Can Startups Save Primary Care?


Today, primary care is considered the bee’s knees of value-based care delivery. Instead of being viewed as the punter of the football team, the primary care physician (PCP) has become the quarterback of the patient’s care team, calling plays for both clinical and social services. The entire concept of the accountable care organization (ACO) or patient-centered medical home (PCMH) crumbles without financially- and clinically-aligned PCPs. This sea change has resulted in rapid employment or alignment to health systems, as well as a surge in venture capital being invested into the primary care space.

Before we get too far in the weeds, let’s first begin with the definition of primary care. The American Academy of Family Physicians (AAFP) defines a primary care physician as a specialist typically trained in Family Medicine, Internal Medicine, or Pediatrics. Some women do use their OB/GYN as their PCP, but these specialists are not traditionally considered PCPs. Now if you’ve gone to your local PCP and noticed that your care provider is not wearing a white coat with the “MD” or “DO” credentials, you are either receiving treatment from a hipster physician, nurse practitioner (NP), or physician assistant (PA). Two of the three professionals are trained in family medicine and can provide primary care services under the responsibility of an associated PCP. At least one of the three has a beard.

The crazy thing is, despite the industries heightened focus on the importance of PCPs, we’re still expecting a shortage of primary care providers. In April 2019, the Association of American Medical Colleges (AAMC) released a report estimating a shortage of between 21,100 and 55,200 PCPs by 2032. Given we just passed 2020, this not that far off. The primary reason for the shortage is the growing and aging population. Thanks mom and dad. Digging into the numbers will really knock your socks off, with the U.S. Census estimating that individuals over the age of 65 will increase 48% over that same time period. Like a double-edged sword, the issue is not just on the patient demand side though. One-third of all currently active doctors will be older than 65 in the next decade and could begin to retire. Many of these individuals are independent PCPs who have resisted employment by large health systems.

Now the easiest solution would be to wave a magic wand and dramatically increase the supply of medical students selecting primary care versus other specialties. However, in the absence of any Hogwarts-trained healthcare enthusiasts, we have to face the realities of today’s medical school situation. 75% of medical school students in the class of 2018 graduated with student debt, with the average loan debt of $196,520. With that loan balance, you’d owe approximately $2,212 a month on a standard, 10-year federal repayment plan. If you compare that with the earning potential, pediatrics and family medicine are consistently among the lowest paid specialties. According to Medscape, in 2019 PCPs earned an average of $237,000, while specialists earned an average of $341,000. That is a big difference. This all despite the fact that according to a Merritt Hawkins report estimating PCPs generated $2.1 million for their affiliated hospitals in the previous 12 months. This referral value to the hospital even exceeded Otolaryngology ($1.9 million), despite the fact the average annual physician compensation for an Otolaryngologist is $471,000.

The other important characteristic that healthcare economists and researchers have closely monitored is increase in hospital employment and alignment of PCPs versus physicians who own their own independent practices. The fear being that employed PCPs have the potential to refer testing, therapies, and services back to the mothership hospital, as opposed to independent specialists, labs, ambulatory surgical centers, or imaging centers. These hospitals charge considerably higher professional and facility fees, particularly for commercially insured patients. The crazy part of value-based care is that some of the clinically integrated network (CIN) provisions and waivers associated with primary care allow hospitals to align networks of independent PCPs and ensure they receive much higher negotiated rates. Aligning PCP networks to highly motivated and sophisticated health systems who are actively involved in significant downside risk contracts has clear benefits, but the potential for inefficiency and shoring up referral patterns does exist.

Now back to the world of healthcare startups as they relate to primary care. I’ve said it before and I’ll say it again, the successful One Medical (NASDAQ: ONEM) IPO was the single best thing for primary care startups. Yes, they focus on a particular clientele (commercially insured in urban markets). Yes, they are charging a $199 annual fee for access to their care that many Americans cannot afford. Yes, they will likely grow revenue through higher volume and negotiated reimbursement contracts by partnering with health systems, referred to as health networks. However, 3 weeks after the IPO, the company maintains a market valuation cap of $2.8 billion with nine months of net revenue equaling $199 million and $34 million is losses during the same time period. That is impressive and should be encouraging for current PCP startups.

There are a ton of other primary care focused startups and companies that should not be overshadowed. Each takes on a slightly different approach, whether they focus on a specific population (e.g., Medicare Advantage), actually employ physicians themselves, or serve as administratively- and clinically-aligned vendors for networks of PCPs. I strongly believe in the value of these organizations using technology-driven communication, remote monitoring, home care, and intensive wrap-around care management services for complex populations to offer a new model of care.

