Home Analysis Evidence mounts in favor of hospitalists for sicker patients

Evidence mounts in favor of hospitalists for sicker patients

November 2007

Published in the November 2007 issue of Today’s Hospitalist

Over the years, both the medical literature and word-of-mouth anecdotes have shown that a big benefit of hospitalists is reducing patient length of stay. But until now, there has been little in the way of data to answer a related “and even more interesting “question: Are there particular patients or diseases where hospitalists truly earn their stripes?

A new study from New York’s Montefiore Medical Center in the Sept. 24, 2007, Archives of Internal Medicine answers that question with a resounding “yes.” Researchers found that while hospitalist care reduces length of stay for all patients, it makes the biggest impact on very sick patients, particularly those individuals who have a complicated discharge.

“The sickest 25% of all admissions account for the majority of benefit that hospitalists offer,” explains William N. Southern, MD, MS, director of hospitalist services and associate medical director for the Weiler Division Hospital of Montefiore Medical Center and lead author of the article.

Data from the study show that hospitalists helped slash length of stay for the sickest of the sick by nearly two days. When these patients were cared for by hospitalists, they were hospitalized for just over 8.7 days. Patients treated by nonhospitalists, on the other hand, spent just over 10.5 days in the hospital.

Difficult diagnoses

The study compared care at an academic medical center “Montefiore’s Weiler Division Hospital “provided by two groups. One was led by a full-time teaching hospitalist; the other was led by a faculty member who is on the teaching service only one or two months a year.

Researchers found that patients cared for by hospitalist-led teaching teams stayed in the hospital on average between one-half and a full day less (specifically, 0.86 of a day less) than those cared for by the teaching teams led by nonhospitalist attendings. But the most interesting results appeared when the study drilled down and looked at the sickest patients on the wards.

The diagnoses that posted the most dramatic length-of-stay differences included stroke (a difference of 3.86 days), sepsis (3.7 days), pneumonia (1.38 days), congestive heart failure (1.18 days), asthma/COPD (0.93 days) and urinary tract infection (0.83 days).

“Our conclusion was that there are specific diagnoses where hospitalists appear to have a large effect,” Dr. Southern says. “These are diagnoses that require close clinical monitoring.”

These are also the conditions “where you need to make more than one decision a day,” he explains. The advantage of being on-site all the time is that hospitalists “can reassess people, adjust their treatment regimen and let them go home if they are feeling better, rather than the more traditional model where a physician makes only one or two visits a day.”

Complex discharges

Not surprisingly, the study found that diagnoses of conditions that don’t require such intensive monitoring, like acute MI or fluid/electrolyte disorder, produced less dramatic length-of-stay results. With those conditions, hospitalists cut length of stay by only about one-third of a day.

In addition to analyzing length-of-stay data by looking at specific diagnoses, Dr. Southern measured patient acuity as it related to discharge.

When researchers looked at discharge disposition, they found that hospitalists had the greatest effect on patients who had to be discharged to a skilled nursing or rehabilitation facility. Hospitalists had a more modest effect on patients who were discharged home with home care. They also made less of a difference in length of stay for patients discharged home to care for themselves.

“The more complicated the discharge plan became, the larger the effect hospitalists had,” Dr. Southern explains. “For instance, hospitalist care resulted in a two-day shorter hospital stay for patients ultimately transferred to a skilled nursing facility than when these patients were cared for by nonhospitalist attendings. For patients discharged home, the difference was less than half a day.”

Those length-of-stay reductions, he adds, speak to several areas of hospitalist expertise. “I think this goes along with patient acuity, because in general, sicker patients require more complicated discharge,” Dr. Southern says. “But I also think it’s because hospitalists develop very good skills in coordinating ancillary services, discharge services and mobilizing social workers because they do so much of it.”

Experience counts

Dr. Southern’s study also adds to accumulating evidence that hospitalists don’t reduce length of stay by compromising patient care. Researchers found, for example, no differences between the two groups (hospitalists and nonhospitalists) for re-admission, in-hospital mortality or 30-day mortality.

This is a particularly important finding, says Dr. Southern, because many hospital medicine programs are not at the largest, most prestigious academic institutions. As hospital medicine continues to be embraced all types of hospitals, Dr. Southern says the question is whether the broadening use of hospitalists is causing harm. “We were able to show again that there isn’t a worsening of the mortality rate when you add hospitalists to the team,” he adds.

This study also replicated previous research, which found that hospitalists’ level of inpatient experience correlates to greater reductions in length of stay. Dr. Southern says that his next project will explore further the effect of experience. While several studies “including his own “have looked at the total time physicians spend on service each year, he plans to examine how long doctors have held their medical licenses and compare those factors to length of stay, re-admission rates and outcomes.

A hospitalist step-down unit

“One can’t get away from the fact that hospitalists tend to be younger and more recent grads and that more traditional attendings tend to be doctors who have held their license longer,” Dr. Southern says. “The question is, does that experience influence these results at all?”

He adds that his findings should help hospital administrators when they are trying to decide how best to utilize hospitalists. One option he would like to see tried is grouping patients on floors by particular diagnoses and then assigning hospitalists to cover those floors with complicated, very sick patients. He likens that concept to a hospitalist step-down area.

“It would be a way of offering hospitalist care,” Dr. Southern says, “to the subset of patients for whom this study shows hospitalists offer a very efficient, very high quality of care.”

Deborah Gesensway is a freelance writer reporting on U.S. health care from Toronto, Canada.