The Shortage of Primary Care: Potential Solutions

By Robert L. Moore, MD, MPH, MBA, Chief Medical Officer

“If I had an hour to solve a problem I’d spend 55 minutes thinking about the problem and 5 minutes thinking about solutions.”

– Albert Einstein

As we emerge from the COVID pandemic, the health care delivery system is not yet recovered.

A point-in-time survey of medium and large primary care physicians (PCPs) in the Partnership HealthPlan of California (PHC) network showed an overall current vacancy rate of 24% of which two thirds were openings for PCP physicians (about 200 openings) and one third were openings for nurse practitioners/physician assistants (about 100 openings). The current PCP openings are in all 14 counties and in all practice settings. The overall need for PCPs is even greater if we want to have sufficient clinicians to do all the preventive and chronic disease care that is recommended by the United States Preventive Services Task Force (USPSTF), the American Academy of Pediatrics, and the Department of Health Care Services (DHCS).

This is not just true in the PHC service area but also nationally and internationally. The United Kingdom is having the largest primary care physician shortage since the start of the National Health Service in 1948. A few clinicians died of COVID, more were disabled by long COVID, and others retired early or stopped doing primary care. As the supply dries up, the demand per clinician remaining in practice increases causing stress and burnout. This leads clinicians to take a break from practice or reduce their hours per week making the shortage worse.

The rise of hospitalists has been a major driver of loss of primary care access. Twenty years ago, about 40% of all internal medicine residents became PCPs. Now that number is about 15% with most of the shift occurring in the rise of the number of hospitalists caring for patients in the inpatient setting.

The number of PCPs plus hospitalists is equal to the number of PCPs needed in a community. The steady decrease in the number of internal medicine specialists in primary care (as older primary care internists retired) was not accompanied by a sufficient increase in training family physicians, nurse practitioners, and physician assistants to overcome this loss. This trend started reversing in the mid-2010s with an increased number of primary care physician residencies as well as a substantial increase in the number of new nurse practitioners and physician assistant graduates.

While the hospitalist movement promised to improve quality and utilization, this has absolutely not happened. Hospital length of stay in the PHC service area is at an all-time high (6.23 days) and increasing steadily over many years. (Part of this increase length of stay is a shortage of skilled nursing facility beds.)  Poor continuity of care for discharged patients is now common as compared to when PCPs rounded on their own patients. The major benefit for PCPs was that they no longer had to get up in the middle of the night to admit patients; the hospitalist could take care of it. While this may have made their lives feel better in the short run, the disconnection from hospital care and the loss of collegial relationships with hospital specialists makes PCPs more professionally isolated (and unmentored if they are in small practices), and contributes to a loss of a sense of professionalismThis increases stress and burnout.

If any particular medical system raises primary care salaries, they will recruit away some clinicians from other settings, at least temporarily. This happened in the prison health care system in California and the Permanente Medical Group about 20 years ago. The invisible hand of market forces soon intervenes. Other organizations soon raise PCP salaries in response, to stem the loss of staff. With the nationwide and international shortage of primary care, migration of PCPs to greener pastures will contribute to more severe shortages in other communities.

What Can We Do?

Train more PCPs. First and foremost, we need to train more primary care clinicians who will work in primary care. If all the primary care resident physicians in our service area stay in the PHC service area after graduation (unlikely), it would still take ten years just to staff the current open positions. In the past decade, the number of family medicine residencies in our current 14 county region increased from two to seven, with two additional programs in gestation. Nurse practitioner residencies are springing up in many health centers.

Community connections. Secondly, we need to help new primary care clinicians and their families form deep roots in the community and the clinical setting where they practice. The average length of time that new PCPs are staying at their first jobs is now about three to five years in our service area. Whenever a clinician moves to a new setting, the time they invested in learning the systems of care and learning about the lives and medical histories of their patients is lost. Reducing turnover thus instantly has the effect of increasing primary care capacity where the same number of clinicians can care for heavier patient workloads as they get to know their patients better.

A place-based business or career for a physician spouse can build longevity in the community. A friend from residency moved to rural Alaska to practice for what he and his wife thought would be a three-year stint to repay his substantial medical student loans. His wife liked dogs, and this grew into a hobby of racing sleds driven by Alaskan Huskies. They are still in rural Alaska 30 years later. This is an extreme example of a common phenomenon. Most PCPs who stay in one town their whole career do this because of family connections to the community.

