How much physicians get paid is increasingly determined by a formula that penalizes doctors whose patients are more expensive—even when those higher costs stem from services that other doctors perform, according to a new report.
How much physicians get paid is increasingly determined by a "Kafkaesque" payment formula, one that penalizes doctors whose patients are more expensive—even when those higher costs stem from services that other doctors perform.
That's the analysis from a healthcare payment reform consultant and author of a new report, "Fair and Effective Ways to Analyze the Drivers of Healthcare Costs and Transition to Value-Based Payment."
Date: August 27 1:00-2:30pm ET—presentations and live Q&A Leaders from Cleveland Clinic and Catholic Health Initiatives reveal how they have successfully positioned their organizations’ cardiovascular service lines for growth through leadership training, partnerships, and better care coordination. >>>Register
A complex system of physician "accountability" for services is already in place with some Medicare demonstration projects, ACOs, and some commercial health plans, and is soon to be folded into the Medicare's physician fee schedule formula, says Harold D. Miller, president and CEO of the Center for Healthcare Quality and Payment Reform, a research nonprofit he launched.
But, he explains, the system uses an unreasonable method to assign a patient's healthcare costs and quality to a particular doctor based on the plurality of visits to that doctor in a given year, not whether the doctor actually performed or referred the patient to the expensive services.
In what he calls an extremely bizarre example of the worst kind of unintended consequence, the formula could prompt doctors to "try to avoid seeing" patients with certain high-cost conditions, especially if the patients previously saw them or might switch to other doctors by year's end.
The penalty for low quality and/or high cost could affect 10.4% of providers in practices of 100 or more, cutting as much as 1% from their Medicare physician fee schedule payment, Miller says (based on estimates in a January 2014 report).
But Miller says the reduction in payment will go up to as much as 2% in 2016, when the formula applies to practices of 10 or more physicians, and it will increase to 4% in 2017 based on proposed regulations from Medicare, when it will apply to individual physicians as well as larger groups. The study estimates that a higher percentage of providers in smaller practices would be affected than those in practices of 100 or more. What's more, proposed Congressional sustainable growth rate reductions would increase potential reductions in payment over time to as much as 9% by 2021.
Unfortunately, Miller says, "most doctors don't understand the complexity of this yet. The average person out there believes that Medicare is paying for value and that's a good thing. But … it really doesn't do that, and there's a question of what undesirable side effects this creates."
The new algorithm is designed to reward doctors for practicing patterns that combine higher quality, in which they score high on performing certain process measures that some believe lead to better outcomes, and lower costs, in which they demonstrate they avoid duplication or unnecessary tests and procedures.
Costs and quality scores for a patient's care are "attributed" to the practices where that patient had the plurality of primary care visits, regardless of what other expensive services the patient received from other doctors throughout the year.
In a scenario in his report, Miller gives an example of what usually happens with a very large percentage of a doctor's patients in any given year. In an interview, he elaborated on why the accountability system is "Kafkaesque."
In January, a patient visits his primary care doctor because of chest pain while exercising. The doctor orders a stress test. In February, a cardiologist does the test and determines the patient has risk factors and orders medications.
The patient also experiences lower back pain, but instead of going to his primary care doctor, he sees a neurosurgeon who performs surgery at a center 50 miles away from the patient's home. An anesthesiologist delivers sedation. The patient gets physical therapy at a skilled nursing facility. Unfortunately he develops an infection and gets admitted to a community hospital in May. After a hospitalist treats the infection, the patient goes home with a recommendation to see a primary care doctor.
In June and September, the patient visits a new primary care doctor, who makes a referral for GI health. A gastroenterologist performs a colonoscopy.
This hypothetical patient received care from eight physicians and two hospitals, but under the accountability method, all of those costs are attributed to the second primary care doctor because that doctor saw the patient twice. Costs for the stress test, the back surgery, and the hospital readmission after the surgery were all attributed to the second doctor even though those services occurred before the new primary care physician had met the patient for the first time.
"None of the services would be assigned to the other physicians who actually delivered or ordered them," Miller says.
Miller says that this assignment or accountability strategy is rolled into demonstration projects under way at 481 sites in seven states participating in the Medicare Comprehensive Primary Care Initiative. Several commercial health plans use the same strategy.
And through a recently modified Medicare program called the physician value-based payment modifier, the strategy is scheduled to affect payment for physicians in practices with 100 or more eligible professionals starting in January, for physicians in practices with 10 or more the following January 2016, and for all physicians by January 2017.
When and if doctors begin to understand this new model, Miller says, they will realize that it discourages groups with large numbers of specialists from keeping lots of primary care doctors in their group.
While he is not recommending that doctors do this to avoid a penalty, he suggests that —as a way of explaining "how screwed up the methodology is" — doctors who want to take actions to score more favorably might "go dump all your primary care physicians. Because if you have primary care physicians in your group, you'll get attributed patients, but if you don't have any primary care physicians in your group, you won't get attributed any patients and you won't be subject to the cost measure associated with it. And since the quality measures tend to be related to primary care, you won't get many quality measures attributed either."
The group might not receive a reward for better performance on cost and quality measures, but at least it wouldn't be penalized either, he says.
Miller's report highlights six major problems with the accountability system. They are:
1. For many patients, the spending on their care is not assigned to any provider. "Consequently, the spending associated with those patients is ignored in spending analysis. This can cause distortions in comparisons of spending between providers and it can also create perverse incentives for the providers," he says.
For example, patients who aren't getting adequate preventive care will be excluded and providers who take on care of these patients will be financially penalized.
In what he says is a kind of catch-22, the accountability method of attributing patients to particular providers is stating that, in effect, "you're giving us all this money so we don't bring these patients into the office, but if we don't bring them into the office, we won't get the money."
2. Providers cannot control all services and spending assigned to them.
3. Providers are not attributed the spending for many services they provide.
4. Spending measures do not distinguish appropriateness of services.
5. Risk adjustment systems do not adequately adjust for patient needs.
Miller says that risk adjustment equations predict spending on patient care but don't adjust for differences in patients' needs, which can incentivize inappropriate spending or penalize efforts to reduce underuse.
Furthermore, most risk adjustment systems use historical information on patient characteristics, not current information on health problems that affect services patients get.
6. Inadequate adjustments are made for structural differences in costs, for example for providers in rural areas and poor communities, who incur higher costs to deliver the same services than other providers. Adjustments do exist to account for this difference but they are limited and don't reflect factors outside of the doctor's control.