Who Needs an ICU Bed Most? New Study by IHPI members

ANN ARBOR, Mich. -- Just because a patient has had a heart attack, or a flare-up of their chronic heart or lung disease, doesn’t mean their next stop should be an intensive care unit, a new study suggests.

If they’re critically ill, of course, the ICU is the best place. But for patients in less-dire condition, an ICU stay may not increase their chance of survival, even as it costs more, the research finds.

Across America, hospitals vary widely in what they do with such patients – and doctors have little evidence to guide them on the best option for each patient. The new study could help lay the groundwork for more effective – and cost-effective – use of ICU care for these conditions.

Published by a team from the University of Michigan Institute for Healthcare Policy and Innovation in the Annals of the American Thoracic Society, the study is based on an analysis of more than 1.5 million Medicare records.

“ICU care can save lives, but it is also very costly,” says lead author Thomas Valley, MD, MSc, a pulmonary and critical care researcher at the U-M Medical School. “Our results highlight that there is a large group of patients who doctors have trouble figuring out whether the ICU will help them or not."

The data came from people who were hospitalized for a flare-up of chronic obstructive pulmonary disease or heart failure, or for a heart attack, over a three-year period. The researchers tracked how close they lived to the hospital where they were treated, and how often those hospitals admitted such patients to ICUs or general wards.

The researchers zeroed in on those patients who were likely admitted to an ICU solely because they lived near a hospital that tended to send more of such patients to ICU beds. In other words, those in the “margin” or “bubble” between clearly needing an ICU and clearly being able to be cared for in a general inpatient unit.

In all, about one in six of the patients in the study met this description. But the researchers found that despite getting ICU-level care, they were no less likely to have died within 30 days of their hospital stay than patients who stayed in a general hospital bed.

One thing that was different, though, was the cost of their care – at least for the patients with heart attack or heart failure.

Those who stayed in an ICU racked up hospital bills several thousand dollars higher than those cared for in a general ward – more than $2,600 more for heart failure patients and nearly $5,000 more for heart attack patients. This is likely because of the additional testing and procedures that mark ICU care, but also potentially because of the higher chance of developing infections there. COPD patients didn’t have a significant difference in cost between the two care settings.

“We found that the ICU may not always be the answer,” says Valley. “Now, we need to help doctors decide who needs the ICU and who doesn’t.”

Valley and his co-authors on the new paper previously applied their model to older patients admitted with pneumonia, and found that those who were “on the bubble” between needing an ICU or a general bed were actually more likely to survive if they were admitted to an ICU. The acute nature of pneumonia, compared with the chronic nature of heart failure and COPD, may have something to do with this difference, they think.

They’ve also done previous work showing that those hospitals that send higher percentages of their heart patients to the ICU actually perform worse on measures of health care quality.

Valley and the senior author of the new paper, Colin Cooke, M.D., M.Sc., M.S. wrote a commentary last year on the issue of how doctors decide where to send these “on the bubble” patients.

The authors conclude, “These findings suggest that the ICU may be overused for some COPD, heart failure, or acute myocardial infarction patients with an uncertain indication for intensive care, and opportunities exist to decrease healthcare costs by reducing ICU admissions for certain patients.” 

Future studies, they said, should help define which patients with these conditions would benefit from the ICU and which can be treated elsewhere in the hospital without compromising their care.

The study was funded by the National Institutes of Health, the Dept. of Veterans Affairs and the Agency for Healthcare Research and Quality. Valley, Cooke and their co-authors Michael W. Sjoding, MD, MSc; Andrew M. Ryan, PhD; and Theodore J. Iwashyna, MD, PhD are all members of the U-M Institute for Healthcare Policy and Innovation, and all except Sjoding are members of the Michigan Center for Integrative Research in Critical Care.

Reference: DOI: http://dx.doi.org/10.1513/AnnalsATS.201611-847OC