Life-long love of primary care. Third, we need to find ways to help our PCPs feel the joy and pleasure of providing primary care. If clinicians only enjoy their lives outside of the workplace, they will want to work less hours, retire early, and/or change jobs more frequently. Ways to build this joy are building friendships with peers, matching new providers with enthusiastic mentors who love primary care, and finding ways to minimize non-value added administrative activities.

Shake it up. Another tactic that works well is to break up the office practice routine. This could be a clinical activity like becoming the primary care “specialist” in the office for opioid treatment or dermatology or seeing patients weekly in a homeless shelter-based clinic site. The change in routine can also be from working with a team to improve quality outcomes, improving internal operational efficiency, or working on some other special project.

Reduce non-value-added activities. Electronic health records are here to stay, but they are a major contributor to clinicians having less enjoyment in the workplace in the past two decades. The PCP office needs to have support staff off-loading computer documentation activities and adapting the EMR to make it ever more efficient to use and less of a drain on clinician morale. Trying to cut costs by reducing support staff and keeping billable providers will improve the bottom line only temporarily before clinicians start leaving because they don’t feel supported or valued.

As clinical leaders we are all devoting ourselves to this challenge using the five tactics listed above and others. Alleviating the PCP shortage is a long-game imperative for the entire health care delivery system, as well as your individual practices. We at PHC appreciate all you are doing!

Looking for Opportunities to Improve

By Robert L. Moore, MD, MPH, MBA, Chief Medical Officer

“Mistakes are a fact of life: It is the response to the error that counts.”

-Nikki Giovanni, poet

Many years ago, a Primary Care Provider (PCP) unintentionally placed an intrauterine device (IUD) into a women who turned out to be pregnant. The pre-procedure protocol was to time the insertion of the IUD to be a few days after completing menstruation, and checking a pregnancy test. In this case, the medical assistant did not conduct the pregnancy test and the provider did not confirm a negative test before the procedure. The patient later discovered she was pregnant and had a miscarriage.

The PCP instituted a number of process changes. They developed a standardized form which listed the pregnancy test result prominently so the medical assistant is reminded to do the test and the provider could readily see it. A check box, requiring the clinician to confirm that the pregnancy test was negative, was added to the progress note template.

A Partnership HealthPlan (PHC) nurse noted this adverse outcome referred it to the PHC quality team for investigation.

Like many of your organizations, PHC has a peer-review process where we review referred cases in which a potential quality issue is identified. PHC doctors, nurses, and pharmacists review patient care while conducting care coordination or utilization management activities, the results of which could generate complaints from our members (your patients).

PHC typically gathers documents (medical records, grievances etc.) to see if the initial concern is confirmed. If the concern for a quality-of-care issue is not resolved by this documentation, we will typically send a letter to the clinician whose care we are concerned about, asking them to review the case and respond to our concerns.

Clinicians may respond in one of these ways.

  1. Respond giving some additional information or context which makes clear that there were no quality-of-care issues after all.
  2. No response, in spite of multiple attempts. We are then forced to make a finding based on the information we have. These clinicians may be afraid of liability or feel like they are too busy to get around to responding; hard to know.
  3. Respond defensively, saying they did everything right, often blaming the patient in some direct or indirect way for the outcome.
  4. Respond reflectively and humbly, noting the complexity of the case but clearly identifying opportunities for them to improve practices, systems of care, or thought processes they will apply for future cases.

We are all human and make mistakes. When clinicians respond with reflection and humility, looking for ways to learn and improve ourselves, we are making ourselves better diagnosticians and clinical decision makers. This approach is a reflection of a “growth mindset”.

On the other hand, if we look around for excuses and others to blame for poor outcomes – unwilling to look in the mirror at how our biases and thought processes played a role in our own poor choices – we are not learning, and we are likely to make similar errors in judgment in the future. This approach is a reflection of a “fixed mindset”.

Psychologist Carol Dweck’s 2006 book Mindset: The New Psychology of Success suggests that our world view and behavior is influenced by these two mindsets. While our mindset is not easily changed, becoming aware of this dichotomy may help some people consciously move themselves from a fixed mindset in the direction of more growth.

