Abstract
Part 1 of this essay argued that the root causes of primary care’s problems lie in (1) the low percent of national health expenditures dedicated to primary care and (2) overly large patient panels that clinicians without a team are unable to manage, leading to widespread burnout and poor patient access. Part 2 explores policies and practice changes that could solve or mitigate these primary care problems.
Initiatives attempting to improve primary care are discussed. Diffuse multi-component initiatives—patient-centered medical homes (PCMHs), accountable care organizations (ACOs), and Comprehensive Primary Care Plus (CPC+)—have had limited success in addressing primary care’s core problems. More focused initiatives—care management, open access, and telehealth—offer more promise.
To truly revitalize primary care, 2 fundamental changes are needed: (1) a substantially greater percent of health expenditures dedicated to primary care, and (2) the building of powerful teams that add capacity to care for large panels while reducing burnout.
Part 2 of the essay reviews 3 approaches to increasing primary care spending: state-level legislation, eliminating Medicare’s disparity between primary care and procedural specialty reimbursement, and efforts by health systems. The final section of Part 2 addresses the building of powerful core and interprofessional teams.
INTRODUCTION
For decades, the United States has undervalued and seriously underfinanced primary care. For most aspiring young clinicians, primary care is viewed as too much work for too little reward. Too few primary care clinicians means too many patients for each clinician to manage. The 2021 National Academies of Sciences, Engineering and Medicine report on primary care sounded the warning that “primary care in the United States is slowly dying.”1
Part 1 of this essay argued that financial neglect (low primary care spending) and large patient panels are key factors causing primary care’s problems. Part 2 explores practice and policy changes that can allow primary care to thrive. Part 2 begins by discussing improvement initiatives that have enjoyed limited success, but have failed to address low primary care spending and excessive panel size. The final lengthy section, “Hopes for the Future,” proposes far-reaching measures that may revitalize primary care—increasing primary care spending and building powerful teams.
LIMITED IMPROVEMENT INITIATIVES
Improvement initiatives come in 2 flavors: (1) diffuse programs enhancing multiple components of primary care, and (2) focused efforts targeting 1 specific primary care function. The diffuse initiatives discussed here are patient-centered medical home (PCMH), accountable care organizations (ACOs), and Comprehensive Primary Care Plus (CPC+). Three focused initiatives explored are care management, open access, and telehealth. This essay does not consider Direct Primary Care, a model that dramatically reduces panel size; its widespread adoption would leave millions of people without primary care.2 For each of these 6 initiatives, Tables 1 and 2 explore 4 key questions: (1) Is panel size greater or smaller? (2) Has access improved? (3) Has clinician burnout decreased? (4) Has primary care spending increased?
The Impact of Diffuse Improvement Initiatives
The Impact of Focused Improvement Initiatives
Diffuse Initiatives
PCMH
In 2007, primary care organizations adopted principles of the patient-centered medical home. In 2008 the National Committee for Quality Assurance (NCQA) created standards for practices to receive PCMH recognition.17 PCMH standards include such areas as team-based care, access, continuity, and knowing your patients. About 13,000 practices and clinics are recognized as PCMHs, with recognition common among community health centers.18
Many studies have evaluated the impact of PCMH recognition. A 2020 commentary cites limited evidence associating PCMH practices with most clinical and financial outcome improvements.19 Yet PCMHs have not reduced health disparities among vulnerable populations.17
Because PCMH is a diffuse collection of initiatives rather than a focused intervention, evaluation is difficult. “If you have seen one medical home, you have seen one medical home.”20 One PCMH commentator suggests, “Perhaps it is time to study interventions more focused in their content, target population, and desired outcomes.”19 My community practice spent much effort getting PCMH recognition, but nothing changed for our patients.
Primary care practices could increase revenue by reducing hospital costs, if the savings were returned to primary care. But even with cost savings, no standard mechanism exists to return the savings back to primary care.
ACOs
Accountable care organizations are groups of doctors, hospitals, and other health care providers who come together to provide coordinated care to their patients. When an ACO succeeds in delivering high-quality, lower-cost care, the payer shares the savings it achieves.
