Determining the floor plan of a hospital is usually the responsibility of an architect, but a recent study has found reason for doctors to care about design, too.
The study, published in the May issue of CHEST, compared outcomes among more than 600 patients who were admitted to the medical intensive care unit at Columbia University Medical Center (CUMC) in New York. The hospital's ICU is set up as 12 rooms in a rectangle around a central nursing station—beds in some rooms are visible from the nursing station, while others are not.
According to the study's results, severely ill patients (those with an APACHE II score greater than 30) in the less visible rooms had higher hospital and ICU mortality than those in the visible rooms. Among less sick patients, room assignment didn't appear to affect mortality.
Study author Phillip H. Factor, DO, recently spoke with ACP Hospitalist about possible explanations for the study's findings and potential impacts on ICU operation now and hospital design in the future. Dr. Factor is an associate professor of medicine at CUMC.
Q: What motivated this study?
A: It was a simple clinical question that grew from an observation that's been made many times in ICUs: Some rooms seem to forecast a worse prognosis than others. Irrespective of visibility in the ICU, there was suspicion or legend that patients admitted to some rooms had a much greater risk of dying.
Q: What are the most likely causes of the mortality difference you found?
A: The obvious conclusion is that [in] the patients who you cannot see, you don't appreciate when something bad is happening or they're deteriorating. As a consequence, you either don't intervene or don't intervene as soon as [would have been] possible had an event been witnessed.
Q: Why did the relationship only hold true for sicker patients?
A: Those patients are probably more sensitive to small changes or small events or small deteriorations. They're already so sick that they have no reserve and even a small event or small new problem can make a big difference in outcome.
Q: What are the clinical implications of your findings?
A: If our data is confirmed, it will have a big impact on the design of future ICUs. Many ICUs around the world have patient rooms that are not well visualized by the staff that takes care of these patients, with a nursing station that cannot directly view the patient. These units may even have video monitoring, but that often doesn't help. If [our results are] confirmed, future ICUs will need to be constructed so that all patients can be visually monitored all the time.
Q: Have you seen existing ICUs with designs that allow visual monitoring?
A: Many ICUs were designed to achieve that goal but it was done in the absence of data. There's no similar data at all [to our study]. This was a supposition that many people had made that being able to visually monitor patients should be associated with better care and better outcomes.
There are some ICUs where the units are grouped together in semi-circles, so that all the rooms can be seen from a central nursing station. In some ICUs, there are a smaller number of rooms that are put together in pods so that the patients in their beds can be seen from a small nursing station.
Q: A full redesign of an ICU would take some time. Are there other methods ICUs can employ to improve visibility of their patients?
A: And hundreds of millions of dollars. Yes, there are intermediate steps that can be taken. Those would include assuring nurse-to-patient ratios that allow nurses more time either in the room or just outside of a patient's room so they can more directly observe a patient. That may include maintaining nursing ratios of 1-to-1 in many cases. There are compensatory mechanisms that we believe could be achieved through enhanced nursing care of the sickest patients.
Q: Should the sickest patients be the focus of these improvement efforts?
A: Another alternative would be to segregate the very sickest patients into rooms that can be seen from a nursing station.
Q: What other enhancements to nursing care would you recommend?
A: You would include staff education to reinforce that they need to focus more time on their patients than on interactions with their peers. We also believe that in units where visibility is limited, in a setting of a large, central nursing station, by moving nurses out of a central nursing station and minimizing peer-to-peer interaction, this would be less of an issue.