Checklists Can Improve ICU Care, With Prompt Attention

In the complex environment of the intensive care unit, checklists have become synonymous with attempts to simplify care and reduce medical errors, according to a study published online ahead of print in the American Journal of Respiratory and Critical Care Medicine. Although the results of field tests have been disappointing, recent research has shown significant improvement in the quality of care and even in mortality rates of some patients, indicating that the problem lies not within the checklist itself, but in its implementation.

Curtis Weiss, MD, MS, instructor of pulmonary and critical care medicine at Northwestern University in Chicago, found that when ICU treatment teams were verbally prompted to follow a checklist, patients under their care fared remarkably better than those under the care of an unprompted team.

“Previous studies showing disappointing results of checklists may be due to the failure of individual physicians to regularly use the list, rather than shortcomings of the lists themselves,” explained Dr. Weiss.

But even when physicians do use checklists, improvements in care are not automatic. In 2008, Dr. Weiss’s attempt to introduce a checklist at Northwestern University hospital failed. “Even when it was being filled out, it wasn’t leading to changes in patient management,” he said. In one case, a patient had a central venous catheter for six days before it was removed. “You could see by the handwriting on the checklist that each day different people had noted it and left it in,” continued Dr. Weiss. “But that is certainly not the standard of care.”

The failure of checklists to elicit changes in the medical decision-making is a form of alarm fatigue, according to Dr. Weiss. “There are thousands of moving parts and activities in the ICU,” he explained. “Many cases require urgent attention, so it is easy to overlook subtler problems like prolonged ventilation, which can also lead to poor outcomes.”

Dr. Weiss collaborated with colleagues to create a checklist of parameters known to be associated with ICU outcomes and tracked patients admitted to the ICU under two independent teams. One of the teams’ attending physicians was prompted to follow a checklist by a resident physician using a script. “If, for example, the team failed to discuss the central venous catheter (CVC), the prompter would say, “The CVC has been in place for X days. Do you want to continue it?” The prompters had direct contact with or responsibilities for caring for patients. The second team used an identical checklist without prompting.

Patients under prompted care had more ventilator-free days, decreased empiric antibiotic use and CVC duration; increased rates of prophylaxis for deep DVT and ulcers; and decreased length of stay. Overall mortality was also significantly reduced.

“Constant attentiveness to the care practices under investigation, driven by prompting, improved care,” said Dr. Weiss.

While using a resident physician as a prompter is not feasible on a large scale, Dr. Weiss and his colleagues are currently investigating other ways of implementing checklists while avoiding alarm fatigue. “In any case, the idea of prompting physicians to discuss these types of issues can be applied more broadly across the healthcare system, and potentially lead to widespread patient benefit,” he said.