In 2012, the Centers for Medicare and Medicaid Services (CMS) launched ACOs for Medicare patients—the Medicare Shared Savings Program (MSSP). Medicare later added the Pioneer ACO and Next Generation ACO models. By 2018, over 600 ACOs were managing care for nearly 12 million Medicare beneficiaries. Similar programs function in the commercial, Medicare Advantage, and Medicaid markets.
In 2019, ACOs showed improving financial performance, generating small net savings relative to CMS’s benchmarks.21 ACO savings may be overstated, however, because ACOs can “cherry pick” healthier patients, lowering their costs in order to benefit from shared savings.22 And even when savings are generated, they are not necessarily channeled to primary care.
Comprehensive Primary Care Plus (CPC+)
In 2017, CMS launched CPC+, a 5-year program to support over 3,000 primary care practices. CPC+ practices are required to address access and continuity; care management; comprehensiveness and coordination; patient and caregiver engagement; and planned care and population health. CMS pays CPC+ practices a care management fee (in some cases up to $300,000 in 1 year) with additional incentive payments for reducing patients’ utilization or costs. Although a goal of CPC+ is to stimulate alternative payment models, in 2018 most CPC+ revenue remained fee-for-service.11 Geographic areas with CPC+ practices, compared with non-CPC+ areas, have higher median incomes, fewer households in poverty, higher mean educational level, fewer people on Medicaid or uninsured, and a healthier Medicare population, thereby raising questions about the validity of CPC+ evaluations.23
Focused Initiatives
Care Management
In 1996, Ed Wagner published his first of many papers on the Chronic Care Model.24 The model pioneered the concept that good care for patients with chronic illness is fundamentally different from acute care, and requires practice transformation with 2 central features: (1) chronic disease and preventive care registries, and (2) planned visits to provide patient education, medication management, and self-management support. These activities are now called “care management.” Much of the Chronic Care Model has been incorporated into high-performing primary care.25–27
Care management may be associated with reduced hospital utilization and costs for patients in PCMHs and ACOs.20,28 Patients with diabetes, asthma, or heart failure receiving care management have better outcomes, and sometimes lower costs, than patients without care management.28 Care management can assist in the care of patients with the chronic “long-COVID.”29 To reduce health care costs, care management works best for patients with multiple conditions and high costs.30 Care management patients have lower medical expenses, fewer hospital admissions and bed days, and fewer specialist visits.31 Care management does tackle the problem of large panels by adding care managers to assist clinicians to care for their panels. I did not benefit from care management in my community practice, meaning that time-consuming patient education, self-management support, and care coordination were my responsibilities.
Open Access
In the early 1990s, family physician Mark Murray rearranged his schedule so that his patients could see him the same day they called for an appointment.32 The innovation came to be known as advanced access. Murray showed that good primary care access requires that capacity—number of appointment slots in a year—equals demand for those appointment slots.33 Because reducing demand is difficult, adding capacity is the best option to improve access. When my community practice tried to implement advanced access, however, successes were short-lived and access deteriorated over time.
Murray’s innovation—same-day appointments for all patients—was watered down in a popular access improvement called open access: freeing up same- or next-day appointments for some but not all patients. In 2015, 79% of US family physicians reported that they used open-access scheduling.34 If 20 patients request open-access slots, however, and only 10 slots are available, the other 10 patients are denied prompt care. Without increasing capacity, the total number of appointment slots is unchanged.
Telehealth
Telehealth includes telephone visits, video visits, and electronic patient portals. While telehealth has existed for decades, its uptake was slow before 2020. In 2014, an estimated 15% of family physicians utilized telehealth.35 Kaiser Permanente of Northern California and the Veterans Health Administration were pioneers in telehealth, the latter providing services for rural and homebound disabled veterans.
