Jonathan W. Pearce

Capitation once again seems on track to displace fee-for-service as the industry's predominant payment model. Unlike in the 1990s, the casualties may be the providers that cannot adapt, not the payment methodology.

At a Glance  

  • To succeed under population-based health care, organizations need to understand thoroughly how this approach differs from traditional fee-for- service health care.
  • To manage care under capitation, the contracting organization should have a population of sufficient size and a clear means of assigning patients to that population.
  • To assess performance, the organization requires metrics that view performance in terms of per member per month, while avoiding common pitfalls of misapplying such metrics.

With the increasing participation of healthcare providers in accountable care organizations (ACOs) under the Medicare ACO demonstration, the trend toward population-based health care appears to have gathered significant momentum. And with this trend, although the specific terms used may differ, the healthcare industry is again adopting capitation as a primary method of payment. Moreover, this trend applies not only to Medicare payment but also to commercial payment, as private insurers follow Medicare's lead. For hospitals and most other providers (with the exception of some primary care physicians who have been capitated for years), this trend will require a fundamental change in organizational business models and attention to many new factors, including financial incentives, terminology, metrics, and management paradigms.

Because the move from fee-for-service to population-based health care will be gradual, these providers will need to deal with conflicting incentives throughout the transition period, and probably indefinitely. Failure to understand and manage these conflicting incentives was a primary cause of the failures of capitation during the 1990s. In the present healthcare climate, however, providers that cannot adapt-rather than the payment methodology-may end up being the casualties.

Healthcare finance leaders whose organizations are now evaluating options such as ACOs and capitated contracts-particularly if they are accustomed to working primarily in a fee-for-service payment environment-require a deep understanding of the new metrics, accounting processes, financial dynamics, and incentives that characterize population-based health care. This need is particularly pressing if the organization is in a state where provider capitation is being considered to alleviate financial shortfalls in the Medicaid system. In general, given the overall strong trend toward population-based health care, all healthcare finance leaders should have a framework for financial management under a population-based health scenario.

Capitation Requisites

Under a population-based healthcare model, a specific patient population is assigned in some way to an entity that functions as the contracting organization, which will receive capitated payment for the services delivered to the population's members. (See the sidebar below for a discussion of what constitutes a member and how these individuals are assigned to a specific population under population-based health care.)

It also is essential that the contracting organization be capable of functioning under the capitated payment methodology, which means there must be some way to associate the specific population with the entity, and only that entity, for specific types of service.

Capitation works well only when it involves a single capitated entity providing a specific type of service to patients. Primary care physicians are capitated only for primary care services, for example. It would not work, for example, for two physical therapy providers to be capitated for the same patient population, because patients could choose to receive services from only one of the providers, which would leave the more-often-selected provider undercompensated and the other overcompensated.

For this reason, capitation poses problems with certain provider types-for example, in those instances where members typically expect to have a choice of provider. A hospital could be capitated for certain services, for example, only if all members were required to use that hospital for all the covered hospital services. Similarly, an orthopedics group could be capitated only if all members were required to receive all orthopedic services from the group. Because such limited networks for these types of providers are not common, these types of capitation arrangements have rarely been implemented.a  

Typically, therefore, organizations delivering population-based health services and receiving capitation payments are organized to include a wide range of providers such that the members have the freedom to choose providers as allowed by their type of insurance. The Medicare ACO "capitation" amount will apparently include all Part A and Part B costs, and commercial capitation contracts frequently include prescription drug costs as well as medical costs.

In general, under the population-based model, the contracting organization receives the capitation payment and must compensate the providers who are part of the organization. It should be noted that working out these compensation arrangements is often the most complex part of forming contracting organizations. (Read a discussion of some of the challenges that are likely to be encountered in such an effort.)

