Published in the August 2013 issue of Today’s Hospitalist
WHEN RESEARCHERS at MedStar health, a multihospital system in the Baltimore-Washington area, decided to look at deep vein thrombosis prevention to see whether practices at its eight hospitals were evidence-based and cost-effective, they expected to find some variation across institutions. What they didn’t expect was that the divide between the hospitals would be so great “and geographically based.
The study, which appeared in the April issue of the Journal of the American College of Surgeons, found that the system’s four hospitals in Washington primarily used low molecular weight heparin (LMWH) and sequential compression devices (SCDs) as prophylaxis in both the OR and on the units postoperatively. But the four hospitals in Baltimore used unfractionated heparin (UFH) in most cases and steered clear of SCDs except for single-day surgeries.
Despite the substantially different practice patterns between the two sets of hospitals, their rates of hospital-acquired DVT were essentially the same. Per-hospital rates ranged between 0.8% and 1.8%, with an average rate of 1%.
What was different, however, was the cost. The Washington hospitals spent more on DVT prevention efforts ” roughly $4 million more “per year.
“The big surprise was the practice differences in our own system between Baltimore and Washington, when they’re only 40 miles apart,” says John Kirkpatrick, MD, MBA, a professor of surgery at Georgetown University and a principal with The Surgical Advisory Group, a surgical research and consulting group, and lead author of the study. “When we looked at cases managed with either unfractionated heparin or low molecular weight heparin as an independent variable, outcomes were slightly better with unfractionated heparin.”
The study was designed to use value-based analysis, a methodology that compares usual practice to best evidence-based practice. The goal is to identify potential cost differences if a usual practice was replaced with the best practice. Dr. Kirkpatrick’s group chose to evaluate DVT prophylaxis because the system’s eight hospitals treat more than 120,000 patients for the condition each year.
“This is what the Affordable Care Act is asking us all to do: use value-based decision-making to see if value can be improved and costs reduced without compromising quality,” says Dr. Kirkpatrick. “DVT prophylaxis is a good practice to look at because it’s so common.”
The biggest cost difference between the two practices was the use of UFH rather than LMWH. Switching to UFH for all prophylaxis in the eight hospitals would save $2.9 million annually. Getting rid of SCDs for all but short-stay cases “where patients stay less than a day “would save an additional $1.1 million.
The literature supports both changes, Dr. Kirkpatrick explains. Based on the most recent evidence, LMWH is no more effective than UFH for DVT prophylaxis, although it is more effective for treating an existing DVT or pulmonary embolism. Dr. Kirkpatrick adds that while LMWH has a more predictable effect than UFH and is easier to administer, it is contraindicated in patients with renal impairment and can be more challenging to reverse.
Sequential compression devices are notoriously cumbersome (and potentially dangerous) both in the OR and on the floors and are contraindicated in several types of patients. The devices have also proven to be minimally effective in preventing DVT.
Based on the researchers’ evaluation of practice patterns and their meta-analysis of the literature, the system adopted a single best practice for all eight hospitals. That includes using UFH rather than LMWH in the hospital for prophylaxis, except in patients who are admitted with or develop a PE, and those with unstable angina or an ST segment elevation myocardial infarction. The protocol also includes discontinuing SCDs in all but short-stay patients or those undergoing outpatient procedures.
“Usual practice” pitfalls
How the practice patterns in the two hospital groups diverged over time makes an interesting case study in itself. The variation was largely due to process and timing, Dr. Kirkpatrick explains.
“Each MedStar hospital committee came up with an algorithm for what it thought was the best practice, and that became the usual practice,” he says. However, the Baltimore hospitals repeated their literature-review process more frequently. “The Washington committees were using 2002 data, while the Baltimore folks used 2006-07 data.” As a result, the Baltimore hospitals had already transitioned to what would become the system’s new protocol: removing SCDs in surgical inpatients and converting primarily to UFH rather than LMWH for prophylaxis.
“The key message of the study is that you have to be contemporaneous with usual practices,” Dr. Kirkpatrick says, “or suddenly, your usual practice is no longer the best practice.”
The new protocol has met with relatively little resistance, although some surgeons have expressed concern about the more limited role of LMWH. (Using UFH in the OR, Dr. Kirkpatrick explains, does require close pre-procedure coordination with the anesthesiologists because local anesthetics must be in place before UFH is administered.) Once those surgeons reviewed the literature, they became more receptive.
The researchers did have some interesting behind-the-scenes conversations with surgeons, chief medical officers and service chiefs, including the chiefs of medicine. For one, some were not aware that SCDs were continued after surgical patients went to the floor. The researchers also found a range of opinions about exactly when prophylaxis should begin, with more than 60% thinking that SCDs alone constituted sufficient prophylaxis.
Cost awareness varies
In terms of cost awareness, researchers found that chief medical officers and service chiefs had a reasonable handle on the relative costs of the two heparin drugs and the tab for SCDs. The surgeons, however, were less informed.
“The surgeons were quite surprised that there were so many dollars’ difference between the two protocols,” says Dr. Kirkpatrick. “For the most part, they embraced the opportunity to save money.”
The DVT initiative is one of several ongoing value-based analyses of common tests and practices within the MedStar system. Dr. Kirkpatrick recommends this relatively straightforward way to launch and sustain best-practice efforts: First, establish a best-practice committee within each clinical department and give a high priority to value-based analyses in each committee.
“Every single department needs a best-practice committee just like the M&M committee,” he suggests. “These committees could handle four or five very common clinical initiatives a year and look at the costs.” Such analyses, he adds, “should happen about every five years because that’s how much time it takes for a change to occur.”
Hospitalists are in a good position to facilitate such efforts because they’re involved in so many different clinical areas.
“Hospitalists should push hard for identifying common clinical initiatives,” says Dr. Kirkpatrick, “and bring pressure on other departments to do the same.”
Bonnie Darves is a freelance health care writer based in Seattle.