When confronted with an complex adverse outcome with many causes, we should take the time and effort to look at how our decisions or systems could have been different, actively looking for opportunities to change systems and cognitive processes to help future patients.

When the PHC peer-review team receives responses indicating a growth mindset –acknowledging errors and explicitly listing changes for better care in the future – we are reassured these clinicians will learn and improve throughout their careers. When the peer-review team receives no response or a response showing no introspection and searching for others to blame, we are more concerned, and more likely to escalate to appropriate oversight organizations.

As medical leaders within your organization, you play a key role in this process. If you have been a medical leader long enough, you have had to confront the issue of clinicians without a growth mindset. To enable you to conduct your own investigations and assess competency concerns, PHC will ensure the medical director overseeing a clinician is aware of potential quality issues. We hope you can mentor and counsel your clinicians, using potential quality investigations as opportunities for them to grow and learn.

One approach to building a culture of growth is the “Just Culture” approach pioneered in patient safety work. Ensure that peer review and mentoring activities are supportive and don’t shame individual providers. This helps model the a learning approach to feedback with a growth mindset.


Health Equity: What it Means for Primary Care

By Robert L. Moore, MD, MPH, MBA, Chief Medical Officer

“Change was indeed possible, one person at a time.”

Mark Bello, Author

The topic of Health Equity is now everywhere: special issues of Health Affairs and JAMA®, legislation, regulatory mandates, etc.

What does it mean for Partnership HealthPlan of California (PHC) and primary care?

The Department of Health Care Services (DHCS) has a nice framework for thinking about Health Equity (from the 2022 Comprehensive Quality Strategy).
It divides Health Equity into three categories of actions:

  1. Measurement: Activities related to accurately measuring disparities and inequities. This includes a planned several-year effort to collect Sexual Orientation/Gender Identity information as part of the Medi-Cal eligibility process (in addition to current gathering of race/ethnicity/language data). Overall quality data and outcomes can then be analyzed for each of the demographic groups for which we have data, to look for differences or disparities. These differences may be found at the county, health plan, or provider level, depending on the individual measure characteristics. Example: the rate of BP control is significantly lower for Black men, resulting in increased incidence of stroke and heart attack.
  2. Interventions: Once disparities or inequities are documented, they can be analyzed to look for drivers or underlying causes. This analysis can then be prioritized into some potential action plans or interventions to close disparities/reduce inequities. There are two major potential causes of confirmed inequities in vulnerable populations:
    1. Bias/Discrimination. This can occur if a health care professional or support staff make treatment decisions or recommendations based on implicit bias or even explicit assumptions about patients’ values and options, based on their race/ethnicity/language or other trait, and withholds valuable care disproportionately to one group compared to another. Example: Post partum nurses in one PHC contracted hospital don not educate new Latina moms about breastfeeding because they assume that Latina mothers prefer to bottle feed. Provider Intervention: Staff education to understand implicit and explicit bias and learn to counteract it. Societal Intervention: Educate each other and particularly our children at a younger age to understand implicit and learn to compensate for it, so that those who choose to work in health care don’t come with biases to be unlearned.
    2. Systemic or Socio/Cultural/Economic Factors. Different levels of income, family support, educational attainment, quality of housing, neighborhood safety may be associated with different demographic groups and be a major driver for unequal health outcomes. Example: Differential rates of obesity, hypertension and diabetes in different demographic groups, which lead to differential morbidity and mortality ratesProvider Intervention: Extra in-reach and out-reach activities, including addressing key social needs to have the health care team overcome the underlying socio-cultural factors and improve health care quality outcomes despite the underlying systemic factors. Societal Intervention: Support federal, state and local policies and interventions which reduce the underlying social and economic factors driving the difference.
  3. Supporting a Culture of Diversity, Equity, and Inclusion (DEI)in the workplace. Strictly speaking, activities to support DEI in the health care workplace are a societal-level intervention to reduce bias and inequities, but the ideals of DEI in a health care organization support that organization’s propensity to the meaningful interventions listed above. Examples: Designating someone in the health care organization’s leadership to spend time and energy focusing on equity; supporting an employee equity committee that reviews HR data, and practices for potential interventions.

