter featuring the latest actionable ideas and practical innovations from NEJM Catalyst.

Learn More »

More From Leadership

Small Change, Big Consequences — Partial Medicaid Expansions under the ACA

The stakes are high in the debate over partial Medicaid expansions, with far-reaching consequences for patients, state governments, and the federal budget.

The “Stranger Effect” — A Look at Interactions Between Consultants and Care Teams Through the Lens of Social Science

Why do miscommunications between teams and consultants occur, and what can we do about it?

Harm Across the Board - Preventable Harm Metrics Tracked at Seven Mission Health Hospitals

Defining “Directionality” in Quality for the Board

Mission Health’s Quality Team adapted the Harm Across the Board index to seven different hospitals and combined the indices into one regularly reportable systemwide measure.

UCLPartners Leadership Competency Framework

Building a New Kind of Leader for an Era of Cooperation

Moving from competition to collaboration requires different leaders, and a different way to evaluate leadership potential.

Wanted: Talented, Energetic, Creative People to Work on Difficult, Boring Problems. No Perks.

Let’s admit the inconvenient, boring truth.

Physician Coaching: Clinicians Helping Clinicians on the Things That Matter Most

A physician coach reflects on who gets better, what works, and how to create a learning culture.

Jessica Dudley and Namita Mohta head shots

Professional Fulfillment: Where We Want to Be

How the CMO of Brigham and Women’s Physicians Organization combats burnout at her institution and empowers physicians with leadership skills.

Texas Medical Center Houston Methodist Checklist to Maintain Operations During Hurricane Harvey and Future Flood Events

Lessons in Leadership: How the World’s Largest Medical Center Braced for Hurricane Harvey

Due to learnings from previous hurricanes, five success factors kept the Texas Medical Center operational during one of the worst natural disasters in recent U.S. history.

Lessons in Leadership: How the BayCare Health System Withstood Hurricane Irma

BayCare’s extensive disaster preparations were crucial in weathering the storm.

The VA Is a Quality Leader, Contrary to Many People’s Assumptions

A defense of the Veterans Health Administration, from its former Acting Under Secretary.


A weekly email newsletter featuring the latest actionable ideas and practical innovations from NEJM Catalyst.

Learn More »


Physician Burnout

52 Articles

Survey Snapshot: How Do You Know…

The NEJM Catalyst Insights Council discusses strategies for clinical engagement.

Team Care

111 Articles

What AI Means for Doctors and…

Physicians must hone the “four Cs” — critical thinking, communication, collaboration, and creativity — when…

Leading Teams

171 Articles

What AI Means for Doctors and…

Physicians must hone the “four Cs” — critical thinking, communication, collaboration, and creativity — when…

Insights Council

Have a voice. Join other health care leaders effecting change, shaping tomorrow.

Apply Now