Institute of Medicine recommends new P4P system for Medicare
P4P should be phased in, IOM committee recommends
Joining a growing group of organizations that finds merit in a pay-for-performance approach to quality improvement, the Institute of Medicine (IOM) has recommended that Medicare should gradually replace its current fee-for-service payment system with a new pay-for-performance system for reimbursing participating health care providers.
In the report, Rewarding Provider Performance: Aligning Incentives in Medicare, the IOM committee that wrote the report asserts that "Medicare's current payment system places no emphasis on whether the care delivered is of high or low clinical quality, or is appropriate." However, the report adds, since pay for performance does not yet have an established track record, the new system should be phased in so that involved parties can build on successes along the way and avoid unintended negative consequences.
For an initial period of three to five years, the committee recommends, Congress should reduce base Medicare payments across the board and use the money to fund rewards for strong performance. At the same time, efforts should be made to evaluate other ways to fund bonus payments that could be used longer term. Many large, institutional health care providers and organizations that already have the capacity to begin participating in the pay-for-performance system should be required to do so as soon as it is launched, the report adds.
The committee recommends the following design principles for pay for performance and its implementation:
- Use performance measures that reliably define good care and optimal health outcomes.
- Reward care that is of high clinical quality, is patient-centered, and efficient.
- Reward significant provider improvement as well as achievement of excellence.
- Foster care coordination among providers.
- Reward data collection and reporting functions and encourage adoption of improved information technologies.
- Report provider achievement in ways that are both meaningful and understandable to consumers.
- Develop performance measures and structure rewards to maximize participation of all providers over time.
- Be fiscally responsible.
- Implement in deliberately planned phases, evaluate progress, and learn from experience in each phase.
One of the weaknesses of Medicare's current fee-for-service payment system is that "the more you do, the more you get paid. It just doesn't help quality," notes Steven A. Schroeder, MD, distinguished professor of health and health care, University of California, San Francisco, and chair of the IOM committee. If you look at the previous IOM report, Crossing the Quality Chasm, he continues, "They have six different measures, and one was efficiency. The current system actually pays you more for doing what is costly, rather than for care that is cost effective."
Report carries weight
The urgency of the situation, says Schroeder, "demands that steps be taken now to encourage health care institutions and clinicians to improve their quality. Pay for performance has demonstrated sufficient promise based on early experience that it should be pursued, albeit cautiously and in a manner that allows for learning and adjustment as needed. And we should remember that pay for performance is just one part of the solution; other interventions will be needed to achieve the level of quality that Medicare patients deserve."
"IOM is right to say that payment rewards alone won't get the job done," says David Schulke, executive vice president of the American Health Quality Association (AHQA). "Take surgical infections, for instance. When a patient has a post-surgical infection, hospital costs per case double, and their profit margin per case falls by 85% — a huge financial incentive. If financial rewards were all we needed to stop surgical infections, they would have stopped long ago. Clearly, financial rewards are not enough to produce better quality; they are just one important piece of the quality puzzle."
Nevertheless, say quality observers, having the IOM express support for a P4P model carries a great deal of weight.
"We definitely believe IOM's support is significant," notes Denise Remus, PhD, RN, vice president of clinical informatics for Premier, Inc., in Charlotte, NC. "The IOM is obviously highly regarded in the industry, and they have a wonderful process through which they create reports. I believe this report carries significant weight in the industry and with decisionmakers in Congress — and with CMS [The Centers for Medicare & Medicaid Services]."
"I think it's a great report — very well thought out — by a group of people who are absolutely undisputed experts in the field," adds Christine Bechtel, director of government affairs for AHQA in Washington, DC. "We are grateful to have a body that does such a deliberate job in considering what is a very complex issue — and their conclusion is dead on. P4P is promising, but a lot of design issues remain."
Why the caution?
Despite all of the industry buzz about P4P, and some encouraging initial findings from sources such as the Premier/CMS demonstration project, the IOM withheld a full-throated endorsement of the model.
