Susan D. DeVore
Results from the nation's largest hospital-based pay-for-performance project suggest that value-based purchasing could be an ideal model for healthcare payment reform.
At a Glance
Some of the lessons hospitals that have participated in the Hospital Quality Incentive Demonstration project have learned include:
- The need to tie in quality-of-care initiatives to the organization's strategic plan and to incentive plans for all employees, from executives on down
- The value in allowing hospital physicians to "own" quality improvement initiatives
- The importance of making results of the initiative available to all staff
- The benefit of creating best-practice teams to address improvements in specific clinical areas
As lawmakers continue their quest to reform the nation's healthcare system, one undeniable truth has emerged: The status quo system of paying hospitals is broken. Many believe rewarding hospitals based on quality of care and improved outcomes makes sense, especially when there is substantial evidence to suggest that pay for performance, or value-based purchasing (VBP), creates the right incentives to drive continuous improvement.
Results from the national's largest hospital-based pay-for-performance program-the Hospital Quality Incentive Demonstration (HQID) project, coordinated by the Premier healthcare alliance on behalf of the Centers for Medicare & Medicaid Services (CMS)-show that this type of payment model can and does work. Hospitals participating in the HQID project have made significant improvements in their quality scores across five clinical areas. CMS has awarded more than $36.5 million in incentive payments to participating hospitals during the first four years of the project, with $12 million in payments awarded to 225 providers during year four of the project (2006 to 2007), the most recent year for which statistics are available.
About the HQID Project
The HQID project is the first national project of its kind, designed to determine whether economic incentives to hospitals improve the quality of inpatient care. The 250 hospitals participating in the HQID project include small and large, urban and rural, and teaching and nonteaching facilities that volunteered to report their quality data for five high-volume inpatient conditions using national measures of quality care.
Through the project, Premier collects a set of more than 30 evidence-
based clinical quality measures, developed by government and private organizations, from participating hospitals. These measures are based on scientific evidence and, for continued effectiveness, are often reviewed to account for medical breakthroughs and new research. They are tested in care settings, validated by third parties, and proven over time to improve quality.
Hospitals can receive additional payment from CMS if they achieve top performance, achieve top improvement, and/or attain or exceed a set quality score based on results two years prior, across five clinical areas:
- Acute myocardial infarction (heart attack)
- Coronary artery bypass graft (CABG)
- Heart failure
- Hip and knee replacement
Participants are eligible to receive 10 total awards in these areas.
Among hospitals participating in the HQID project, average composite quality scores (CQSs)-representing an aggregate of all quality measures within each clinical area-have improved by 17.2 percent over the project's first four years in all five clinical focus areas. Average CQS scores rose:
- From 87.5 percent to 96.3 percent for patients with AMI
- From 84.8 percent to 98.5 percent for CABG patients
- From 64.5 percent to 92.2 percent for patients with heart failure
- From 69.3 percent to 92.6 percent for patients with pneumonia
- From 84.6 percent to 97.2 percent for patients with hip and knee replacement
These improvements helped to save the lives of an estimated 4,700 heart attack patients in four years, according to a Premier analysis of mortality rates at hospitals participating in the project. The treatments of more than 1.5 million patients in these five clinical areas at the 230 participating hospitals also included approximately 500,000 additional instances of adherence to recommended, evidence-based clinical quality measures, such as smoking cessation education, discharge instructions, and pneumococcal vaccination, during that same time frame.
Additional research by Premier using the Hospital Compare dataset disclosed that, by March 2008, HQID participants scored, on average, 6.9 percentage points higher (94.64 percent compared with 87.36 percent) than nonparticipants with respect to 19 common Hospital Compare measures.
One of the more exciting aspects of these results has to do with the level of improvement in quality of care that these hospitals have achieved throughout the project. Thirteen hospitals moved from the bottom 20 percent to the top 20 percent of hospitals in one or more clinical areas, improving quality scores by an average 28.1 percentage points in four years.
An analysis of 1.1 million patient records from the more than 250 participating hospitals found that if all U.S. hospitals were to achieve the three-year cost and mortality improvements experienced by HQID project participants for five high-risk patient populations, these hospitals could save an estimated 70,000 lives per year and reduce hospital costs by more than $4.5 billion annually. Moreover, the average cost of care per patient in these high-risk populations could be decreased by more than $1,000 within three years.
VBP and Healthcare Reform
A national VBP project is a major part of healthcare reform discussions. President Obama and a number of lawmakers, including the "Blue Dog Democrats," have advocated it. After specifically endorsing the HQID project, Sen. Max Baucus (D-Mont.), chairman of the Senate Finance Committee, proposed a VBP program similar to HQID in the committee's health reform legislation.
