Research: Patients Spent 15 Hours Longer than Necessary in the ICU

Delaying transfers to other inpatient hospital rooms limits availability of costly beds

In the battle to rein in medical costs, health care providers may be overlooking one glaring inefficiency: patients are kept too long in hospital intensive care units.

ICU patients who are ready for transfer to a lower-tier hospital bed often stay in the ICU for longer than medically necessary. They are “occupying a critical care bed and thereby delaying admission for other incoming patients,” write UCLA Anderson’s Elisa F. Long and Kusum S. Mathews, a critical care physician at the Icahn School of Medicine at Mount Sinai in New York City.

Capacity-constrained hospitals “could benefit from even modest improvements” in moving patients more quickly through the system, “particularly by reducing unnecessary time spent in the ICU,” the authors write in a study published by Production and Operations Management.

Of the 36 million hospital admissions each year in the U.S., more than one-quarter involve a stay in an ICU, the study says. Total ICU-related costs top $130 billion per year, or 4 percent of all health care expenses.

Long and Mathews studied ICU occupancy at two major academic medical centers in the Northeast, one during 2010–11 and the other during 2012–14. Each patient’s length of stay in the ICU was divided into two periods: “service time,” when a critically ill patient received treatment and was stabilized; and “boarding time,” when the patient awaited transfer to other hospital wards or facilities.

The authors found that ICU patients experienced an average service time of 3.3 days and an average boarding time of 15.1 hours, though both periods were highly variable.

Their research shows that service times in ICU were not affected by room occupancy levels, implying that staff were taking whatever time was necessary for initial treatment. However, they found evidence that ICU boarding times shrank during periods of high ICU occupancy, suggesting that staff sought to free up beds by moving stabilized patients out of ICU and into other hospital wards more quickly.

But when Long and Mathews studied periods when beds were scarce in other hospital wards, they found a bottleneck: ICU boarding times were as much as 67 percent longer compared to periods of highest bed availability in the wards, the authors found.

What’s more, even when other wards were only partially full (75–85 percent of beds occupied), ICU boarding times still lengthened significantly, “suggesting that hospital staff may be too conservative” in delaying transfers out of ICUs, the study says.

Extended ICU boarding times have effects that ripple through hospitals, the authors write. They cite studies from 2005 and 2011 showing that ICU congestion creates chokepoints for other units, “especially the ED [emergency department] and post-surgery care area, leading to overcrowding, delays in care, and negative financial consequences due to revenue loss from ambulance diversion and patients leaving without being seen.”

Long and Mathews suggest several ways to attack the problem of ICU chokepoints:

• At a minimum, hospitals should begin to measure their ICU boarding times. The authors write that “virtually every large U.S. hospital has implemented electronic medical records, making timestamps for bed assignment and discharge readily available.” Yet “few hospitals also record timestamps for bed requests” — when the attending physician requests a hospital admission, internal transfer, or discharge — which could help quantify ICU boarding times after patients are stabilized.
• During periods of strained capacity, hospitals could “temporarily boost staffing and mobilize nursing teams” to attend to patients who require an ICU bed but are unable to immediately receive one. Shifting to a “flexible-bed” model, where ICU capacity temporarily surges during high demand, could reduce rejections of critical patients from other hospitals because of limited bed availability.
• Communication among different wards should be strengthened. Slowdowns in ICU transfers may occur because ward beds are spread across many independent units (e.g., oncology, cardiology, etc.) that are “each behaving myopically without considering the aggregate effect on ICU bed strain,” the study says. “Better coordination across the ICU and all medicine wards could streamline transfers out of the ICU, reducing ICU admission delays for incoming patients.”