1018 — Interruptions and Patient Hazards in the ICU

Drews FA (VA Salt Lake City), Markewitz B (University of Utah), Benuzillo J (VA Salt Lake City), Wallace CJ (VA Salt Lake City), Samore MH (VA Salt Lake City)

Objectives:
Actual or potential adverse medical device events are a recognized, but incompletely understood problem in the intensive care unit (ICU). Error-producing conditions (i.e., task interruptions) can negatively impact task performance and contribute to adverse events, as has been demonstrated in non-healthcare industries. We developed a tablet computer-based tool to capture nursing task interruptions in the ICU and to assess the actual or potential medical device-related harm via direct observation. The measurement tool was developed with a long-term research objective to develop strategies for reducing adverse medical device events in the ICU.

Methods:
ICU nurses served as observers. They utilized a mobile tablet computer that time stamped all nursing activities categorized as direct care (e.g. patient assessment, hygiene), indirect care (e.g. charting), or medical device use (e.g. administering medications, interacting with ventilators). Additional information recorded included the occurrence, cause, length, and impact of task interruptions and actual or potential medical device-related harm.

Results:
174 observation sessions totaling over 249 hours of observation were analyzed. Direct care, indirect care, and medical devices constituted 30.3%, 47.2%, and 16.6% of task hours, respectively. 14,797 nursing tasks have been observed (approximately one task per minute), of which 11.6% (1,722) were interrupted. The most frequent causes of interruption were device alarms (33%), other nurses (17%), and physician-caused interruptions (11%). The most frequent response to the interruptions was multi-tasking (611; 36%), switching to the interrupting task (477; 28%), delaying the interrupting task (356; 21%), or ignoring the interrupting task (233; 14%). Analyzing the odds ratio of hazards in the context of interruptions we found an odds ratio of 2.5 [95% CI: 2.23-2.74], indicating a significant increase of the occurrence of hazards in the presence of interruptions.

Implications:
Interruptions of nursing tasks in the ICU are common and frequent and significantly increase the likelihood of patient hazards with potential negative impact on patient care.

Impacts:
Interruptions of nursing tasks may be a cause of adverse events in the ICU. These findings can help to develop interventions to successfully reduce patient hazards.