PHC will follow DHCS’ lead and direction in using the the National Committee for Quality Assurance (NCQA) Health Equity Framework as a foundation for extending our previous work in population health/health equity to a new national standard. This new standard includes all three activities above: better measurement of disparities, focused and effective interventions to reduce/eliminate disparities, and activities to support DEI among our staff.

We encourage you to work toward improving health equity in these same three realms.

  1. Use data you have in your electronic systems to analyze disparities that you may be able to detect better than PHC, which now uses only county-provided demographic data usually self-declared in the Medi-Cal eligibility form.
  2. When you identify potential disparities either from the data analysis in step 1 or from direct feedback from patient complaints or feedback, evaluate options for how to reduce this disparity.
  3. Discuss any changes your company can make to support staff having a greater understanding of implicit bias, the historic roots of racism and other discrimination, and ways to compensate for this.

Finally, put your organization’s plan in writing and review it periodically at all-staff meetings, to keep the momentum in the right direction.

We all need to do our part to make our health care delivery system more equitable.

For more materials on health and racial equity, see the California Healthcare Foundation, the Center for Health Care Strategies, and Health Begins.

Collaboration to Achieve System Wide Changes: Part III

By Robert L. Moore, MD, MPH, MBA, Chief Medical Officer

“Leadership is the art of getting someone else to do something you want done because he wants to do it.”

– Dwight D. Eisenhower

As we emerge from the constraints of the COVID pandemic, we are spending important and deserved energy on restoring our systems and staffing. At the same time, we are trying our best to respond to a flood of new regulations, programs, and provider types.

Given all these activities and external stresses, we have been less engaged with community partners and advocacy efforts. This impacts our effectiveness in addressing challenges in our communities. Lack of engagement also leads to less robust feedback in the policy development process, which impacts the effectiveness of these policies as they roll out.

There are many levels of collaboration in which clinical leaders can and should participate. October’s newsletter focused on county medical societies, and November’s newsletter focused on community collaboratives. This newsletter focuses on state-wide trade organizations.

State-wide Trade Organizations: These may include specialty organizations like the California Academy of Family Physicians or the California Chapter of the American Academy of Pediatrics.

They may also represent your practice setting like the California Primary Care Association; the California Rural Indian Health Board; the California Rural Health Association; the California Association of Rural Health Clinics; the California Association of Public Hospitals; and the California Hospital Association. Most of these organizations have committees for clinicians and opportunities for clinical leaders to give input. They make better policy recommendations and formulate better strategies when practicing clinicians are involved.

Trade organizations allow clinician leaders to develop a support network of peers, share best practices with each other informally, and give input on current policy issues that are complex or controversial.

Summary: In addition to leading within your organization, clinician leaders need to engage with outside organizations. It is hard to find time to participate optimally in all these areas. One option is to divide up the workload with other budding clinician leaders in your organization. They often enjoy this break from their clinical work, and it makes them feel like they are a part of something larger than their own clinical practice.

Collaboration to Achieve System Wide Changes: Part II

By Robert L. Moore, MD, MPH, MBA, Chief Medical Officer

“Out of clutter, find simplicity. From discord, find harmony.
In the middle of difficulty lies opportunity.”

Albert Einstein

As we emerge from the constraints of the COVID pandemic, we are spending important and deserved energy on restoring our systems and staffing. At the same time, we are trying our best to respond to a flood of new regulations, programs, and provider types.

Given all these activities and external stresses, we have been less engaged with community partners and advocacy efforts. This impacts our effectiveness in addressing challenges in our communities. Lack of engagement also leads to less robust feedback in the policy development process, which impacts the effectiveness of these policies as they roll out.

There are many levels of collaboration in which clinical leaders can and should participate. Last month’s newsletter focused on county medical societies. Next month will cover statewide trade organizations.

This newsletter focuses on local community collaboratives.

Local Community Collaboratives:
In one form or another, each of our counties has one or more health collaborative stakeholder groups. Examples include:
•    Hope Rising, Lake County
•    Health Action 2.0, Sonoma County
•    Shasta Health Assessment and Redesign Collaborative (SHARC)
•    Live Healthy Napa County
•    Vida Del Norte

Health care provider stakeholders meet together to either keep each other up to date on their major initiatives (helpful for preventing surprises and to look for opportunities to work together) or to actively work across organizations on a common problem (like prescription opioid overuse or adverse childhood events). Major stakeholders may include hospitals, larger medical groups, FQHCs, the county medical society, and the county health department. Sometimes additional community-based providers or even consumers are also included. As a community-based health plan, Partnership HealthPlan of California (PHC) tries to participate in all coalitions of which we are aware.