Although more than 100 incentive programs have been launched in the private sector in the past few years, the report notes, fewer than 20 studies have assessed the impact of these programs on quality of care and health outcomes. However, noting that both private- and public-sector groups are eager to move forward with pay for performance, the committee concluded that a gradual implementation would enable officials to assess the program along the way, adapt to knowledge gained, and monitor for unintended negative effects — such as providers avoiding certain kinds of patients or withdrawing from Medicare.
"In the study itself, we summarize all that is known about pay for performance; some research shows there was benefit, some shows there wasn't," notes Schroeder. "It looks like the amount of money Medicare would be able to pay, especially at the physician level, would be very small, and not many studies show [positive] effects of relatively small bonuses."
He adds that as the committee sees it, there are ideally two components of P4P — one is payment; the other is public reporting. "We felt both components together had the potential to be quite powerful," he says.
Schroeder also recognizes that "there's a lot of sentiment in Congress and elsewhere to do [P4P], so this makes it timely, but by itself we felt it wouldn't be the sole answer to major improvements in quality."
"We agree; there's no question that there's a paucity of data, as well as conflicting data," says Bechtel. "We also agree that the current reimbursement model is not set up to accelerate improvement, and P4P is one approach, depending on design, that can do that."
AHQA has two main issues with P4P, she continues. "One is evidence, and the other is the impact, and that will be driven by design. Can it sustain itself over time, generate a high enough level of participation, and accelerate transformation?"
"One important point to consider is that when you hear debates about whether P4P will work or not, we tend to use the term as if it means one program, or one model, where in reality there is not a national standard," Remus emphasizes. "Yes, there have been some models implemented in the industry that have not been successful or have not rewarded top performers, but our experience is that we have had an extremely successful impact on quality of care delivered and helped hospitals align value around quality. But even our program is not what we believe a national model should be; we are still learning."
Seeing positive parallels
Remus goes on to observe that the major elements of the IOM report align well with Premier's own experiences in its program, formally called the Hospital Quality Incentive Demonstration (HQID) project.
"We concur that P4P should be implemented; we believe that aligning payment to quality is the right thing to do," she asserts.
Rewarding health care that is of high clinical quality and is patient centered and efficient is another area of clear agreement, she continues. "In our project, we use measures that include process and outcome; we have proposed an extension that would include looking at HCAP (CMS's patient experience survey), but we also believe there needs to be more engagement nationally to develop additional measures that are patient-centered — and that look at efficiency."
In the "Performance Pays" program, she adds, "When we looked at just process, when we received [data on] all measures we had fewer costs, lower lengths of stay, fewer mortalities and re-admissions. You can really help hospitals not only identify opportunities for improvement but help them set up efficient, effective systems of care."
Another important point noted in the report, she says, is that P4P should reward not only those facilities that improve significantly, but also those that continue to achieve high performance. "As we talked to CMS staff about what's going on, one of the things they mentioned was there really was not as much benefit to Medicare and their beneficiaries for hospitals to move from 94%-96% as one that was stalled at 60% to move to 80%," she explains. "There's clearly a huge gain in that improvement, but we also want hospitals that are at 96% to stay there."
New model elements
Quality experts acknowledge that additional elements will be required before an ideal model can become a reality. "One thing they did in the U.K. was give very substantial payments for performance around certain criteria for GPs, but I don't think our Medicare program is prepared to put the same money on the table that they did," says Schroeder. "But a number of programs looking at hospitals saw they were willing to improve with relatively smaller dollar rewards, so it may be the hospital sector is riper for this."
The IOM committee deferred to Congress to determine by how much to decrease Medicare base payments to create a pool of funds for bonus payments.
Using a reduction in base payments to fund bonuses should be used initially while other, more sustainable long-term strategies are explored, they recommended. Sustaining the rewards pool through savings generated by improved efficiency and cost-reducing reforms has great potential, the committee said, and it urged CMS to test ways to make this funding source work.
