Published in the August 2012 issue of Today’s Hospitalist
ONE NEW JERSEY HOSPITAL offers a cautionary tale for hospitalists wondering how they can help their hospitals stay in the black when Medicare’s value-based purchasing (VBP) program debuts this fall.
The hospital was pretty pleased with itself; it had managed a 90% success rate in getting heart attack patients to primary percutaneous coronary intervention within 90 minutes. That’s one of the 12 clinical processes of care that will determine VBP’s financial winners and losers, along with eight “patient experience of care” measures.
A score in the 90s must be good, right? It’s good, but not good enough, said Nell Wood Buhlman, MBA, vice president for clinical compliance products at Press Ganey Associates, who spoke about the VBP program during a precourse session at this spring’s Society of Hospital Medicine annual meeting in San Diego. Scoring 90% on that measure, which is reported on the Hospital Compare Web site, puts the hospital below the national average of 93%.
A closer look at the data turned up this fact: In that particular reporting period, Ms. Buhlman explained, the hospital had 37 cases eligible for that measure; the hospital’s score meant it missed the 90-minute cutoff on just four cases. Hospitalists need to know, she said, that they could cost their hospital tens of thousands of dollars simply by letting a few patients slip through the cracks.
“At a big hospital, missing on those four patients could potentially drive a loss of $100,000,” said Ms. Buhlman. “This helps crystallize the importance of hitting the marks every single time.”
Hospitalists’ key role
Because hospitalists provide care covered by most quality and satisfaction measures, the specialty is expected to play an oversized role in value-based purchasing.
For some measures, hospitalist involvement may be indirect, encompassing their committee work and advocacy of quality initiatives. But more often, their impact will be direct, due to medications and procedures they order and how they interact with patients.
“You folks are now driving the bus,” said Peter H. Short, MD, chief medical officer and senior vice president of medical affairs at Northeast Hospital Corporation in Beverly, Mass., and the former medical director of the hospitalist program at Beverly Hospital.
Dr. Short echoed the advice of all the precourse panelists: Assume leadership. That could entail something simple like making the case to administrators for more chairs in patient rooms because doctors who sit down to talk to patients generally score higher on patient satisfaction surveys than those who stand.
It could mean developing scripts that hospitalists can follow when introducing themselves to patients, or insisting that all providers “even community-based doctors who see only a patient or two “need to follow measure-based order sets so crucial antibiotics or blood tests aren’t forgotten even once. Or for some hospitalists, it may entail becoming a medical staff officer or jockeying for a seat on the hospital board.
“Hospitalist programs that get this right are going to make themselves indispensable to their hospitals,” added panelist Patrick J. Torcson, MD, director of hospital medicine at St. Tammany Parish Hospital in Covington, La., and chair of SHM’s performance measurement and reporting committee. “This is a new era.”
Nuts and bolts
On Oct. 1, Medicare will start basing some reimbursement on how well hospitals perform on a dozen quality measures and how satisfied patients say they were with the service they received. While very small hospitals will be exempt, about 3,100 will participate in the new program. To start, hospitals will have 1% of their base DRG operating payments at risk; at full implementation in 2017, it will be 2%. According to a report from the consulting firm Press Ganey Associates, the bulk of U.S. hospitals will have between $500,000 and $850,000 of their Medicare reimbursement at risk each year under this one program.
When added to other payment reforms slated to take effect at the same time “penalties for preventable readmissions and hospital-acquired conditions, along with incentives tied to meaningful use and inpatient quality reporting “hospitals in five years may have up to 7% of their expected DRG payments at risk. According to a Thomson Reuters report, the median operating profit for all U.S. hospitals in 2011 was only 5%.
“This is what is keeping C-suites up at night,” said Ms. Buhlman. “If the quality people are chasing you down the hall asking you why you didn’t follow an order set, you know why.”
Dr. Torcson characterized the program as “a quiet revolution that’s going to transform Medicare from a passive payer to an active purchaser of higher quality, more efficient care.” Private payers have already jumped on the bandwagon, panelists noted, rolling out their own versions of value-based purchasing and putting even more reimbursement at risk.
The concept of value-based purchasing grew directly from a number of pay-for-reporting and pay-for-performance programs that began nearly a decade ago.
The difference between the new program and its predecessors, however, is one of magnitude and menace. Not only is the public reporting of data on quality and patient satisfaction now mandatory, but it is competitive. Not everyone will win; only hospitals that score at or above the national average VBP score will get back all the money that the Centers for Medicare and Medicaid Services (CMS) will withhold for each hospital under the terms of the program. Press Ganey predicts that half of all hospitals will lose money under the VBP program.
Another hurdle: Just because you earn money one year doesn’t guarantee that you will the next. That’s because the program will change from year to year, with new domains being introduced and new measures added to existing domains.
“The fluid scope of the program is the greatest challenge,” Ms. Buhlman observed. “This is not something that you can just ‘get it and forget it.’ ” After all, she noted, the goal is to incentivize continuous quality improvement.
Benchmarks and thresholds
For nearly a dozen process-of-care measures, the “benchmark” (the mean of the top decile of all hospital scores, or the best performers) is 100%. The CMS plans to retire these “topped out” measures at some point as it adds new ones.
The “threshold” is another important number for hospitalists to understand, Dr. Torcson said. The threshold is the minimum acceptable score, and it is set at the 50th percentile of scores in a benchmark period.
Generally speaking, when hospital scores are below the thresholds, those hospitals risk losing their incentive payment. The further below the threshold, the greater the risk of loss. As of October 2011, Dr. Torcson said, some measures “such as “Pneumonia 3” (blood culture taken in the ED prior to the first administration of antibiotics) and some surgical care improvement project measures, such as the use of prophylactic antibiotics, have thresholds greater than 95%.