In many collaboratives, all or most of the stakeholders are non-clinicians. These can achieve the first goal above of giving updates and preventing surprises. In a few communities, clinician leaders representing the different health care providers meet monthly or quarterly. This would typically include the CMO of health centers, the county health officer, an officer of the county medical society, and clinical executive leaders for local hospitals. These can be in parallel to the non-clinician collaboratives. Clinician-level sharing is at a different level than non-clinicians, so having these separate cross-provider meetings can bring additional ideas and group interventions on community health problems.

Many county health departments have a requirement for a community stakeholder process to contribute to their county public health priorities. Often these stakeholder meetings are run by consultants, include some sort of survey (which clinical leaders should try to contribute to), and a series of community meetings. While this process is very time consuming, participating in community prioritization exercises can have some impact on county funding priorities, so it is important to hear the voice of community clinician leaders in that process.

What local community collaboratives exist in your county? Is there someone on that collaborative that represents the primary care providers like yours?

If you are unsure what groups are active in your county or want to strategize on boosting effectiveness of current groups, reach out to your regional PHC staff: your Regional Medical Director, Regional Manager, county Provider Relations Representative, or regional QI staff. None of our work is really siloed; we all need to build relationships to tackle the most challenging issues in our communities.

Achieving System Wide Changes: Collaboration is Key

By Robert L. Moore, MD, MPH, MBA, Chief Medical Officer

“I think the reason I’ve been so committed to advocacy is because I see so many people in pain.”

Olly Alexander, British Actor

As we emerge from the constraints of the COVID pandemic, we are spending important and deserved energy on restoring our systems and staffing. At the same time, we are trying our best to respond to a flood of new regulations, programs, and provider types.

Given all these activities and external stresses, we have been less engaged with community partners and advocacy efforts. This impacts our effectiveness in addressing challenges in our communities. Lack of engagement also leads to less robust feedback in the policy development process, which impacts the effectiveness of these policies as they roll out.

There are many levels of collaboration in which clinical leaders can and should participate in. I want to highlight three that are particularly worthwhile: County medical societies (more detail below), local community collaboratives, and statewide trade organizations. The latter two will be the subject of future newsletters.

Organized Medicine/County Medical Societies 
County medical societies can serve several important functions:

  1. Promote the value of professionalism in the field of medicine
  2. Provide a social support network for individual clinicians
  3. Promote policies and programs that support and advocate for public health
  4. Promote policies aimed at improving the health status of the entire population

The California Medical Association (CMA) has a respected and powerful political voice in Sacramento. There are times organized medicine may vote to promote a policy we individually disagree with; however, the process of directing the policy priorities is remarkably democratic. That process depends on participation by local county delegates and regional trustees that take the time to learn about health policy and participating in the process.

Some medical groups in more populous areas have more influence in the CMA because they have all their physicians join their local medical societies. Conversely, medical society membership and delegate representation is low in rural areas often leading to unaccounted physician views on policy issues.

At a minimum, the clinical leaders of a Primary Care Provider (PCP) organization should be active medical society members. If your organization’s budget can accommodate it, covering the cost for any interested physician to join the medical society is a best practice. Having the organization or medical group pay for all physicians to be members will give your organization a bigger voice and is certainly worth considering.

Next month: Engaging in Community Collaboratives.

Diagnostic Inaccuracy in Primary Care: How Much Can We Blame the System?

By Robert L. Moore, MD, MPH, MBA, Chief Medical Officer

“The search for a scapegoat is the easiest of all hunting expeditions.”

-Dwight D. Eisenhower

Case Example: A 45 year old woman sends a secure electronic message to her primary care clinician, asking what she should do for her “heartburn.” Her PCP works a 3 day per week schedule at a chronically understaffed rural health center, and finally gets to the patient’s message at the end of a long day. The PCP does not ask any additional questions, but suggest the patient try OTC famotidine for a few days and call back if the “heartburn” is no better. Two days later, the PCP receives an electronically communicated consultation note from the local hospital, that the patient was admitted to the ICU in cardiogenic shock from a massive myocarial infarction. Glancing through the electronic health record, the clinician notes that the patient’s mother and father both died in their 40s of heart attacks.

