Skip to main content

To Risk or Not to Risk? Or Is Risk Contracting Too Risky?

February 2019, Vol 9, No 2
Dawn Holcombe, MBA, FACMPE, ACHE
President, DGH Consulting, South Windsor, CT

The word on the street is that value-based, or risk-based, oncology contracts for physicians are the inevitable wave of the future. But are they? I had a conversation late last year with someone in the federal government. We were discussing whether oncology practices should be accepting the modified alternative risk model that has recently been offered to them under the Oncology Care Model (OCM) program. This individual felt very strongly that payment programs that put oncologists at full risk for their patients’ treatment costs (including drug costs) were so imminent that it would be irresponsible for practices to not accept the modified alternative risk model as soon as possible so that they can be prepared.

That conversation was very wor­risome to me. There are building blocks that are critical for the management of financial risk, and I am not yet convinced that most cancer centers and oncology practices are even close to being ready.

Consider the following:

  • Volume
    An important concern is volume. Risk models for insurers are based on per capita calculations. The number of patients with cancer who are being treated by one cancer center for one insurance program is likely to be far less than 1000—in reality, the number is probably in the low- to mid-­hundreds. No insurance company could open its doors with a covered population of less than 1000 and expect to be financially stable. Why then is there an expectation that medical practices for a single specialty should be able to absorb significant financial risk for relatively small numbers of patients?
  • Distribution
    The incidence rates of new cancers in the United States are 439.2 per 100,000 men and women per year.1 In a general insured population, there will be a large number of covered members who do not have cancer to help offset the costs of those who do have cancer. However, a cancer center cares exclusively for patients who have the disease. Therefore, the risk pool they deal with is already in the high-cost range, with less opportunity for absorbing variation.
  • Analytics
    Risk management requires effective use of technology to track, manage, and predict patient populations, cost corridors, and predictive modeling for proactive management. Oncology centers have tools that are minimally useful for electronic medical records and revenue cycle management, and none of these are equipped for risk management.
  • Interoperable technology
    Vendors for the technology available to oncology centers do not make it easy for these centers to connect their patient demographics and care details with external systems for necessary additional analytics. Most of these vendors have a history of making it difficult, if not impossible, for oncology centers to transfer data into other programs. The vendors also place financial and operational barriers on the oncology centers, despite their inability to provide the centers with the complex analytics required for effective risk management.
  • Financial reserves
    The financial reserves necessary for engaging in risk models may be beyond the scope of most cancer centers. There are state regulatory reserve requirements established for statutory capital reserve ratios on insurance companies before they are allowed to conduct business. Standard levels may run from 8% to 12% of the insurer’s total revenue, but are also likely to vary depending upon the type of risk.2 We have not yet sufficiently explored the consequences on healthcare providers entering into risk contracts. Will they be subject to state financial reserves requirements? Will they be somehow restricted from practicing should those requirements not be met?
  • Real-time data management
    Med­ical practitioners who attempt to navigate through payment models that include not only their own services at risk, but also risk for high-cost cancer drugs or other medical services, need as much data as possible. It is likely that they will be asked to manage risk for the total costs of care, and yet have no direct information on costs incurred outside of their own offices. The Medicare OCM is the largest single payment model moving toward a risk model. This model provides complete cost data for every beneficiary in the program, but the data provided to the managing physicians are months behind real-time.

The OCM model offers options for participating practices to move voluntarily into 2-sided risk models, but few, if any, practices have yet volunteered. At the end of 2018, the OCM 2-sided risk model was modified to limit risk only to a percentage of Medicare revenues billed under the organization’s tax identification number. However, with reimbursement for drugs used to treat cancer included in those Medicare revenues, many practices are still unwilling to accept the potential of having to repay Medicare what could amount to millions of dollars if government-created targets are not met.

Barbara L. McAneny, MD, is the current president of the American Medical Association, and a noted oncologist recognized for her innovations in physician payment models, such as the COME HOME model that was funded by Medicare as a precursor to the OCM model. Dr McAneny praises the COME HOME model for its ability to save Medicare more than $2000 per patient annually by prioritizing triage protocols, clinical pathways, same-day appointments, and better patient education.

“We created savings through the things that the doctors can control,” she said. She has recently introduced a new real-time oncology payment model led by physicians called Making Accountable Sustainable Oncology Networks (MASON), which seeks to create real-time quality measurement so that practices can respond to outliers and problems right away. The model “creates an accurate cost target that will be a valuable tool for optimizing patient management while avoiding actuarial risks of adverse patient clinical characteristics,” the proposal states.3

The challenge for practices is that while innovative payment models are being presented by Dr McAneny and others, there may be considerable pressure within the payer communities, including Medicare, to assume that 2-sided risk is inevitable and possibly at odds with the more physician-friendly payment models being proposed. Our mission is now to clarify—to ourselves and to those developing payer policies—the inherent hurdles for cancer centers and full-risk models. We need to have open discussions across practices and networks about the financial obligations and regulatory restrictions on assumption of risk in the medical community. We need to seek out appropriate software tools and face the challenges of trying to extract our own data from our electronic medical records systems. Some groups are starting to discuss the potential of shared risk pools, to combine different practices’ exposure. Their hope is that these would pool any losses of one participant across all the participants.

Even with the recent revisions to the risk parameters, a negative risk event for some of the current OCM practices could yield an obligation to repay Medicare hundreds of thousands of dollars, if not more than $1 million, for a given 6-month performance period. This is a huge bullet to swallow for any cancer center. Physicians and payers remain on opposite sides of the table as to whether the costs of oncology treatments (oral and administered) should be considered as part of the risk calculation. When drug costs remain in the equation, the option of accepting 2-sided risk looks less likely for most cancer centers. This discussion of risk or no risk will continue during 2019, with no clear resolution in sight.


  1. National Cancer Institute. Cancer statistics. Updated April 27, 2018. Accessed January 21, 2019.
  2. Ross S. What level of reserve ratios is typical for an insurance company to protect against large losses? Updated August 22, 2018. Accessed January 23, 2019.
  3. Caffrey M. AMA’s McAneny calls for real-time oncology payment model led by physicians. November 17, 2018. Accessed January 21, 2019.

Related Articles