The COVID-19 pandemic catalyzed an instantaneous shift to telehealth, which became a prominent feature of primary care and is expected to continue as the substrate for many primary care encounters. The proportion of telehealth ambulatory encounters increased from 10% just before the pandemic to more than 90% during the pandemic’s height.36
E-visits through the patient portal have been associated with improved patient access and increased capacity. A study at Kaiser Permanente found that face-to-face visits fell 25% after instituting the patient portal.37 A concern is that patient portal scheduling is used more heavily by younger, White, and commercially insured patients, which increases racial and economic inequities in access.38,39,40
Health Care Consolidation and Primary Care
US health care is increasingly provided within large health systems. Health systems have the potential to improve quality and efficiency, but also erode clinician autonomy while making health care more expensive and less responsive to patients.41 Health care consolidation often means vertical integration—1 or more hospitals plus medical groups within a single ownership structure.42 From 2012 to 2018, the proportion of physicians employed by hospitals rose from 26% to 44%.43 In 2020, 58% of family physicians were employed compared with fewer than 40% for surgical subspecialists.44 From 2010 to 2016, market concentration increased almost 29% for primary care compared with 5% for hospitals and specialist physicians.45 Consolidation spawns the relentless growth in practice size.46
Consolidation has not increased primary care spending. The percent of national health care expenditures across commercial payers going to primary care decreased from 4.88% in 2017 to 4.67% in 2019.47 Practices with 1-2 physicians have 33% fewer preventable admissions than practices with 10-19 physicians.48
When primary care physicians move to a vertically integrated practice, they reduce their clinical output by 10% to 20%, seeing fewer patients, generating less revenue, and threatening patient access.49 Clinicians owning their practices report less burnout compared with those in health system–owned practices.50 In a survey of 17,000 patients cared for by 367 physician offices, patients preferred small practices to large ones and reported better access in small practices.51 Consolidation is not primary care nor patient-friendly.
HOPES FOR THE FUTURE
The limited initiatives reviewed here have not increased primary care spending nor reduced panel size. The final section of Part 2 examines far-reaching policy and practice changes to increase primary care spending and build powerful teams that can assist clinicians in caring for their panels.
Increasing Primary Care Spending
In 2016, the United States spent 5.4% of total health expenditures on primary care, compared with 7.8% (other studies estimate 12%) by 22 Organization for Economic Co-operation and Development (OECD) countries.52 Ways to increase primary care spending include state legislation, federal action, and policies within health systems.
State Legislation
From 2009 to 2014, Rhode Island’s Office of the Health Insurance Commissioner required commercial insurers to raise their primary care spending rate by 1 percentage point per year. Commercial insurer payment to primary care increased from 5.7% in 2008 to 12.3% in 2018, increasing primary care dollars from $47 million to almost $80 million. The increased payments were designed to improve care; for example, hiring nurse care managers, implementing components of the Chronic Care Model, and increasing after-hours care.52–54
In 2017, the Oregon legislature mandated that large commercial insurers, Medicaid coordinated-care organizations, Medicare Advantage plans, and health plans serving public employees spend at least 12% on primary care by 2023.52 In 2018, primary care spending percentages varied widely among different insurers, from 4.3% to 22.6%. Fee-for-service Medicare is not included because the state lacks jurisdiction over Medicare.55
Overall, primary care spending legislation is challenging because more primary care financing does not reduce total health care costs, making it less attractive to policy makers and politicians.56
Federal Action
Primary care spending could increase nationwide by changing how Medicare pays primary care. Changes in Medicare payment are generally copied by Medicaid and commercial insurers. To alter Medicare payment requires severing the tight bond between Medicare and the American Medical Association’s (AMA’s) Relative Value Scale Update Committee (RUC), a procedural-specialty controlled committee that recommends how physicians are paid. Medicare—rather than evaluating the RUC’s recommendations—accepts them 90% of the time.57
One analysis found that 30-minute primary care office visits for complex patients generated 40% of the fee for gastroenterologists performing 30-minute colonoscopies—emblematic of the vast disparity between payment for ambulatory visits vs procedures.58 Given the staying power of fee-for-service,59,60 increasing primary care spending requires reducing the payment disparity between cognitive visits and procedures.
