Critical Path Network

Demonstration project claims $1 billion in potential savings

Premier, CMS collaborative projects 3,000 fewer deaths

Data from Premier Inc.'s pay-for-performance demonstration project with the Centers for Medicare & Medicaid Services (CMS) indicate that improving the care of pneumonia and heart bypass patients alone can save as much as $1 billion a year, as well as thousands of lives. Projected outcomes include 3,000 fewer deaths, 6,000 fewer complications, 6,000 fewer readmissions, and 500,000 fewer days in the hospital.

Through the demonstration project, Premier collects a set of 33 quality indicators from more than 250 hospitals across the country. Because these indicators are not collected from all hospitals, Premier researchers extrapolated national implications using statistical methods. If patients receiving a smaller percentage of widely accepted care measures had instead received most of the measures — 76% or more — hospitals costs would have been approximately $1 billion lower in 2004, according to Premier's analysis.

"The first year the project collected data was 2003," says Denise Remus, PhD, RN, Premier's vice president of clinical informatics. Formerly, she was senior research scientist at the Association for Healthcare Research & Quality (AHRQ), where she was responsible for the development of AHRQ's quality indicators. "In spring of last year, we closed the first-year data set, and results were released in November 2005. When I joined Premier, I took the first year of the demonstration project and began to do an analysis of the relationship between cost and quality," she says.

As she began her analysis, Remus recalls, she did not see a relationship between individual process measures and outcomes. "This actually made sense," she says. "Several of the measures have to do with discharge. For example, with heart failure, one of the measures is discharge planning; in AMI, one is beta-blocker prescribed at discharge. So, it made sense to not see a relationship between individual process measures and outcomes during stay."

While her "gut" told her there was a relationship, Remus couldn't see it. "I stepped back and said, 'Let me ignore where the patient was taken care of and merge all the patients together and look at the quality on the patient level.'"

In essence, Remus says, she relied on a "pathways of care" approach. "What that says is, we know we have patients who are eligible for a certain number of measures. In my mind, high quality is when the patient gets everything they are supposed to get. So, instead of just looking at whether the patient was getting aspirin, we looked at a proxy — patient process measures — and the rate of how many interventions patients were eligible for and how many they actually received."

In other words, a patient who goes through the health system and receives all the recommended experiences around the measure is receiving highly reliable care. Accordingly, the patients were placed in four groups from low to high, in segments of 25% reliable care — the highest being 76% or more. "Then, we looked within each of the clinical conditions [pneumonia and heart bypass]," says Remus.

Using the data

In order to come up with the estimated savings, she says, "We took the analysis we had done, and then estimated what the impact might be nationally, based on the Healthcare Cost & Utilization Project database of AHRQ — and identified how many patients were discharged in pneumonia and heart bypass. Then, we looked at the total number of patients discharged in those groups and estimated how many would be in the low, medium, and high categories based on how the patients in our projects were distributed."

Remus thinks her numbers are actually conservative, "because in our year of data, our hospitals tended to have higher quality than other hospitals in the country."

These data carry an important message for hospital quality managers, says Remus. "As Medicare moves into value-based purchasing, or linking payment to quality, hospitals will have to manage their costs better," she says. "That's the group we want to address."

She continues: "We believe we've established clear evidence that higher quality can improve outcomes and save costs," Remus asserts. "We've also found that it doesn't cost more to get high quality, which, of course, reduces LOS and complications — all of which increase the costs of care for patients."

Readiness program

To help hospitals move in the right direction, Premier has launched a pay-for-performance readiness program, which is provided free of charge on the company's web site ( The program includes a pay-for-performance calculator.

"Anyone who has a Medicare provider number can enter it," says Remus. "Then, in the next screen it asks them to enter a numerator and denominator for the 18 process measures in AMI, heart failure and pneumonia."

Once those numbers are entered, the site will come back with a report that links cost and quality. "Using their information, we calculate an 'appropriate care score' at the hospital level," Remus explains. In other words, the report will show how many patients treated at your hospital for AMI and pneumonia received appropriate care. "Then, we model what their potential cost savings would be," she adds.

"This just helps support awareness that there is a changing reimbursement environment out there," she continues. "Payments can potentially be reduced or increased based on quality. In that environment, hospitals may be at risk if they do not provide high-quality care, so they need to take a look at the reliability of their care system."