This case of diagnostic error is certainly partly due to the PCP not performing an adequate assessment of the patient: not reviewing the electronic health record for background information, not asking additional questions to find out what the patient meant by “heartburn” or asking about red flag symptoms.

However, several system issues also contributed to the diagnostic errors: the overworked clinician, more prone to shortcuts to get through their day; the PCP’s reduced in-office schedule; the promotion of electronic communication to increase access to clinician; the lack of the PCP’s experience with rarer but more serious causes of “heartburn” because an emergency room/hospitalists sees all patients in the hospital.

The Institute of Medicine’s 2015 book “Improving Diagnosis in Health Care,” categorizes such system issues that contribute to diagnostic inaccuracy into five groups:

  1. Organizational factors, such as schedules, staffing models, payer mix, and leadership.
  2. Physical environment, including clinical ambiance, proximity to co-workers.
  3. Tasks, which may compete for the attention of the clinician, like electronic medication refills, or which may not occur when they need to, like following up abnormal lab results.
  4. Technologies and tools, such as the configuration of the electronic health record, and the use of electronic modalities for virtual care.
  5. Diagnostic team members, including who is included on the team (adding a triage nurse for example), how communication occurs with the team, and the sense of responsibility team members feel to collaborate (in the case above, none of the physicians caring for the patient in the hospital called the PCP about the admission).

These same five factors contribute to clinician burnout and to health care inequities, so we as clinician leaders have a triple responsibility to spend our discretionary time working to steadily optimize this work system, the external environment that our clinicians work within. This is true for all clinician leaders, whether we work in primary care, specialty care, institutional settings like hospitals or skilled nursing facilities, at a health plan, or a government regulator.

At the same time, individual clinicians must continually strive to improve their own diagnostic processes, being aware of their own cognitive biases and short cuts, and reinforcing a sense of professional responsibility to achieve diagnostic excellence even with the shortcomings of the system we work in.

Knowledge Management: Don’t Reinvent the Wheel

By Robert L. Moore, MD, MPH, MBA, Chief Medical Officer

“Those who do not remember the past are condemned to repeat it.”

-George Santayana, Philosopher

One casualty of the Covid pandemic is institutional memory. There has been a big turnover in staff, management, and leadership positions in many health centers, offices and hospitals in the Partnership HealthPlan of California (PHC) service area.

Example A: A new clinic manager tries to increase provider access by making provider appointments shorter. This results in more patients being seen per day. However, over time the provider feels burnt out and quits to join another clinic, resulting in even longer waits for patients.

While new staff can bring new ideas, they often don’t have the benefit of knowing what has been tried before, and how well it worked. Often problems are approached as if no one has ever tried to fix them before. As a result, they repeat previous ideas and the lessons learned are no different from the first try.
The concept of Knowledge Management includes three components:

  1. Using existing knowledge to inform pilots and interventions, instead of starting from scratch.
  2. Generating new knowledge by processing and analyzing information and existing knowledge in new ways.
  3. Capturing new knowledge in a way that can be searched and accessed easily in the future.

The first component can be summarized succinctly: “Don’t reinvent the wheel.”

When mentoring new staff with shiny new ideas, show them how to find out what is known about that idea. This includes internet searches, talking to staff and outside experts with more institutional wisdom, and looking though company files about past projects.

Background research also includes subjecting articles and papers to scrutiny, to identify bias and statistical fallacies that could reinterpret their findings.

Example B: A new physician decides to implement a scribe system to improve provider efficiency. They consult articles and talks to a colleague at a neighboring health center to look for additional information. This results in phone conversations with clinical leaders at several other health centers, where they learn that clinic productivity does not tend to improve with the use of scribes, but that chart accuracy, clinician happiness, and quality scores improve when a scribe program is implemented well.

Sir Isaac Newton (a co-discoverer of calculus and father of the field of physics) understood this principle of building on prior knowledge. “If I have seen further, it is by standing on the shoulders of giants.”

As leaders we must model this framework, setting an expectation for the new staff that join our organizations, that they will search out existing knowledge and apply it to the problems they confront.