Medicare P4P demonstration project shows significant QI

Participants improve in all five clinical areas over three-year period

Quality of care has improved significantly in hospitals participating in a groundbreaking Medicare pay-for-performance demonstration project, according to preliminary reports from more than 270 participating hospitals on their experience during the project’s first year.

Called the CMS/Premier Hospital Quality Incentive (HQI) Demonstration Project, after its partners, the Centers for Medicare & Medicaid Services and Premier Inc., a health care alliance entirely owned by more than 200 of the nation’s leading not-for-profit hospital and health care systems (a total of about 1,500 hospitals), the initiative tracks hospital performance on a set of 34 nationally standardized and widely accepted quality indicators and pays annual incentives to top performers among participating hospitals.

To date, four quarters of preliminary data have been gathered, which show median quality scores for hospitals improved:

  • from 90% to 93% for patients with acute myocardial infarction (AMI);
  • from 86% to 90% for patients with coronary artery bypass graft (CABG);
  • from 64% to 76% for patients with heart failure;
  • from 85% to 91% for patients with hip and knee replacement;
  • from 70% to 80% for patients with pneumonia.
  • The median performance composite score for all hospitals — just one measure of improvement — went up 7.5% in the project’s first year. A number of individual hospitals scored impressive results. For example:
  • Nineteen hospitals improved their quality score for care of heart failure patients more than 30%.
  • Eighteen hospitals improved their score for care of pneumonia patients by more than 20%.
  • Seventeen hospitals improved their score for care of hip and knee replacement patients by more than 15%.
  • Twenty-one hospitals improved their AMI score more than 12%.
  • Sixteen hospitals improved their CABG score more than 10%.

In addition to overall improvement in quality scores, the variation in quality of care among participating hospitals is narrowing as all hospitals are demonstrating improvements. This means the gap between top performers and lower performers is shrinking.

Hospitals participating in the project cared for more than 400,000 patients in the five conditions during the first year. During the course of the three-year demonstration project, which began in October 2003, Medicare will reward high performers with bonuses totaling $7 million per year for a total of $21 million. Poorly performing hospitals may face financial penalties in the third year.

Under the Premier demonstration, a hospital can receive bonuses in its Medicare payments based on how well it meets the quality measures. Hospitals are scored on measures for each condition, and those in the top 10% for a given condition will be given a 2% bonus on their Medicare payments for that condition. Hospitals in the second 10% will be given a 1% bonus. Hospitals in the remainder of the top 50% get recognition for their quality, but no bonus.

At the end of the first year, baselines will be set for the bottom 20% and the bottom 10%. These levels remain static, and CMS and Premier expect that all hospitals will be above the baselines by the final year of the demonstration. If any hospitals are below the 10% baseline in the third year of the demonstration, they will get a 2% reduction in Medicare payments for the clinical area involved, and those between 20% and 10% will get a 1% reduction.

Sponsors pleased

Not surprisingly, both CMS and Premier are pleased with the results to date. "We were hoping and expecting that the provision of incentives would serve to enhance quality," says Mark Wynn, PhD, CMS’ director of the division of payment policy demonstrations. "I think we are all delighted we’ve seen such immediate and good improvement in the quality of care provided as measured under the demonstration project. Our focus is on improving quality for Medicare beneficiaries, and we can see, at least in this initial test, that paying for performance can move quality in the right direction," he notes.

"The results have been absolutely outstanding," adds Stephanie Alexander, MBA, senior vice president with Premier. "The participating hospitals were already using data to improve and were performing at higher quality levels than other hospitals, but when I looked at the first quarter of data and began tracking them, it’s just amazing that the whole group in aggregate really moved forward."

Of course, cautions Wynn, "We’re not going to prove the whole concept until we finish our evaluation, but the initial numbers are so dramatic so far, it appears true that the incentives and the focus surrounding these incentives are the reasons for the improvement."

He adds, however, that the incentives may not have been the only reason for the improvement, but that they are a means of focusing on important quality improvement issues. "I think both the Medicare program and the participating hospitals are all focusing on QI to the extent possible and that this project is giving a meaningful focus for these efforts," Wynn says. "We find from our initial discussions with the hospitals that they use this as a motivating focus to really drive their QI work."

Closer look at project scoring

The CMS (Centers for Medicare & Medicaid Services)/Premier Hospital Quality Incentive Demonstration Project’s overall composite score is calculated by combining the process of care and outcome measures, including mortality and complications, from five clinical focus areas.

The clinical focus areas are acute myocardial infarction, coronary artery bypass graft, heart failure, pneumonia, and hip and knee replacement surgery.

Premier collects, processes, and reports data for the project through its Perspective clinical comparative database. The data remain preliminary until CMS completes a rigorous auditing and validation process.

While the preliminary data reveal significant improvements in all focus areas, the most dramatic improvements were seen in indicators reflective of care provided to heart failure and pneumonia patients.

For all hospitals participating in the project, the overall quality score for heart failure, which includes assessment of heart function, provision of detailed discharge instructions, and appropriate use of selected medications, improved 12%.

Similarly, the overall quality score for pneumonia care, incorporating indicators such as assessment of patients’ oxygen status and selection and implementation of appropriate antibiotics in a timely manner, improved 10%.

Assessing impact of P4P

Alexander says she’s convinced that the P4P arrangement contributed to the quality improvement at participating hospitals. "We have quite a few health systems, and some came in with all of their hospitals, while others did not.

"One of them showed us that those who were in the project were performing better than those who [were] not. The project creates a sense of urgency and focus, and bringing this group together to focus on improvement brought results that much faster," she explains. (For more on Premier’s interpretation of the results, see box, at right.)

Clearly, numerous structures are being used within the broad category of "pay for performance." What does Alexander think of this particular model? "I think there are absolutely some very good parts to this structure from which to move from a national pay-for-quality model to a pay-for-performance model," she notes, making an important distinction.

"This only measures effectiveness of care — in the future, we will focus both on effectiveness and efficiency. The other challenge we have is that we must find a way not just to incentivize physicians or hospitals but to incentivize them as a team together. That’s a fundamental flaw in our Medicare system today — there are two different payment structures," Alexander points out.

"The point is, these are not huge increases," Wynn notes. "But even with these limited amounts of money, it’s helping to focus the attention of these hospitals and CMS on these improvements. We will have a web site on which the winning hospitals are prominently displayed, and hospitals are really interested in that."

He adds that the Bush administration "is very interested in evaluating and pushing on pay for performance," and that to that end, the results of the project will be forwarded to the appropriate administrator.

Need More Information?

For more information, contact:

  • Stephanie Alexander, MBA, Senior Vice President, Premier Inc. Phone: (704) 733-5446.
  • Mark Wynn, PhD, Director, Division of Payment Policy Demonstrations, Centers for Medicare & Medicaid. Phone: (877) 267-2323.
  • Comprehensive information about the CMS/ Premier Hospital Quality Incentive Demonstration Project is available at and on the CMS web site at