Strengths and weaknesses

Major teaching hospitals have the lowest mortality rate of all hospital categories, according to a new study conducted by Truven Health Analytics. But they also have significantly higher incidence of adverse patient-safety events and complications.

Those seemingly incongruent findings might be explained, in part, by considering the patient population that major teaching hospitals serve and their capacity for addressing complications, says Vinita Bahl, director of the Office of Performance Assessment and Clinical Effectiveness for the 919-bed University of Michigan Health System, Ann Arbor.

“Major teaching hospitals treat patients that are more complex and more severely ill than nonteaching hospitals, and these patients are at higher risk for complications,” Bahl says in an e-mail. “Though it seems like a paradox, the data suggest that major teaching hospitals are equipped to … effectively rescue patients through timely recognition and management of complications once they occur.”

Truven, which recognizes high-performing hospitals through its annual 100 Top Hospitals list, conducted the analysis exclusively for Modern Healthcare to explore whether hospital size or teaching status affects performance on key quality and cost metrics. The takeaway: No hospital category has consistently superior performance compared with the other classes, but significant differences in performance were found in almost every measure.

According to Truven, major teaching hospitals are those with at least 400 beds in service and a high intern- and resident-per-bed ratio, while teaching hospitals have 200 or more beds and a less-intense training component. Large community hospitals are those with 250 or more acute-care beds; medium-sized community hospitals have 100 to 249 beds; and small community hospitals have 25 to 99 beds.

“We're really demonstrating that each category of hospitals has a different set of issues,” says Jean Chenoweth, Truven's senior vice president of performance improvement and the 100 Top Hospitals program.

David Foster, principal investigator for Truven's Center for Healthcare Analytics, says the study is noteworthy because it is one of the first to examine the effect of structural characteristics—in this case, size and teaching status—on hospital performance. With the increasing focus on value in healthcare delivery, the findings might provide clues to where hospital leaders need to focus their efforts. For example, small and medium-sized community hospitals might need to work on reducing mortality, while teaching hospitals need to improve patient safety.

That said, structural characteristics do not determine a hospital's destiny. “There are hospitals that have recognized the issues specific to their class of hospital and developed ways to achieve 100 Top national benchmark performance,” Chenoweth says.

Kyle DeFur, president of 873-bed St. Vincent Indianapolis Hospital, a seven-time 100 Top teaching hospital, believes management trumps structural factors in determining the value of care a hospital delivers.

“I think it has more to do with culture and the focus of the organization than it has to do with the size of an organization or even the kind of organization that it is,” DeFur says. “You can make the argument that the larger the ship, the more time it takes to turn it. But I don't think (success) is as size-related as it is tied to the resources that are being dedicated to move the dial on those value-based measures.”

Truven analyzed the performance of 2,922 hospitals bucketed into five classes defined by the number of acute-care beds in service and teaching status: major teaching, teaching, large community, medium community and small community. Data sources were Medicare Provider Analysis and Review (MedPAR) public use files for 2010 and 2011, the CMS' Hospital Compare database and Medicare cost reports. Noteworthy findings include:

• Average length of stay increases with facility size. With an ALOS of 5.31 days, major teaching hospitals had significantly longer lengths of stay than any other class. Small community hospitals had the shortest, at 4.83 days.
• Major teaching institutions have significantly higher inpatient expenses per discharge—an average of $7,303—than the other groups. The analysis found that small hospitals had the second-highest inpatient expense per discharge, $6,428. Large community hospitals had the smallest average figure at $6,092.
• Large community hospitals had the highest operating profit margin—7%—in 2011, while small community hospitals had the lowest at 3.5%. Major teaching hospitals had a 4.5% margin.
• All classes of hospitals comply with the CMS' core measures at between the 95th and 97th percentile, meaning that more than 95% of all patients are treated with evidence-based level of care for those measures.

Truven investigated hospital performance on a single Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) survey question: Would you recommend this hospital to your family and friends? Smaller teaching hospitals had the highest score, on average, among the five classes, while medium-sized community hospitals scored the lowest.

Ted Townsend, president and CEO of 363-bed UnityPoint Health-St. Luke's Hospital in Cedar Rapids, Iowa, reads the Truven analysis as good news for smaller teaching hospitals, which outperform major teaching hospitals on every quality and cost metric in the study.

That means smaller teaching hospitals could have a natural advantage under the CMS' value-based purchasing formula, which ties payment rates to quality, cost and patient experience measures.

“We have some education going on, but a far greater degree of experience at the bedside on a day-in, day-out basis,” he says. “And that's the sweet spot from a cost standpoint as well, because we don't have to have all the infrastructure of a major academic medical center, but we do have a lot of the value that comes from teaching people.”

The University of Michigan Hospitals & Health Centers has been named to the 100 Top Hospitals list eight times, and Bahl agrees that the Truven analysis might provide hints about how major teaching hospitals are likely to fare under the government's value-based purchasing program. Although Truven's analysis shows that major teaching hospitals perform relatively well on the one willingness-to-recommend HCAHPS measure, academic medical centers generally tend to score lower than other hospital classes on the full HCAHPS survey, Bahl says. Further, the Truven analysis shows that major teaching hospitals had no advantage over other classes in terms of core-measure performance.

“In value-based purchasing, where even small differences in measurement results can translate to large differences in overall program performance, major teaching hospitals may not thrive, but they can be competitive,” Bahl says.

Medium-sized community hospitals had two potentially worrisome findings in the Truven analysis: At 13%, their 30-day mortality rate was the highest of any class while their willingness-to-recommend HCAHPS score was the lowest. However, the medium-sized hospitals' mortality rate was just one-tenth of a point higher than the small-hospital category.

With between 100 and 249 beds, hospitals in that class are challenged to have staff dedicated to patients based on their medical situation, such as orthopedic surgery versus cancer surgery, says Dave Graebner, president of 130-bed Aurora Sheboygan (Wis.) Memorial Medical Center. The need to congregate patients on medical/surgical floors may explain the patient-satisfaction score.

Aurora Sheboygan has made the 100 Top Hospitals roster five times, proving that medium-sized facilities can overcome challenges identified in the study. Graebner cites two advantages hospitals in that class may have as they seek to deliver high-value care: the ability to focus on a well-defined patient population—in his case, Sheboygan's 145,000 residents—and the ability to work closely with a relatively small medical staff to provide patient care at the right place at the right time.

Aurora Sheboygan's medical staff has only about 150 physicians, all of whom are employed by the Aurora system. Graebner says that small number of physicians and physician employment work together to support the hospital-physician alignment needed to succeed in the CMS' value-based purchasing program.

The different strengths and weaknesses of the various hospital classes could reflect the different roles they play in the healthcare system, Chenoweth says. For example, the fact that small and medium-sized community hospitals have higher mortality than larger hospitals might not reflect the quality of care they deliver but rather the lack of hospice services in their communities.

“We need to do more research to understand why those mortality rates are higher … but my suspicion is that they are providing some end-of-life care,” she says.