“It’s pretty scary how high the thresholds are being set,” said panelist Nikolas Matthes, MD, PhD, MPH, vice president of research and development at Press Ganey Associates. “A lot of hospitals thought they were a top performer because they were in the top 10%. But suddenly, the thresholds show that you have to be in the top 5%.”While you may be improving, he added, “everyone else is improving too.” An individual hospital’s level of performance needs to keep pace with or outpace the comparison group’s rate of improvement.
A moving target
Threshold levels on patient satisfaction measures, taken from the CMS’ Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) survey, tend to be lower. They range from about 60% on explaining new medications to 80% on doctor communication.
As a result, hospitals have more room to improve in these areas. But as panelists explained, challenges abound here too. In particular, the only patient answer that improves a hospital’s VBP program score on surveys is “always,” not “usually” “let alone “sometimes.”
Take the physician communication dimension, which is made up of three questions. When a hospital scores 75% for physician communication, the average percentage of top box (“Always”) responses for those three questions was 75%.”What we don’t see online is that another 17% of patients responded ‘usually,'” Ms. Buhlman says. “What can you do to get that 17% checking the ‘always’ box instead? It can be very difficult.”
Another challenge is the fact that VBP is designed to be a moving target. Not only will topped-out measures be retired and new measures added, but how much weight each measure has toward the total score will change.
For instance, patient experience measures will together count for 30% of this year’s total score, with process-of-care measures making up 70%.
But for fiscal year 2014, VPB is adding outcomes measures, which will account for 25% of the total score; process-of-care measures will drop to only 45% and patient experience will stay at 30%. (Outcomes measures include risk-adjusted 30-day mortality for AMI, heart failure and pneumonia patients.) VBP-watchers expect a fourth category “efficiency of care “to be added in 2015.
Hospitalists also need to know that VBP payments (and withholds) don’t just apply to revenues from those clinical areas being measured, such as heart failure.
“They apply to revenue from all DRGs, so they’re going to have a real impact on hospital revenues,” Dr. Torcson said. “Hospitals that don’t do well will have less revenue overall, with less funds to invest in things like hospital medicine subsidies.”
Lessons from top performers
How can hospitals succeed under VBP? Panelists suggested looking at top performers under the program’s trial runs.
Consider Beverly Hospital, where Dr. Short is chief medical officer. According to Press Ganey, the 220-bed hospital has ranked in the top 10% of hospitals in all four CMS measure sets: acute myocardial infarction, congestive heart failure, pneumonia and surgical care improvement.
But as Dr. Short pointed out, Beverly Hospital’s experience with the heart failure measures has been a challenge, as it has been for many hospitals. Unlike MI or pneumonia, a heart failure diagnosis listed at the time of discharge may not be the diagnosis on admission, making it tough to focus performance improvement programs on a problem known only in retrospect. As a result, some patients aren’t initially treated according to heart failure protocols, which raises the risk that a hospital under VBP could be dinged on heart failure measures.
What Beverly Hospital did was confront the problem head-on. Staff actively search patient rolls for possible heart failure patients, ordering BNP tests to look for elevations that might signal higher risk. They then set up a system to track potential heart failure patients, with case managers visiting every one of them in the hospital.
“We ID patients on the front end so we are prepared on the back end,” explained Dr. Short. The goal is to make sure doctors and nurses are on top of these cases so that no one slips through the cracks.
Tracking patients in real time and then directing staff and resources to getting patients the care spelled out in process measures is a key way top performers distinguish themselves, said Dr. Matthes. According to Press Ganey research, top-scoring hospitals also tend to set goals with deadlines, have a multidisciplinary quality improvement department that meets frequently, and enforce measure compliance.
“That enforcement has to be timely and specific,” Dr. Matthes said. “It helps a lot if it comes from the head of service, not just from the quality department.”
Top-scoring hospitals are also distinguished by physician and executive engagement. Dr. Matthes notes that the highest scoring hospitals have physicians leading their practice improvement initiatives, championing measure sets, writing guidelines, and teaching and counseling other providers about best practices.
That engagement has to flow from a real interest in quality improvement, not a fear that reimbursement may be cut. “What top performers don’t say,” said Dr. Matthes, “is ‘This is a mandate. You have to do it. The CMS says so.’ ”
Deborah Gesensway is a freelance writer who covers U.S. health care from Toronto.
The top 20
WHAT MEASURES WILL COUNT toward a hospital’s score in Medicare’s value-based purchasing program that kicks off this October? For this year, scores will be based on 12 clinical measures, as well as on eight patient satisfaction measures. Here they are:
- Fibrinolytic therapy within 30 minutes of hospital arrival
- Primary PCI within 90 minutes of arrival
- Discharging heart failure patients with written instructions or educational materials
- Performing a blood culture in the ED before initial antibiotics
- Initial antibiotic selection in community-acquired pneumonia patients
- Prophylactic antibiotics within an hour before surgical incision
- Prophylactic antibiotics for selected surgical patients
- Prophylactic antibiotics discontinued within 24 hours after surgery
- Cardiac surgery patients with controlled 6 a.m. postop serum glucose
- Surgery patients who take a beta-blocker before arrival receiving perioperative beta-blockers
- VTE prophylaxis ordered for surgery patients
- Surgery patients receiving VTE prophylaxis within 24 hours before and 24 hours after surgerySatisfaction measures
- Nurse communication
- Doctor communication
- Clean and quiet room
- Staff responsiveness
- Explanations of new medications
- Pain management
- Discharge information
- Overall rating for the hospital