A major input into fee-for-service payment is the time required to deliver the service. The RUC surveys physicians—chiefly procedural specialists—asking them how much time each procedure requires. The surveyed specialists make more money if they overestimate the procedure time. A study using electronic medical record (EMR) time stamps for 293 procedures found that the objective procedure times were on average 20% lower than the specialists’ estimates accepted by the RUC. Another investigation found that the RUC overstated times by 18% to 61% depending on the procedure.57,61 Most RUC members are appointed by specialty societies, with only 5 of 32 from primary care specialties.1
In 2015, the US Government Accountability Office (GAO) recommended that Medicare more thoroughly review RUC’s payment recommendations, but changes did not occur.62 The National Academies of Sciences, Engineering and Medicine report on primary care concluded that the RUC could not be reformed and that Medicare should value physician services independently of the RUC.1
Increasing Health Systems’ Primary Care Spending
Primary care spending can be augmented by health systems or insurers without governmental action. Currently primary care spending rates vary widely among health systems and insurers. Previous data found that Geisinger spent 9% on primary care, Intermountain HealthCare 8%, and Group Health (now Kaiser Permanente Washington) 14%.53 Most health systems have far lower primary care spending rates or do not track this metric at all. One caveat: the variation in how primary care spending is measured makes comparisons treacherous. Change in primary care spending over time is more reliable.
Value-Based Payment May Not Increase Primary Care Spend
Alternative payment models give primary care more flexibility to encourage team-based care.63 Changing the payment model, however, may not bring primary care more revenue. For example, in my practice, insurers set capitation rates equal to, but not more than, their estimate of what they would have paid for those patients under fee-for-service.
Increasing primary care spending can be accomplished rapidly given political will. Primary care seems to be the only health care institution, however, that is expected to save money for the health system overall. In truth, primary care’s value lies in providing care and improving health outcomes for tens of millions of people.64
Powerful Teams
As shown in Part 1, primary care panel size is too large, and cannot decline due to the clinician shortage. To address panel size requires a powerful team sharing the care of the panel.65 Although not all patients benefit from team care, the role of teams is to assist clinicians in caring for their panel. Powerful teams add capacity while reducing burnout, yet few teams have shown that they can accomplish these goals.66
The team narratives described in this section rely on visits to “bright spot” practices—practices seeking to overcome the impact of large panels. The bright spots featured here are practices at which I have conducted site visits, and are thus only examples. Many bright spots exist throughout the country that are not featured here. Moreover, bright spots seldom shine forever. They can lose their luster if leadership changes, if the business case fails, or if key personnel leave. Some bright spots described here have already dimmed. Yet evanescent bright spots continue to teach us ideas that work.
Primary care teams are often composed of a core team or teamlet (commonly a clinician working with a medical assistant) and an interprofessional team (for example, registered nurses [RNs], pharmacists, behaviorists, and physical therapists). The core team is responsible for its panel of patients. The interprofessional team assists several core teams for patients requiring more services.65
Powerful Core Teams
Bellin Health, in Northeastern Wisconsin, initiated team-based care in 2014. By 2019, all 130 primary care clinicians were involved in team-based care.67 The central innovation is the expansion of the core team to 2 upskilled medical assistants (renamed care team coordinators [CTCs]) per clinician. Table 3 describes how clinician visits have become team visits. Relieved of documentation tasks, clinicians see more patients each day, adding capacity while increasing clinician satisfaction.67
Contrasting Bellin Health’s Team Model With the Traditional Model
By 2018, a core team model similar to Bellin Health’s was up and running in several University of Colorado primary care clinics. Hypertension control, colorectal cancer screening, and most diabetic quality metrics improved. New patient appointments grew markedly, leading to increased revenue and better access. Clinician burnout dropped from 56% to 25% in one clinic and from 40% to 16% in another. Staff burnout in one clinic fell from 42% to 21%, perhaps because medical assistants—traditionally excluded from the clinician visit—have a more interesting job as participants in the “room where it happens.”68,69
Scribing is a core team innovation in which 1 core team member performs in-room documentation. 2018 data from 100 million patient encounters with 155,000 physicians shows that physicians spent an average of 16 minutes per encounter using the EMR, with primary care physicians on the high end of the specialty distribution.70 Working with scribes is associated with reduced burnout, decreased charting time, and high physician and patient satisfaction.71–75 In 1 study, visits per clinic session increased 29%, adding revenue that more than paid for the scribes. Physician time after hours went down by 38%.75 Scribe use has been associated with a 60-minute daily time saving for clinicians.76