"Number one for us is the technical assistance program," adds Bechtel. "This is not just about money; you can't throw money at the problem and say 'improve.' You also have to give the hospitals help from experts who know how to improve."
The committee, she notes, says it's important to expand QIOs' ability to provide help to anyone who wants it.
"We also need improvements in IT — for a lot of reasons," she continues. "We need it for data. We have to know, for example, if a diabetic's hemoglobin A1c is in control; you can't get that information from a claim, but you can get it from an EMR [Electronic Medical Record]. It's an important tool to help improve performance." What's required in addition to the technology, she emphasizes, "is the technical assistance on how to implement it, so that hospitals and physicians are using the systems to the fullest to improve quality and safety of care."
"I believe we should also keep adding areas to look at for quality improvement," says Remus. "For example, performance in heart care is high, but for stroke you may find some facilities down at 60%. I see that concept as something that can help – the conditions and measures hospitals will look at should continue to expand. You also want to use quality measures that are evidence-based."
What you can do
What should quality managers do in response to this report — or, more importantly, prepare for the new environment it portends? "Number one, they should read the report carefully," Schroeder advises. "They should try to figure out what things are going to be measured that aren't measured now — and look at the quality of coordinated care."
By that, Schroeder says, he means avoiding condition "silos." "We tend to try to do it that way, but many Medicare patients who are quite ill have multiple conditions — heart disease, diabetes, kidney disease — so you have to look at how to coordinate that care. Also, they should pay attention to how they collect the data and how they report it and should be conscious that there is growing sentiment among private payers to move to P4P."
Schroeder also predicts an increase in public reporting. "You've got to be sure to capture your data accurately and be knowledgeable, or you may see stuff about your hospital in news or TV stories," he warns.
"If I had any recommendations for quality managers, it would be that they look toward early discussions about what's happening in the Hospital Quality Alliance, the Ambulatory Quality Alliance, and measures being submitted to NQF for evaluation — and, of course, the Joint Commission," says Remus. "That really seems to be the pool [of measures] that Medicare will require. Hospitals will do well to not only make sure they look at what's required, but at how they can start to position themselves for these other measures of outcomes — before CMS mandates them."
She also recommends using comparative data to benchmark against the top performers. "Ask yourself who is performing well, what you can learn from them, what the barriers are for your facility, and how you can learn from those to do better," Remus observes. "We have a lot of programs in place to facilitate knowledge transfer; it's been a huge piece of our success."
The path to P4P
"I think looking at technology and the level of health IT they use in their hospital is really smart – not just in preparation for P4P but because there's a big push for health information exchanges like RHIOs [Regional Health Information Organizations]," Bechtel recommends. "Then, they should look system-wide at already established measure sets and what assistance they may need to improve."
Hospitals, she continues, "are the group everybody is watching to see the path to P4P — because that's where the model will be established. In the report, the IOM indicates that hospitals are a pretty good place to start, so the rest of the provider world will look at them to see what the path is."
And where does she suggest the hospitals get their own guidance? "Make sure you know everything about the Premier demonstration project," she recommends. "It will tell hospitals what's coming down the pike."
For more information, contact:
Steven A. Schroeder, MD, Distinguished Professor of Health and Health Care, University of California, San Francisco. Phone: (415) 502-1881.
Denise Remus, PhD, RN, Vice President of Clinical Informatics, Premier Inc., 2320 Cascade Pointe Blvd., Charlotte, NC 28208. Phone: (704) 733-5634.
Christine Bechtel, Director of Government Affairs, The American Health Quality Association, 1155 21st Street NW, Suite 202, Washington D.C 20036. Phone: (202) 261-7569.
Joining a growing group of organizations that finds merit in a pay-for-performance approach to quality improvement, the Institute of Medicine (IOM) has recommended that Medicare should gradually replace its current fee-for-service payment system with a new pay-for-performance system.You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content