Auditory and visual alarm designs impact clinicians' perceived cognitive workload

Healthcare settings heavily rely on clinicians' abilities to interpret vital sign alarms indicating patient decompensation. Meanwhile, clinicians are bombarded with many multisensory stimuli necessary for patient care, including simultaneous visual and auditory displays. Here, we aim to assess how our modified auditory and visual alarm designs impact clinicians' perceived cognitive workload. This experimental study, conducted at Vanderbilt University Medical Center (VUMC) between March and September 2023, included 26 clinicians (nurse practitioners, residents, and fellows). Auditory trials involved 15 clinicians and non-clinicians (university students) to validate design intuitiveness. Clinicians participated in visual and auditory trials to identify simulated mean arterial pressure (MAP), utilizing standard and modified alarms. Visual modifications incorporated a line-graph display with a moving dot for MAP. Auditory modifications introduced harmonic overlays indicating severity and direction of MAP values. After each trial, participants completed the National Aeronautics and Space Administration Task Load Index (NASA-TLX) to assess perceived workload across 6 domains (temporal demand, physical demand, mental demand, effort, performance, frustration) on a 1-20 Likert scale with increased scores represent greater workload. For analysis, Wilcoxon signed-rank and rank-sum tests were used. Demographics for auditory alarm trials averaged an age of 26.2 and 54% identified as male. Visual display trials included 26 clinicians with an average age of 30.1 and 59% identified as male. In visual trials, clinicians reported significantly lower temporal demand with the modified monitor (median, interquartile range (IQR)) (8.0, 4.2-11.8) compared to the conventional monitor (13.0, 6.5-16.0; p = 0.022). In auditory trials, clinicians reported significantly higher perceived performance with conventional auditory alarms as compared to non-clinicians (10.0, 5.0-13.0) vs. (4.0, 2.0-9.0; p = 0.022). Non-clinicians reported higher perceived temporal demand for conventional auditory alarms when compared to clinicians (6.0, 3.0-10.0) vs. (2.5, 1.0-5.0; p = 0.024). Our findings suggest modifications to both visual and auditory alarms can reduce elements of perceived cognitive workload, especially temporal demand, while preserving clinician performance without deterioration of other measured components.