Blog Home


Creating Value-Based Incentives For Primary Care

June 2nd, 2011

In a remarkable recent interview, Donald Berwick MD, Administrator of the Centers for Medicare and Medicaid Services (CMS), eloquently described his vision of value-based health care.

Paying for value is an incentive…The underlying idea of improvement is that American health care, historically built in fragments, often cannot achieve for patients what it really wants to achieve…Health delivery system reform refers to really reconfiguring care into much more seamless coordinated-care operations so that people, especially those with chronic illnesses, experience continuity of care over time and space.

So when patients come home from the hospital, there is a smooth handoff, and all the necessary information follows them. When they are seeing a specialist, that specialist is coordinating care with their primary care doctor.

This description probably resonates with most health care professionals as a better approach than the current paradigm’s fragmentation and lack of continuity of care. But as with many things in health care, it won’t be easy getting to a value-based health care approach in Medicare and Medicaid. Despite wide acknowledgement that fee-for-service perpetuates our health system’s most undesirable characteristics, the mainstream of American health care seems stuck. One wonders whether CMS can rise above the special interest lobbying, get beyond the interminable pilots and decisively act on payment reform with the conviction required to help save health care from itself.

Still, the idea of value-based reimbursement begs questions. What payment methodology will incentivize the best quality and most efficient care? What path can take us there?

Primary care should be at the heart of this discussion. While much of specialty care has been overvalued over recent decades, the undervaluing of primary care has weakened its moderating influence over downstream services, with dramatic cost growth that now threatens all health care and the nation.

Let’s recount what we know about primary care and its impact on specialty services.

  1. More primary care in a market lowers overall health care cost.
  2. Primary care physicians (PCPs) who aren’t rushed with patients tend to develop stronger patient relationships and handle problems immediately, making fewer (unnecessary) specialty referrals. By contrast, volume-based primary care reimbursements that incentivize shorter established office visits increase  cost.
  3. Fee-for-service reimbursement encourages more services, independent of appropriateness, and so is antithetical to medical homes that focus on ensuring appropriate care throughout the continuum.
  4. Over the past 20 years, a specialist-dominated political process has driven an enormous disparity between primary and specialty care reimbursement. Low primary care reimbursement has resulted in a primary care labor crisis.
  5. When referrals are made, an open line of communication creates greater specialist accountability to the PCP, moderating unnecessary services. This approach appreciates PCPs as full-continuum patient advocates and guides rather than as “gatekeepers.”

The lessons above constitute a basis for a revised approach to primary care payment. The goal here should not be to simply pay primary care physicians more for the same work, but to change medical management in a way that increases efficiency throughout the continuum. Here are guidelines that should be reflected in any new primary care payment scheme:

  1. Separate Valuation Mechanisms. The work of primary and specialty care physicians can be very different and must be evaluated differently, through separate mechanisms.
  2. Valuation Absent Financial Conflict. Specialty physicians have a financial interest in how primary care physicians practice, so should not dominate the determination of primary care reimbursement.
  3. Financial Parity. To rebuild our primary care workforce, generalist physician income should be on par with average specialist income.
  4. Incentivize Appropriateness. Payment should incentivize appropriate care and, unlike straight fee-for-service or capitation, avoid encouraging the delivery of unnecessary care or the denial of necessary care.
  5. Incentivize Teamwork. Payment should reward population-level health improvements that can only be achieved through more primary-specialty collaboration and accountability. Of course, this assumes that physicians have better access to comprehensive patient information.
  6. A Focus on Value. Payment should be based not only on a service’s “inputs” – both inside and outside the patient encounter – required to accomplish care, but on its value. Care valuation should include patients, purchasers and health economists as well as clinical practitioners.
  7. Encourage Investments for Better Performance. Payment should encourage investment in technologies and programs demonstrated to improve quality or safety at lower cost.

Payment that reflects these elements would liberate primary care, organically reducing unnecessary specialty care, and saving money without reducing payment for individual specialty procedures.

Fixing primary care reimbursement is a critical first step toward healing primary care and the larger American health system. For this reason, CMS should consider these design criteria within the frame of Dr. Berwick’s vision, and move with all speed to change the way it pays for primary care.

Email This Post Email This Post Print This Post Print This Post

 to the #1 source of health policy research.

3 Trackbacks for “Creating Value-Based Incentives For Primary Care”

  1. Will the Bubble Burst? | Brian Klepper, PhD
    August 20th, 2012 at 5:56 am
  2. Will the Bubble Burst? - The Doctor Weighs In | The Doctor Weighs In
    August 19th, 2012 at 11:12 am
  3. Care And Cost
    June 3rd, 2011 at 4:26 am

3 Responses to “Creating Value-Based Incentives For Primary Care”

  1. james rickertmd Says:

    Financial parity between specialists and primary care docs would solve many of the problems in our current system. First, higher payments would give PCPs more time to spend with their patients to organize their care and treat their illnesses. Second, if docs were paid equally for evaluation and management (office visits) or procedures, there would be much less incentive for inappropriate treatment by specialists. I say all of this as a procedural specialist who has seen the corrosive effect of the discrepancy between payments for procedures and payments for conservative management of disease on our health system in general and on individual physician behavior specifically.
    Finally, currently specialty practices typically are much more lucrative than primary care practices and, therefore, specialists could easily attempt to purchase primary care medical homes to ensure inappropriate referrals. As you note, this should note be allowed.