Still, there is skepticism surrounding the national implementation of a VBP plan. Some believe that rural and safety net hospitals-those with limited resources-will be disadvantaged. Although hospitals serving a large percentage of disproportionate share patients performed below others at the outset of the project, an analysis of HQID results for the first three years of the project showed that there was no significant relationship between urban or rural designation and either awards or recognition. The same analysis showed that even though hospitals with a large percentage of disproportionate share patients performed below others at the outset of the project, they had the same likelihood of receiving awards by the end of the third year.
Elements of an Effective VBP Program
Analysis of HQID results provides insight into ways to design an effective VBP program. Because it takes time for disparities between safety net and nonsafety net hospitals to dissipate, a national VBP program should phase in payment incentives gradually, giving disadvantaged hospitals time to adjust. Likewise, a national VBP program should provide technical assistance and resources to hospitals that fall below the national thresholds to ensure that all are equipped to succeed.
VBP payment models should reward both attainment of quality-of-care benchmarks and overall improvement. Doing so will create new opportunities for hospitals to receive incentives and continue to make rapid improvements that ultimately close the quality gap between them and others in the program.
Perhaps most important, a VBP model should provide payment policies that reward hospitals for performance and share savings with them. Under the current system, if outcomes result in fewer hospitalizations and improved overall wellness, hospitals are punished because they perform fewer procedures and receive reduced reimbursements. VBP, on the other hand, provides hospitals additional reimbursement for doing a better job of providing care, thereby addressing the perverse incentives that are hardwired into the current system.
The Need for Action
As the HQID initiative, which took place from October 2003 through September 2009, comes to a close, the logical question is, What are the next steps that our nation should take regarding VBP and healthcare reform?
It goes without saying that all patients-no matter where they live or what hospital they visit-should receive the same level of evidence-based, high-quality care. The results of the HQID project clearly indicate that a national VBP program that is based on rewarding-not punishing-hospitals with financial incentives and transparent quality reporting is an effective way to reduce care variations while driving quality improvements. It's a model that has been tested and proven to improve quality and control costs.
Given the impetus for healthcare reform and our nation's current economic crisis, hospitals should act today to help to fix the healthcare system from the front lines. It's time to move beyond demonstration tests and enact changes to our healthcare payment and delivery system so that our nation can effectively expand the quality-of-care measures and clinical conditions that are tracked and the incentives that are rewarded.
The current model of healthcare payment and delivery is clearly broken. Let's fix it now while we still can.
For more information, see Web Extra: Performance of Hospitals Participating in the HQID Project.
Susan D. DeVore is president and CEO, Premier healthcare alliance, Charlotte, N.C. (email@example.com).
Snapshot of VBP in Action Aurora Health Care, Milwaukee
Patrick D. Falvey
Prior to joining the HQID, our health system had protocols in place that had previously helped to drive quality improvement successes. Ironically, these protocols were both an advantage and a disadvantage as our organization participated in the HQID project.
As a system, we ranked in the middle of the pack at the beginning of the demonstration. Our organization was already well rounded in quality improvement, but we found through participation in the HQID project that our approach was not going to be as effective as needed if we wanted to be a top performer. For instance, if we were at 85 percent compliance with a bundled measure, our organization was geared to improve incrementally. Due to the competitive nature of HQID, we quickly learned that such incremental improvements would leave us behind the rest of the pack, so we made a number of adjustments that created an overall cultural shift in our organization and a drive toward perfection. Discussions quickly turned from "How do we improve from 85 percent?" to "How do we impact the 15 percent not in compliance with a bundled measure?"
Key changes in our organization included the standardization of quality approaches across the entire system. Upon joining HQID, we realized that we had close to 50 approaches to quality across our facilities. In response, we decided upon an "Aurora Approach to Quality," which is based on Lean processes and plan-do-study-act approaches. This shift allowed for improvements in efficiencies -time and money saved-as well as enhancements in quality.
Additionally, our president and CEO at that time made a structural change, requesting that accountability for the HQID results be transitioned from the quality departments to the administrators at our individual facilities. This increased the opportunities to accomplish quality improvement with each other versus separately. This shift of accountability to leadership has played a major role in the successes that we see as a part of this project.
Results for our HQID performance were posted quarterly and available to all caregivers. A lessons-learned database was created so successes and pitfalls could be quickly shared. This open approach to both performance and improvement activities allowed everyone to be engaged in the effort. Our results reflect this impact: During year three and year four of the HQID project, Aurora had the most hospitals of any system to receive awards for their performance.
Patrick D. Falvey, is senior vice president and CIO, Aurora Health Care,
Snapshot of VBP in Action Baptist Health South Florida, Coral Gables, Fla.