If there was ever a downside for creating the primary care-led revolution towards value-based care, I believe it would be the production of charlatans claiming to “primary care consultants” or point-solutions that only affect a sliver of the problem. I see lots of claims related to SAAS startups who utilize an AI-based, machine learning analytics program that spits out a list of high-risk patients. After many years trying to squeeze clinical and financial value out of total-cost-of-care models, that is no longer enough. In my opinion, PCPs should seek partners who can not only provide technology, but also have clinical resources and are willing to stand by their performance in terms of compensation. No guaranteed PMPMs if performance doesn’t add measurable and definitive value, unless the organization is willing to take downside risk exposure. They also need to help your PCP organization understand how to set the rules of the game in your advantage. Otherwise, you will never win regardless of performance (e.g., trend rates, minimum savings thresholds, rebasing, shared savings, etc.).

Now I’m not claiming I know everything, but my experience was borne out of helping health systems and physician groups across the country manage total-cost-of-care contracts in Medicare fee-for-service (FFS), Medicare Advantage, Commercial, and Medicaid managed care while at Evolent Health. For those unaware, Evolent Care Partners is a solution focused on enabling independent PCPs with the capital and resources needed to participate and succeed within two-sided contracts. In addition to Evolent Health, there are a bunch of other primary care startups that I appreciate. They did not pay me (although I should’ve asked before), but here are a few startups that I would research before thinking about primary care in a value-based care world.

  • One Medical: Provider for Commercial
  • Iora Health: Provider for Medicare
  • Oak Street Health: Provider for Medicare
  • ChenMed: Provider for Medicare
  • Privia Health: Population health management partner for primary care
  • VillageMD: Population health management partner for primary care / provider
  • Aledade: Population health management partner for independent primary care

At the end of the day, primary care still receives a pitiful amount of the total spend in healthcare. The best estimates believe only 5-7% of healthcare spending devoted to primary care. In a RAND Corporation study, researchers predicted 2.12-4.88% of total Medicare fee-for-service medical and prescription drug spending. However, the power of referral, care management, and addressing the social determinants of health (e.g., housing, food, transportation, etc.) holds the promise of a better tomorrow. I am hopeful that the trends over the past few years will continue and new startups will be developed that further innovate on the $260 billion primary care market in the U.S..

Andy Mychkovsky is the creator of Healthcare Pizza, where this article first appeared.

4 replies »

  1. No.

    Start ups will not save primary care by trying to reinvent a new delivery system or returning to the “golden age of medicine”, as depicted on TV in the 20th century.

    Family physicians missed their chance, at the beginning of the 21st Century, when they were offered the opportunity to thread their established medical practices through a nationwide network of retail pharmacies, thus giving a world of clinical and business opportunities to PAs and NPs.

    Tech-savvy entrepreneurs, downstream medical specialists and staffing companies by-passed the line by using telemedicine to “cap the wellhead,” without depending upon primary care referrals.

    So, why is it that American medicine and medical entrepreneurs haven’t applied the lessons of conversion franchising and international co-branding?

    Ron Hammerle
    Tampa, Florida

  2. Why bury the lead? Why end this piece with a minor mention of the financial design when this is by far the major reason for shortages in generalists, in general specialists, in rural settings, in underserved settings, and across 2621 counties lowest in health care workforce? The shortages are specific to the practices and populations and places least valued that are forced to deal with the worst public and private insurance plans – for providers and for patients. Lower payments, lower collection rates, increasing usual costs of delivering care, and relatively higher costs to adjust to metrics, measurements, and micromanagement are killing primary care where most needed.

    The financial design does not even consider the usual disruptions that hurt small and medium size practices most – changes in billing, key personnel, ownership, location, or EHR (Mold Annals of FM). It rewards procedural, technical, and subspecialized and destroys cognitive, basic, office, most needed, most prevalent services across generalists and general specialists. This has to end. Some sacrifices will need to be made by those doing well – who often prevent true reform.

    The declines in primary care visits, particularly for the elderly and in lower income areas in MEPS data attest to the financial design. They are a smoking gun pointing to CMS and state plans, or lack thereof.

    Why believe all of the propaganda from MD DO NP and PA leaders about training more or special training to address shortages? The AAMC continues to promote more GME as a solution despite reports and evidence to the contrary. Medical education leaders helped to reign in diploma mills 100 years ago but now we have four sources increasing annual graduates at 6 to 12 times the annual population growth rate now slowed to 0.6%. This massive expansion alone attests to the failure of training to be able to resolve shortages.

    We learned this training dogma in medical school and believed it – for too long. Our associations and meetings and literature promote this. The data indicates that expansions of US MD graduates are proceeding but there is a decline in primary care result. US DO graduate primary care result has been cut in half with each doubling of annual graduates since the 1960s (60% to 35% to 18%) until the matches were unified and then the DO primary care levels dropped even more. The financial design is what shapes career choices, salaries, benefits, numbers of delivery team members, and abilities of the team members. The case can be made that the financial design is a primary influence on burnout, lower productivity, higher turnover frequency, and higher turnover cost in addition to more spent on costly locums and brokers.

    There is absolutely no way to higher functioning or patient centered primary care because the designs shape fewer and lesser team members – not more and better. Value based care is a cruel joke with higher costs of delivery for minimal if any changes in outcomes (because outcomes are about the patient and population, not the practice or process).

