Do hospitalists really provide better care?
According to one skeptic, research on hospitalist quality doesn’t ask the right questions
Keywords: Research on hospitalist impact meets skepticism - a call for improved measures of quality care
by Deborah Gesensway
Published in the November 2009 issue of Today's Hospitalist
IN THE LATEST SALVO in the long debate about whether hospitalists improve the quality and safety of inpatient care, University of Alabama general internist Robert Centor, MD, has once again been cast as the skeptic.
Along with his colleague Benjamin B. Taylor, MD, Dr. Centor—known to many from his DB’s Medical Rants blog and frequent writings in Medscape—penned his most recent editorial the topic. In the Aug. 10 issue of Archives of Internal Medicine, that editorial poked holes in the latest published study (also included in that issue),
which concluded that hospitals with hospitalists provide better quality medicine than those without. The editorial adds to Dr. Centor’s growing record of questioning whether hospital medicine translates into safer care and greater patient satisfaction.
|“I’m much more willing to say out loud what some hospitalist leaders have tried to keep quiet.”|
–Robert Centor, MD
University of Alabama Huntsville Regional Medical Campus
“People keep giving me a chance to express my opinion,” says Dr. Centor, who is associate dean at the University of Alabama’s Huntsville Regional Medical Campus and has served as a ward attending for 30 years. “I’m much more willing to say out loud what some hospitalist leaders have tried to keep quiet.”
What research fails to take into account, he points out, is this oft-repeated maxim: “Once you’ve seen one hospitalist program, you’ve seen one hospitalist program.” In an interview with Today’s Hospitalist, Dr. Centor explains that his concern is not with hospital medicine per se or hospitalists, who he believes are here to stay. Instead, he keeps pointing to the variability—of both clinical care and group organization—within the hospital medicine model. What he found misleading in the latest Archives research on hospitalists, for example, is equating compliance on performance measures with quality and hospitalists with effectiveness.
Quality of care, he argues, is much more “multidimensional,” and it’s a vast “oversimplification” to say that hospitalists improve care. The likely conclusion, says Dr. Centor, is that some do and some don’t.
What’s important, he adds, is determining how high-performing groups operate, and how engaged they need to be in hospital processes and systems to provide safer care.
According to Dr. Centor, a major flaw in most studies of hospitalist vs. non-hospitalist care, including the most recent Archives research that he weighed in on, is that hospitalists self-identify. As a result, he points out, there are no standards of what constitutes a hospitalist.
Some hospitalists work 100% of the time in inpatient care, but others may attend on inpatient wards only a few weeks or months a year. Dr. Centor holds his own experience up as an example: While he rounds with residents 150 days a year, he doesn’t call himself an academic hospitalist. (He notes, however, that he doesn’t contradict people who refer to him as an academic hospitalist.) “I call myself an academic general internist,” Dr. Centor says.
In his mind, time counts when it comes to honing the skills you need. “We don’t know what the magic number is,” Dr. Centor explains, “but you need a certain amount of patient exposure each year to maintain your skills and knowledge.” One good example is MRSA, he points out, which hospitalists picked up on more quickly than internists who were coming into hospitals only periodically.
The way hospitalist groups are organized and managed also affects the quality of care. But exactly what types of organizational structure lead to better quality remains an unanswered question. In essays he wrote for the June 23, 2008, Archives, Dr. Centor noted that organizational variability among hospitalist groups has a major impact on patient safety, satisfaction and trust for physicians.
Quality is not just compliance
If hospitalists really want to prove their worth, Dr. Centor says, studies need to look at how to put together a good hospitalist group, what type of “group sociology” might cause newly-hired residents to leave after only a year or two, and what type of hospitalist leadership makes a difference in care quality.
“A really strong hospitalist group,” Dr. Centor points out, “infiltrates the entire hospital culture and works with the entire hospital—the nursing staff, the physical therapists, everyone—to try to figure out how it can make lengths of stay shorter and more efficient, with fewer errors.”
The Archives study published this August looked at more than 3,600 hospitals across the country. Researchers found that regardless of “size, location, ownership type and staffing availability,” hospitals with hospitalists posted better performance scores on some important quality indicators, including Hospital Quality Alliance measures for acute myocardial infarction and pneumonia, as well as in counseling and prevention.
“Hospitalists are likely to be one of several factors that contribute to high quality care,” the study authors concluded.
In their editorial, Drs. Centor and Taylor agreed. But they pointed out that the data used by the researchers were “not persuasive enough” to conclude that “hospitals with hospitalists are associated with better performance.”
A key problem, they wrote, is conflating “quality” with “adherence to disease performance measures.” In addition, the fact that a hospital has hospitalists on staff may not cause—but only correlate to—improved adherence to practice guidelines, not to mention better quality.
“Without patient and physician level data,” they wrote in the editorial, “the analysis has too many potential confounders.” Alternative explanations, such as higher nursing staff ratios or better hospital systems and infrastructure, may have more of an impact than the presence of hospitalists.
Or it may be that more progressive hospitals invest in hospitalists as well as in other types of infrastructure that help improve quality, such as information technology or more nursing.
What’s the right metric?
The real test for hospitalists, Dr. Centor adds, is how much they improve safety. “That’s where hospitalists can really make a difference to improve morbidity and mortality by reducing MRSA or C. diff rates,” he explains. “Some hospitalists are having an impact on safety, and all should be striving to do that.”
Research should be refocused, he says, to understand how programs should be structured and how hospitalists should be utilized within the hospital to improve safety. Studies of various groups’ safety records in multiple sites, for instance, would begin to yield “best practices” for hospitalist group organization.
“The next step for hospital medicine is to try to develop standards and convince hospital administrators that investing in the right type of hospitalist leadership and structure actually does pay off,” Dr. Centor says. “You can’t just hire hospitalists. You have to invest in the system.”
While some studies are beginning to provide answers about the optimal number of patients a hospitalist can take care of each day, he says that research is needed on how much hospitalists should be involved in hospital processes to improve patient safety. Research should, for instance, be able to show the impact of hospitalists in spearheading safety initiatives that prevent “never” events.
“We need to know what best practices are,” says Dr. Centor, “so that hospital administrators will buy into them.”
Deborah Gesensway is a freelance health care writer who covers U.S. health care from Toronto.