Why have only a few practices adopted a powerful core team model? Trust among team members can dissipate quickly with one negative encounter. As much as clinicians dislike the EMR, giving up the keyboard and cursor to another person is a stretch. Patients may reject care by non-physicians.77 In my community practice, patients initially refused appointments with our nurse practitioner (NP), but after one NP visit, they often switched to her care. The additional personnel and their training cost money and regulatory requirements can be tricky. Staff absences and turnover throw a wrench in the system. Lower burnout among clinicians can be associated with higher burnout among practice staff.78 To succeed, everyone needs to win: patients, clinician, staff, and the health system.79
In summary, core teams are powerful if they add capacity and reduce burnout. To achieve these goals, they need to save clinician time, particularly EMR documentation and the heavy burden of in-box messages. The AMA STEPSforward Saving Time Playbook proposes a menu of time-saving and burnout-reducing activities core teams can perform.80
Powerful Interprofessional Teams
Members of the interprofessional team vary from clinic to clinic; we focus on RNs, pharmacists, behaviorists, and physical therapists, all of whom can also offer in-person and telehealth encounters with good quality. On some occasions, interprofessional team members can manage a subpanel of patients within their expertise—for example patients with diabetes—with minimal clinician oversight. Too many team members for one patient are confusing for patients and team members alike. Facilitators and barriers regarding interprofessional teams are summarized in Table 4.
Interprofessional Team Facilitators and Barriers
Registered Nurses
RNs can contribute to primary care in several ways. Two of these are RN co-visits (Table 5) and care management. A study of RNs at 13 community health centers found that RNs confined to telephone triaging are often frustrated, but those doing co-visits and care management fully utilize their professional skills.89
RN Co-Visits
Community Care of North Carolina (CCNC) situated RN care managers around the state to provide care management for patients of small practices with chronic conditions. Hospital admissions and emergency department visits for high-risk Medicaid patients dropped dramatically and risk-adjusted costs were 15% lower than for non-CCNC patients.90
The most effective care management involves RN authorization to manage medications through (1) physician-created standing orders that allow RNs to pend prescriptions in the EMR or (2) patient-specific orders for RNs to prescribe a particular medication to a particular patient. A few state nursing boards allow RNs to adjust medication doses under physician-approved standing orders.91 RN care managers able to make medication changes can significantly improve hemoglobin A1c levels in patients with diabetes compared with usual care.92 Intermountain HealthCare found that physician productivity was 8% higher for clinics with care managers. The additional revenue outweighed the program’s cost.93
Clinicians performing care management themselves suffer greater burnout than clinicians delegating coaching to team nurses.94 More task delegation to nurses is associated with lower burnout among clinicians but more burnout for RNs on the team.95
Pharmacists
When pharmacists manage medication-related care, physicians have time for additional patient visits.96 Diabetes care provided by pharmacists improves diabetes and hypertension outcomes.97,98 Primary care clinicians report that pharmacists performing medication management decreased workload, reduced mental exhaustion, and increased patient access.99,100 At one hospital, 27% of chronic disease patient appointments were converted to pharmacy appointments, opening access for other patients.101 Small practices, unable to hire a pharmacist, can share pharmacist time with similar practices in their health system or network.
Behavioral Health Professionals
Behaviorists include psychologists, licensed clinical social workers, marriage and family counselors, drug/alcohol counselors, and others. The integration of behavioral health into primary care has spread over the past 30 years,102 though only 26% of family physicians reported working with a behavioralist in 2018.103 Behavioral health and primary care can be co-located, with warm handoffs to behavioralists working in physical proximity; or integrated, with clinicians and behaviorists creating one treatment plan with behavioral and medical elements.104
Behavioral health–primary care integration is associated with improved mental health, diabetes, cardiovascular, and chronic pain outcomes; it can reduce the number of physician visits, adding capacity.104–106
Physical Therapists
These interprofessional team members are experts on musculoskeletal conditions that make up about 25% of primary care visits. Patients with direct access to physical therapy (seeing the therapist first) vs physician referral had more fully achieved goals, less average pain at discharge from care, fewer missed days from work, higher satisfaction, fewer imaging studies, and lower health care costs.107 Bellin Health co-locates physical therapists in primary care teams to receive warm handoffs for patients with musculoskeletal complaints. Therapists see about 8 patients per day, adding primary care capacity.85
Large Interprofessional Teams
Table 6 provides a follow-up to the thought experiment in Part 1 of this essay, visioning how an interprofessional team could add capacity and reduce burnout.