At first glance, Medicare ACOs may not look like capitated entities, because their payment is referred to as "shared savings" and there are complex rules regarding the distribution of those savings. Fundamentally, though, the ACO demonstration is a capitation system: Members of a participating ACO are inherited from the ACO's primary care physicians, and its capitation payment is determined from historical cost of those members. Despite some differences in details involving the shared savings model, quality payments, and other factors, the fundamental paradigms, accounting treatments, metrics, and incentives of capitated systems also apply to Medicare ACOs.

Financial Statements

The financial statements for providers-both physicians and larger healthcare entities-under population-based health care and the capitated contracting organizations, such as ACOs, are interrelated, as illustrated in the exhibit below. These relationships are critical to understanding how the capitation arrangements affect providers that are part of the contracting organization.

Exhibit 1


Revenue. Healthcare providers commonly derive their revenue from fee-for-service models based on services provided to patients-i.e., revenue is based on numbers of services delivered to patients. By contrast, revenue under a population-based model is based on the size of the covered population, without regard for the services provided, as well as on the payment rate per member per month (PMPM).

Members can be associated with a capitated entity in several different ways. As noted in the sidebar below, addressing this process requires either direct patient selection or inheritance or attribution of a population from another provider.

The PMPM payment rate may be determined in several ways. Under many managed care commercial contracts in the past, the contracting organization was paid a percentage of the actual insurance premium paid by each member. The problem with this approach is that because the premium is under the control of the insurance company, the company incurs no negative effects from lowering insurance premiums to gain market share. The Medicare ACO payment methodology is far more complex and deals with the costs incurred (i.e., the payments made to providers) throughout the ACO demonstration period as compared with benchmark costs established before that period. The payment methodology also includes payments for meeting certain quality targets. (An executive summary and section highlights of the Medicare ACO final rule.)

Again, revenue from population-based health care is distinct from fee-for-service revenue in that it is not based on the healthcare services provided to the members. Therefore, with respect to revenues (but not expenses), the services provided to members are irrelevant to the contracting organization, because the ACO will receive the same payment regardless of the services provided.

Expenses. Understanding expenses for a contracting organization requires a conscious effort to separate that organization from the providers that constitute it. This separation is particularly important when the contracting organization is itself a provider, such as in a hospital-sponsored physician-hospital organization (PHO). The distinction is important because payments (i.e., revenue) to providers are an expense to the contracting organization. And because most providers are paid on a fee-for-service basis, utilization of their services creates a cost to the contracting organization.

This is the most critical paradigm shift for hospital finance managers. Under fee-for-service payment, increases in utilization are good; they generally increase hospital revenue. But under capitated payment, those increases in hospital revenue are generally countered by corresponding increases in expenses to the contracting organization. This relationship requires a fundamental management change that will pervade the entire hospital, from the finance and management staffs through the clinical staff and ultimately to the board. It was the failure of these constituencies to promptly understand this paradigm that led to the failure of many PHOs during the managed care era of the 1990s.

Assets and liabilities. Just as revenue for the provider is an expense to the contracting organization, a similar transformation occurs on the balance sheet between these entities under population-based health care.

Simply put, accounts receivable are an asset on the balance sheet. For providers, accounts receivable represent services that have been provided but not yet paid. Because payments to providers constitute an expense to the contracting organization, however, a delay in accounting for those expenses (because the providers have not yet been paid for the respective services provided) creates a liability to that organization. This liability relates to incurred-but-not-reported (IBNR) claims for services that have been performed but have not yet been expensed because their amounts are not yet known.

In a contract with a commercial payer, claims are generally reported back to the contracting organization on a monthly basis so that the contracting organization can properly account for its medical expenses. Because of the payment lag, however, the contracting organization must also record an IBNR liability for expenses that have been incurred (because services have been rendered by a provider) but that have not yet shown up in the claims data. Although proper estimation of IBNR requires actuarial assistance, finance managers should, at the very least, understand the critical importance of accurate estimation. Many contracting organizations during the managed care era believed they were profitable until the "run-out" claims showed up months later and exceeded the IBNR liability that had been accrued.