One of the major issues hospitals are facing from a clinical standpoint is sustaining resources to maintain ever-growing responsibilities for measuring quality. The challenge is not just to sustain performance at a high level, but also to drive toward perfect care. Participation and success in the HQID project has allowed us to address this issue, and we've benefited from a number of other aspects of this project.
The collaborative nature of the HQID project offered us a chance to benchmark against other leading hospitals, as well as learn from them regarding how they've improved quality. The opportunity to prepare for likely government regulations also has been invaluable. As HQID participants, we have had six years' experience with VBP, well before a plan is implemented across the country.
Ultimately, it's the opportunity for continuous quality improvements that really drives us. HQID is consistent with our goal of providing care that is of the highest quality and highest value.
This quality-improvement road has had its share of bumps for our organization. As a physician, I am well aware that some of my peers are not big believers in the pay-for-performance model. HQID measures are ultimately hospital-centric measures and not necessarily physician-centric measures. So we included HQID in a larger context of improvement across our enterprise: the need to focus on the big picture of true patient outcomes and improved collective patient care. We didn't just focus on measures, as we knew we'd run into opposition. By accommodating for the physician viewpoint, and letting our physicians own the process, the outcome has been improved patient care.
Thinh Tran, MD, is corporate chief quality officer, Baptist Health South Florida, Coral Gables, Fla.
Snapshot of VBP in Action Cleveland Regional Medical Center, Shelby, N.C.
From the beginning of the HQID project, our physicians were engaged in and concerned about improving care for our patients. Hospital case managers conducted reviews of patient charts and would ask physicians about care measures that mirrored the national benchmarks. It did not take long for improvements to take hold, creating improved outcomes for our patients and a culture of quality.
Unfortunately, the county that we serve has some of the highest levels of heart disease in North Carolina. So during our first year of participation in the HQID project, our organization focused on the process measures related to the clinical area of CHF. Our goal was to uncover the reasons why patients with CHF were readmitted to the hospital so frequently, then create processes to ensure that trend would not continue.
What we found was that many patients had difficulty understanding care instructions, they lacked adequate support systems at home, or their financial situations prevented them from taking proper care of themselves. As a result, they were frequently readmitted, and some were labeled as "noncompliant," as it appeared they were not attempting to manage their condition.
Our staff reached outside the box of traditional health care, working to develop consistent education with home healthcare companies and nursing homes. In addition, community care managers assisted CHF patients with self-management skills and identified needed resources for those patients who had previously been termed noncompliant. Often, it was as simple as helping patients understand our healthcare language-for example, alerting them to how much salt was in the soup they were eating daily. We learned that the words we choose-for example, "salt" over "sodium"-helped patients better understand their dietary needs and how to keep themselves healthy.
As a result of initiatives such as this, our adherence to the quality measures across all the clinical areas improved, many significantly. In particular, targeted discharge instructions for CHF patients improved almost 60 percent throughout our participation in the HQID project.
Elizabeth Popwell is chief ancillary executive and safety officer, Cleveland Regional Medical Center, Shelby, N.C.
Snapshot of VBP in Action Palomar Pomerado Health, San Diego
Participating in the HQID project has allowed us to enhance our focus on, among other things, evidence-based care. We have worked diligently to hardwire our key processes to ensure that our patients are receiving evidenced-based measures whenever possible, across all of our areas of care.
Three years ago, Palomar Pomerado created "best practice teams" to focus on each of the HQID project's five clinical areas. These multidimensional teams are chaired by nursing leaders and are facilitated by quality staff, and include a designated physician champion. The teams work with staff-level caregivers to identify the action steps needed to ensure that they are adhering to the evidenced-based measures that are a part of the project. Through these experiences, the teams have been able to create order sets, education tools, and checklists.
Because communication is key, we ask our staff to partake in huddles to ensure consistent communication. The need for ongoing feedback from frontline caregivers is essential, and our teams continue to create mechanisms that allow for communication among nurses, physicians, and others.
Palomar Pomerado also has made a point to publicly recognize caregivers while consistently communicating our results to all levels of the organization. Our success in administering the quality-of-care measures and in the overall project is tied to our systemwide initiatives for strategic planning. Results of these initiatives also are tied to employee incentive packages at all levels, from executives on down.
Evidence-based care has truly become an integral aspect of the care Palomar Pomerado offers our patients. We'll continue to monitor key indicators and measures to ensure success with these processes, and we are in the process of linking them to specific outcomes, such as mortality rates, length of stay, and more.
Opal Reinbold is chief quality officer, Palomar Pomerado Health, San Diego.
Publication Date: Friday, January 01, 2010