Demand and Capacity for Interprofessional Teams
Few primary care practices, however, have large interprofessional teams. In a 2017-2018 family medicine survey, 38% reported working with a team including one or more behaviorist, physical therapist, and pharmacist. Small practices can build interprofessional teams by sharing personnel with other practices in that health system. Working with interprofessional teams, physician burnout was 21% when teamwork was effective but 69% when teamwork was poor.108 Care management is the mechanism through which interprofessional team members shift the time-consuming function of behavior-change counseling from physicians to team members.
CONCLUSION
My 50 years in practice, teaching, and policy writing have been a love affair with primary care. Primary care means patients from all backgrounds struggling with thousands of symptoms or diagnoses, placing their trust in us. Primary care means knowing the intimacies of patients, their families, and their lives over years or decades. Primary care is built on patients knowing us and we knowing them. Primary care is unique in the panoply of health care services.
This 2-part essay reflects my thinking about primary care. I believe that the root causes of our problems lie in financial neglect and too many patients to handle. As a result, patients have a hard time getting enough time with us, and time is the coin of the realm. Limited initiatives trying to mitigate these problems have scarcely made a dent in our fortunes. Lurking behind this disappointment is low primary care investment.
To counter these difficulties, teams in bright spot practices give us hope for the future, but sustaining these teams is challenging and requires more primary care spending. Conversely, new primary care dollars are best focused on sustaining these teams. Primary care spending and powerful teams need each other. Primary care needs both.
Barriers are daunting. But consider the status quo. Patients can’t get appointments while exhausted clinicians spend hours on documentation. We cannot continue to care for too many patients without teams to share the care. With adequate primary care spending and powerful teams, primary care can become accessible to patients and joyful to all.
Footnotes
Conflicts of interest: author reports none.
- Received for publication December 5, 2021.
- Revision received May 16, 2022.
- Accepted for publication May 27, 2022.
- © 2022 Annals of Family Medicine, Inc.
REFERENCES
- 1.↵
- 2.↵
- 3.
- 4.
- 5.
- 6.
- 7.
- 8.
- 9.
- 10.
- 11.↵
- 12.
- 13.
- 14.
- 15.
- 16.
- 17.↵
- 18.↵
- 19.↵
- 20.↵
- 21.↵
- 22.↵
- 23.↵
- 24.↵
- 25.↵
- 26.
- 27.↵
- 28.↵
- 29.↵
- 30.↵
- 31.↵
- 32.↵
- 33.↵
- 34.↵
- 35.↵
- 36.↵
- 37.↵
- 38.↵
- 39.↵
- 40.↵
- 41.↵
- 42.↵
- 43.↵
- 44.↵
- 45.↵
- 46.↵
- 47.↵
- 48.↵
- 49.↵
- 50.↵
- 51.↵
- 52.↵
- 53.↵
- 54.↵
- 55.↵
- 56.↵
- 57.↵
- 58.↵
- 59.↵
- 60.↵
- 61.↵
- 62.↵
- 63.↵
- 64.↵
- 65.↵
- 66.↵
- 67.↵
- 68.↵
- 69.↵
- 70.↵
- 71.↵
- 72.
- 73.
- 74.
- 75.↵
- 76.↵
- 77.↵
- 78.↵
- 79.↵
- 80.↵
- 81.
- 82.
- 83.
- 84.
- 85.↵
- 86.
- 87.
- 88.
- 89.↵
- 90.↵
- 91.↵
- 92.↵
- 93.↵
- 94.↵
- 95.↵
- 96.↵
- 97.↵
- 98.↵
- 99.↵
- 100.↵
- 101.↵
- 102.↵
- 103.↵
- 104.↵
- 105.
- 106.↵
- 107.↵
- 108.↵