Because population health is based on members rather than patients, it will involve entirely new metrics with denominators that differ from typical provider metrics. Although the typical "per patient day" and "per admission" metrics will still have their place in hospital management, the primary metrics for a contracting organization are denominated by member months.

Member months are computed by counting the number of members assigned to the contracting organization in each month of the metric. And the number of member months in the year is computed by summing the number of members in each of the 12 months. Thus, a critical metric for a contracting organization is "amount paid PMPM," which refers to total payments to all providers during a specific period (which need not be a month), divided by the number of member months in that period. This amount can be stratified by different cost categories-for example as "inpatient cost PMPM."

Utilization measures are also denominated by member months; however, the more common metric is "per 1,000 members per year" often shortened to "per 1,000." The denominator is adjusted in this way to avoid metrics with very small values. Member months are converted to a per 1,000 metric by dividing them by 12,000. Many typical population-based inpatient metrics are expressed on the per 1,000 basis (e.g., "patient days per 1,000" and "admissions per 1,000").

These metrics measure the "incidence" of utilization of services-the proportion of the population that utilizes the various types of healthcare services. In a fee-for-service healthcare system, incidence is not a metric that is commonly measured, except as a percentage of market share or in assessing the demand for healthcare services within a geographic area. But under population-based health care, incidence is a critical driver of costs to the contracting organization. Fee-for-service providers want incidence to increase; population-based health organizations want it to decrease. Therefore, understanding and reviewing these metrics frequently are key to creating to paradigm shift necessary for finance managers to operate successfully in a population-based world.

Matching numerators with denominators. Care should be taken when applying population-based metrics to make sure that the population associated with the numerator is also the population associated with the denominator.

For example, computing "cancer costs PMPM" (computed by dividing the costs of patients having cancer diagnoses by the total number of member months) combines the incidence of cancer in the population with the costs of treating cancer patients. This approach makes it difficult to isolate and analyze those two factors individually, which have significantly different causes. A high relative cost using this metric may suggest incorrectly that one population has higher cancer treatment costs when instead it has a higher proportion of cancer patients.

It may be more useful to compute one or both of the following metrics: "patients having cancer diagnoses as a percentage of the membership," and "cost of cancer per cancer patient member months"-the latter using the number of cancer patients, rather than the entire population, as the denominator. These metrics are more specific and provide more useful information as to their causes and effects.

Similarly, it may be tempting to compute a PMPM cost of an individual provider when that provider is not the sole provider of its type for the population, but doing so is incorrect. For example, a metric such as "Dr. Smith's cost PMPM," computed by dividing the amount of payments made to Dr. Smith by the total number of member months of the contracting organization, is not a valid metric because Dr. Smith doesn't have "members"-he has only "patients," because some members may be utilizing other providers in his specialty. A low PMPM cost may simply indicate low utilization of his individual services, rather than a low cost of treating patients.

PMPM costs for a type of service (such as cardiology costs PMPM) are valid because they encompass the entire population. However, the only time a provider-based PMPM metric is valid is when the provider actually has specific assigned members, as may occur among primary care physicians but rarely occurs for most other specialties.

Risk Adjustment

A "risk adjustment" refers to a quantification of the relative costs of a particular population based on its medical and demographic conditions. It is typically used to adjust capitation payment rates to account for differences between the populations that provided the bases for computing payment rates and the populations to which the rates are applied. For example, if a contracting organization had a higher-than-average percentage of diabetic patients, its risk-adjustment factors should reflect that difference.

Many different types of risk adjusters are available, but the most common and easiest to implement follows a "hierarchical condition categories" (HCC) grouping process that is used for Medicare Advantage and Programs of All-Inclusive Care for the Elderly (PACE) programs. This process uses clinical factors (based on diagnosis and demographic data) that have been shown to predict cost, identifies those conditions when they occur in each patient, assigns the respective risk scores from those factors to that patient, and then sums those factors to arrive at the total risk score for each patient.