    I cannot fathom why AAFP still supports performance based payments with members serving populations most likely to have lesser outcomes. How can you claim to understand and support social determinants and not understand that SDOH is incompatible with value based design? Outcomes are relatively fixed in place. Do you really believe that a few minutes a year – a tiny fraction of the life influences impacting a person – is enough to make a major change in outcomes?

    Who can support 15% of practice revenue stolen by innovation and regulation and another 15% to cover the costs of turnover of a primary care physician ($300,000 per loss, loss each 3 years)?

    How can practices survive with zero adjustment for the increased costs of delivering care? Primary care is about people delivering care to people. Punish the practice budget and you punish the people and their care and caring.

    Do not feel bad about clinging to training dogma. I was a slow learner, too. Rural practice taught me this lesson of the financial design closing my practice, other practices where most needed, and rural hospitals in the 1980s. But I was on a mission to fix rural health. I went on to promote rural medical education schools and pipelines in leadership roles state, national, and international. But I was ignoring the obvious. Yes there is value in specific training for a specific location, but no there is no sustained improvement in basic health access.

    Watching the county maps over 15 years in Nebraska 1992 t0 2008, I saw no change in the primary care delivery capacity. The names and the initials and the types changed, but not what mattered. Why do all of the tracking going on for 20 years and not see the lack of change?

    Schools and programs can look good in measurements involving proportions or ratios for target locations, but workforce remains insufficient. This is a rearrangement of the deck chairs effect. Fixed finances with more from one source results in less from other sources. UNMC graduates choosing family medicine residency at the peak of their pipeline were 12 times more likely to be found instate in one of the 73 counties lowest in health care workforce (14 without any, 6 with average to higher concentrations). A pipeline starting at age 14 was working, but the primary care remained short. KU graduates had much the same impact when choosing family medicine and Kansas has the same deficits by financial design.

    Only 200 to 250 billion in annual spending for primary care has long been a pittance. CMS and designers have made this worse with stagnant revenue despite worsening usual costs of delivering care plus the added innovative, regulatory, metric, measurement, and micromanagement costs.

    Whatever happened to health care designs that focused on support of the team members that deliver the care. What happened to the original Medicare and Medicaid designs that one time only added more billions to these counties and populations for their care? Look back and many of the claims made about improving primary care levels or distribution (FM departments, FM interest groups, FM or primary care schools, Dual Medical Schools doing research and primary care both, NHSC, scholarships and loan repayments) worked mainly during the 1965 to 1978 years of substantial injections of health care dollars into the counties of need – the counties that have long had concentrations of lower income, elderly, disabled, and disadvantaged populations.

    Add to this the millions more people now being added to these counties as they can no longer afford the cost of housing and living in the higher concentration counties that enjoy the property value increases and increases in rentals that most Americans cannot afford.

    Yes, the designers ignore the obvious. Stagnant to declining health care workforce where the American population is increasing fastest. The 2621 counties lowest in health care workforce have been increasing fastest for decades and should reach 50% of the US population by the 2050s. This could accelerate as Americans get poorer as they age and as health care and housing costs go up.

    The designers have created this monster. They continue to destroy Basic Health Access. It appears that this may well be due to incredibly poor understanding with regard to the majority of the American people most behind by health, education, and economic designs.

    There is not a rural crisis or an underserved crisis – there has always been a disparity. Nothing is new except worse designs.

    True reform is possible only if those least valued are at least moved to equity. Support for the basic human infrastructure nurses, teachers, primary care and public health team members, social workers, and public servants is required – and it has to be over generations of time.

    Designing away more billions of health care, education, and social support dollars removes 3 out of the 5 most important sources of jobs and social determinants. Bad show designers. Bad show both parties. Bad show CMS. Bad show MD DO NP and PA leaders. Each in your own way have managed to make the situation much worse.

    And if you reflect on stagnant and insufficient primary care revenue, increasing costs of delivery, high turnover, and massive expansions of graduates in four sources – you have the formula for the least experienced primary care workforce in the history of the United States.

    The expansions have predominantly increased the non-primary care workforce and have helped to generation much higher health care costs and even greater concentrations of health care workforce, dollars, economics, and social determinants where they are already concentrated.

  3. “Those who pay us need to listen.”

    As do those who claim to represent us. The AAFP et al have bought into one disastrously bad, Rube Goldbergian payment scheme after another.

    That venture capitalists now think they can squeeze a few bucks out of poor, dying primary care and then throw away the rind . . .

  4. Primary care could be and so much but insurers don’t really reimburse us for overseeing and coordinating care anywhere near what they reimburse us for cranking out visits. And, repeating myself here, Medicare is confused, thinking the best use of physicians is to execute public health protocols That nurses or even medical assistance could easily do. What we need is not more middlemen but payment reform. We primary care docs know what we should be doing. Those who pay us need to listen.