Checklists more effective when physicians are prompted to use them

Checklists have been used successfully to improve surgical safety and cut infection rates in the intensive care unit, but a study suggests that checklists are even more effective when physicians are prompted by a colleague to take action on information gathered using the quality improvement tool.

A study of 265 critical care patients at Northwestern Memorial Hospital in Chicago found that intensivists who relied on checklists alone did not reduce mortality rates. However, the death rate was cut in half when the checklist was accompanied by residents who asked the attending physicians how to act on information related to matters such as antibiotic prescribing, ventilator use and central-line placement.

The residents were asked to nudge physicians if the checklist indicated a need for a question, and the residents were given a script of what to ask.

"We always saw the checklist as sort of a data repository meant to trigger a decision," said Curtis H. Weiss, MD, lead author of the study. "But there was a disconnect between the information on that ... piece of paper actually leading to a decisive change. The missing link was the process of prompting a physician to make a decision based on the information."

There were cases, for example, when the checklist showed that a patient had a femoral central line in place for a week. The evidence says that to avoid infection such lines should be removed and placed at another bodily site within one to two days unless there is a significant contraindication. In such cases, residents involved in the study would ask the attending physician, "The femoral central line has been in place for seven days. Do you want to continue it?"

They asked similar questions designed to prompt physicians to remove Foley catheters or begin measures to prevent stress ulcers or deep vein thrombosis, as indicated by the checklist. Patients cared for in the "nudge" group had 45% fewer deaths from sepsis, said the study, published online May 26 in the American Journal of Respiratory and Critical Care Medicine (link). The results were adjusted to account for differences in the severity of patients' illnesses.

Spreading the nudge

If the results can be replicated, implementing such a nudging initiative outside academic medical centers will not be easy, said Dr. Weiss, instructor in the Division of Pulmonary and Critical Care Medicine at Northwestern University Feinberg School of Medicine in Chicago.

"It's hard to ask a community hospital to hire a physician to do this," he said. "We recognize that's not really scalable to other institutions for a longer period of time than what we did. We're designing an electronic checklist and plan to compare that to personal prompting."

Building the nudge into health information technology could have its own drawbacks, Dr. Weiss said. "My main concern with electronic prompting is alarm fatigue."