In Medicare Advantage plans, these aggregate risk factors are then applied to a standard payment rate in the same way that DRG weights are applied to the conversion factor to yield a payment rate for each patient, although ACOs and commercial payers may use different methods to compute the payments.

Risk factors are helpful to contracting organizations in several ways. First, when comparing PMPM costs with the associated risk factors, organizations should find a rough correlation between the patients' costs and their associated risk factors. Although most HCC research has not found a strong correlation on a per-patient basis, it is intuitive that the extreme values of cost and risk factors should bear some relationship to each other, simply because sick patients are generally costly. High costs but low risk scores for a patient may indicate inappropriate or unnecessary treatment for a relatively uncomplicated patient, whereas higher risk scores accompanied by low costs may indicate a patient with critical medical conditions who is being undertreated.

In addition, the risk-scoring process itself identifies patients having the most common chronic conditions, such as diabetes, chronic obstructive pulmonary disease, and asthma. Identifying these patients and targeting them for special treatment is a key component of any population-based health system. Analytics that incorporate risk scores are critical for population health management.

Understand the Differences

The financial incentives, metrics, and accounting components associated with population-based health care differ significantly from those of traditional healthcare systems that are based on fee-for-service payment. Finance managers in organizations pursing ACO or other capitated strategies need to grasp the underlying differences between these components and apply them in managing their organizations.

Jonathan W. Pearce, FHFMA, CPA, is a principal, Singletrack Analytics, Marlton, N.J., and a member of the board of directors of HFMA's Metropolitan Philadelphia Chapter (


Defining What Constitutes a Member Under Population-Based Health Care  

The basic premise of the population health model is that payment is based on the delivery, or potential delivery, of care to a specific large set of individuals identified as being members of the population that the provider serves. This population includes not only the provider's patients (i.e., those individuals who are actively seeking medical care from the provider), but also healthy individuals who are not actively seeking care. Defining this population is not a straightforward matter: Although members is the commonly used term for the individuals composing the population, these individuals may not be actual members of any specific group, and may even be unaware that they belong to the specific capitated population.

Understanding the details of how members are assigned to the population is important to a contracting organization because those members represent the underlying volume of that organization's business.

Choice is the simplest determinant of whether individuals are members of the capitated population.

In the insurance world, members are simply those who choose to purchase insurance from the insurance company. The same holds true for Medicare and Medicaid, because beneficiaries enroll in those programs.

Primary care physicians are often capitated under HMO arrangements that require each insured individual to select a specific primary care physician and use that physician for all primary care services. Again, membership in such a population is a matter of choice.

In other types of organizations, such as a Medicare accountable care organization (ACO), the members do not have the opportunity of choice. Instead of selecting an ACO, each member is "attributed" to the primary care physician who provides most of his or her services, and who is a member of a particular ACO. Moreover, with the Medicare ACOs, members may not be identified until after the demonstration has started, and as alluded to previously, they may not initially even be aware of their "membership" in the organization.

This lack of choice on the part of the member extends also to commercial payer contracts with physical therapy, laboratory, and radiology providers, under some types of arrangements. These providers often "inherit" members from primary care physicians by having been selected by the payer or primary care physician to be the sole provider of those types of services for the physician's capitated members. In such an arrangement, the specialty provider receives a single payment, generally monthly, for his or her entire capitated membership regardless of whether he or she delivers any services to those members. Should members require services, the provider must render those services, generally without receiving additional payment.


a. There are adaptations of capitated payment systems called "contact capitation" that are sometimes applied to a specialist once a member becomes a patient of the specialist, but those systems are actually episode-of-care systems that are patient-based rather than population-based.

Publication Date: Monday, July 02, 2012

Login Required

If you are an existing member, please log in below. Username and password are required.



Forgot User Name?
Forgot Password?

If you are not an HFMA member and would like to access portions of our content for 30 days, please fill out the following.

First Name:

Last Name:


   Become an HFMA member instead