On quality measures, hospitals treating larger numbers of the poor did not perform as well as hospitals serving more affluent populations back in 2003. But by 2006 that changed.
When offered a financial incentive from the federal government's three-year pay-for-performance incentive demonstration project, mortality rates went down and many quality measures improved closer to the level of their counterpart hospitals treating more wealthy patients. The poorer hospitals were able to narrow the gap.
That's the latest news from the third year of the Premier Hospital Quality Incentive Demonstration (HQID) project, which has tracked performance and outcome statistics for 250 hospitals in 36 states on 30 quality measures. The measures deal with care of patients with five conditions: Community-acquired pneumonia, heart attack, heart failure, coronary artery bypass surgery, and hip or knee replacement.
Safety net and non-safety net hospitals are evaluated in both urban and rural parts of the country.
"We are delighted that the demonstration has produced evidence that value-based purchasing can improve the quality of care for Medicare beneficiaries," says Mark Wynn, director of payment policy demonstrations for the Centers for Medicare and Medicaid Services.
Pay-for-performance incentives to improve quality can be applied across the country with similar success, says Susan DeVore, Premier's president and CEO.
"Current value-based purchasing proposals are structured appropriately to allow all hospitals to succeed, regardless of size, location or patient/payer mix," she says. "We are confident that if our recommendations are followed, VBP will be an effective national reform that will improve quality, reduce variation and avoid unnecessary costs."
The HQID project is the basis for CMS' proposal to Congress for a national pay-for-performance program and has been mentioned by members of the Senate Finance Committee and the Democratic Blue Dog coalition.
However, according to Premier officials, concerns were raised that if imposed as a national standard, the pay-for-performance programs would reward only wealthier hospitals. This research refuted that belief by discovering that hospitals regardless of their size or location can succeed attaining pay for performance.
For example, the research found that overall, safety-net hospitals treating heart-attack patients had an 88.8% quality score in 2003, but in 2006 had improved to 95.2% or by 6.4%. Non-safety-net hospitals started out higher, at 90.7% but improved only 5.5% to 96.2%.
CMS has handed out $24.5 million to top performers, and this year, 112 of the 250 hospitals scoring in the top 20% in each of the project's five clinical areas will receive $7 million, a total of 206 awards.
The top 10% scoring hospitals receive a 2% incentive payment per clinical area. Hospitals between 11 and 20% receive a 1% incentive payment. Also, all hospitals in the top 50% of each clinical area receive public recognition on the CMS Web site.
The amounts of each hospital's award will be made public later this month.
Jack Garon, MD, chief medical officer for Mount Sinai Hospital in Chicago, a 280-bed disproportionate share hospital that treats large numbers of the poor, says that in reality, the money was not what propelled his hospital's efforts to improve quality.
"The money is not enormous or staggering. In fact, it isn't enough of a motivator to do this just for the money," he says. "You're also motivated to be able to say you're in the top 10%."
At his hospital, making sure that quality measures were consistently met became like a competitive exercise that was a dynamic and transparent look at how staff was performing.
"We created a scorecard that enabled us to look at these measures at a glance," he says. "We color-coded it. Being in the top 10% would be green. Yellow was top 50%, and red meant we were in the bottom 50%."
Safety net hospitals, also called DSH or "Pickle" (after a Texas congressman) facilities, are defined as those qualifying for disproportionate share money under one of two definitions. For example, they can qualify by being in an urban area with more than 100 beds and receive 30% of net inpatient revenue from state and local revenues for indigent care.
In the category of heart failure, safety net hospitals started out low on adherence to quality measures, 68.4% but improved to 86.3% after three years. Non-safety net hospitals also started out low, at 72.4%, and improved to 88.6%.
In hip and knee replacement, safety net hospitals started at 86.8%, but improved to 94.7%, a 7.9% improvement, while non-safety net hospitals started at 89% and went to 95.5%, a 6.5% improvement.
Scores are achieved by, for example, making sure patients with an acute myocardial infarction receive an aspirin at arrival and that aspirin is prescribed at discharge.
For patients with pneumonia, appropriate care would be in making sure each patient received his or her initial antibiotic within four hours of arriving at the hospital door. For patients with a hip or knee replacement, all would be assessed for their risk of developing a hematoma or hemorrhage.
Also, inpatient mortality rates and whether the patients had to be readmitted for the same illness within 30 days are both scored.
"The research found that hospitals, regardless of size or location, can succeed in value-based purchasing."
The success of value-based purchasing programs like this one organized by CMS and Premier are destined to become much more widespread with a variety of initiatives under discussion in health reform efforts.
Officials for both CMS and Premier suggest three lessons for how to implement a successful national pay-for-performance program:
- Phase in payment incentive programs gradually to give hospitals serving higher numbers of uninsured and the poor more time to adjust.
- Provide technical assistance and resources to hospitals that fall below national thresholds.
- When a hospital's performance scores dramatically improve, reward that dramatic improvement, even if the hospital continues to fall below goals, to create incentives to continue to make rapid improvements.
Overall, improvements in quality of care saved the lives of an additional 2,500 patients who came to the hospital with a heart attack, and all patients received approximately 300,000 additional recommended evidence-based clinical quality treatments. They included smoking cessation, discharge instructions and pneumococcal vaccine.
The demonstration project has been extended another three years and will be extended to new measures, such as consumer assessments, surgical care improvements, and care of patients with ischemic stroke.