    Learn more about the need for physician payment reform @
    James Rickert, MD
    The Society for Patient Centered Orthopedics

  2. Raymond Simkus Says:

    Working as a primary care physician for over 30 years and having been involved in primary care reform in British Columbia I agree with the comments and opinions in this article. Twelve years ago as part of a primary care reform program my office switched from fee for service payments to population based or a capitated system. I would like to point out that not all capitation systems are the same. In the system here a lot of checks and balances were built in that obviate the usual problems attributed to capitation. The system here provides for increased payments related to the disease load of the patient based on the ACG scoring developed at Johns Hopkins. This provided an incentive to seek out disease. Since there was no payment for office visits there was an incentive to reduce office visits but there was counterbalance of negation if patients were seen outside the practice and presumably if you neglected the patient and they got sicker it would take more work or more visits to get them better. The actual amount paid depended on what the fee for service payments across the province were for each specific ACG category. These two points provide incentives to optimize the treatment of patients, particularly for conditions that require multiple visits, that are chronic and that involve many patients and where we could potentially provide better care than what was delivered to the average patient. This led us to focus on the usual chronic conditions that have attracted a lot of attention and have been reported to have many patients that are under treated. It also provides an incentive to have patients get more involved in their own management.

    Our impression is that the number of office visits have decreased and the number of hospital admissions has also decreased. The physicians love it and the patients also love it because they are not called back for what they feel are trivial reasons.

  3. John Welton Says:

    Drs. Klepper and Kibbe,

    From purely a cost standpoint, using the AAFP average salary of $145,000 (36 yr old, 7 yr or less practice experience, see ) for a Family Practice Physician (FPP), and making an assumption that each FPP sees 15 patients, 5 days a week for 50 weeks, 3750 patients/yr., the marginal cost of treating 1 patient is $38.67 (plus overhead). If FPP salary is doubled to be more consistent with specialist MD salary (as you argue), then marginal costs are also doubled to $77.33/pt using the same assumptions. The marginal utility of using family practice physicians compared to specialists decreases with salaries equivalent to other specialty MDs, weakening your argument. On the other hand, if more nurse practitioners are allowed to enter the primary care market place, assuming an average annual salary of $85,000 and using the same assumptions above, the marginal cost/patient is $22.67 (plus overhead). Hence the marginal utility of increasing primary care improves by adding more nurse practitioners (NP).

    I am not arguing to substitute NPs for FPPs, however a more comprehensive approach to redesigning primary care to better serve our patients and improve overall value is more consistent with a team approach combining the best of medicine and nursing. Reconfiguring your arguments above to include a nursing perspective:

    1. Separate Valuation Mechanisms : Valuation of primary care should include both physician and nurse inputs evaluated on services provided to patients and the overall outcomes/goals of care. FPPs and NPs have a different educational, practice, and licensing structures and these should be considered in valuating physicians and nurses individually as well as the combined efforts and contributions of each. Other primary care practitioners such as PAs should also be considering in this framework..

    2. Valuation Absent Financial Conflict: Physicians and nurses should work collaboratively and cooperatively and physicians should not dominate reimbursement or exclude nurses or other primary care practitioners from the reimbursement system. Nurses and other primary care providers should receive reimbursement relative to the services they provide on parity with similar services provided by physicians. All physicians have a financial interest in primary care and should not dominate or influence primary care reimbursement or practice for nurses and other primary care providers.

    3. Financial Parity: Salaries for nurses and other primary care providers should be equivalent to physicians providing similar primary care services. Salaries should be based on overall value, e.g. the contribution of each primary care provider to reach and sustain high levels of health and disease prevention.

    4. Incentivize Appropriateness: Payment should incentivize optimum care and consider the entirety of the primary care effort and interaction of physicians, nurses, and other primary care providers. Payers should pay on parity for alternative primary care delivery models such as nurse led/run clinics or medical homes to the degree they provide equivalent services and outcomes. Sustained benefits of nursing centric models of care should be compared to other primary care delivery models and payment methods should reward those delivery models that provide superior care, e.g. best outcomes at best price.

    5. Incentivize Teamwork : Payment should reward population-level health improvements that are part of a shared accountability framework, that is both physicians and nurses and other primary care providers are each individually and collectively accountable for achieving high quality health outcomes. All primary care providers should have access to the latest clinical guidelines for care and the necessary information technology to improve workflow and information flow to interact with other providers and hospitals.

    6. A Focus on Value : Payment should be based on value. All primary care providers are part of the value chain and as such each should be recognized for their contribution in the payment system. Analysis and evaluation of performance in primary care should be inclusive and focus on high quality and safety based on inputs from all primary care providers. Patient, community, and purchaser input should be part of a comprehensive evaluation of primary care practice models.

    7. Encourage Investments for Better Performance : Payment should encourage the integration of primary health care, the identification of high performing health care teams, and the achievement of optimum outcomes of care as well as the technology needed to support this effort. As such, value-based purchasing models should identify and incentivize collaboration and collegiality amongst all primary care professionals, in particular physicians and nurses.

    John Welton, PhD, RN, Lakeland, FL

Leave a Reply

Comment moderation is in use. Please do not submit your comment twice -- it will appear shortly.

Authors: Click here to submit a post.