148 Typing Speed and Accuracy as Neurocognitive Measures of Emergency Medicine Resident Physician Fatigue

Fatigue impairs performance and productivity and escalates the risk of accidents and injuries. An objective tool to screen for physician fatigue would promote a culture of safety. Typing is an important task in the electronic medical record era. Typing speed and accuracy has been used as a neurocognitive screening tool for fatigue. This study assessed the impact of shift and time of day on emergency physician typing speed and accuracy. This prospective, observational, IRB-approved, single-institution study assessed a cohort of emergency medicine resident physicians’ typing speed (word number) and accuracy (error number) at 3 different times of the day: 0900-day shift, 1600-evening shift, and 0400-night shift. Physicians were given 2 minutes to type one of 3 different cases representative of typical emergency medicine electronic medical records depending on the time slot. Each resident participated twice in each time slot and was assessed based on investigator availability. Typing samples were then screened for word count and number of errors. Data analysis utilized repeated measures of analysis of variance (ANOVA) with program year as the between-subjects factor to assess day, evening, and night word count and error rates for each time slot; both analyses met the assumption of sphericity. Because of error rate skew, they were log-transformed prior to ANOVA and reported as medians with ranges to facilitate interpretation. Data analysis utilized SPSS version 27 (Armonk, NY: IBM Corp) with p < .05 denoting statistical significance and no adjustment for the multiple comparisons. Investigators screened 35 subjects twice during the day, evening, and night shifts. Subjects typed significantly higher mean word counts of 126.0+ 23.5 during the day shift and 140.1+ 24.1 during the evening shift compared to 114.2+ 23.3 in the night shift (p < .001, Table 1). Postgraduate year (PGY) group mean word count interaction trended toward significance based on time of day (Day, Evening, Night; Table 1a, Figure 1): PGY1 121.4, 133.1, 109; PGY2 131.5, 140.6, 119.9; PGY3 125.5; 146.7, 114.3; p .07. Median error rates for the entire cohort trended towards statistical significance based on shift worked (Table 2): day 6 (0-28), evening 5 (0-24), and night 5 (0-21), p .09. PGY status did not significantly impact median error rates based on time of day (Day, Evening, Night; Table 2b, Figure 2): PGY1 3.5, 4.0, 5.0; PGY2 6, 4, 4; PGY3 7, 7, 7; p .17. Physicians typed significantly faster during morning and evening shifts compared to the night shift; senior residents’ typing output trended higher than junior residents during each shift period. The shift worked typing accuracy as reflected in the error rate trended towards significance for the entire cohort but was not significantly impacted by PGY status. Assessing typing speed and accuracy can provide meaningful information regarding physician neurocognitive status based on time of day.View Large Image Figure ViewerDownload Hi-res image Download (PPT)View Large Image Figure ViewerDownload Hi-res image Download (PPT)View Large Image Figure ViewerDownload Hi-res image Download (PPT)View Large Image Figure ViewerDownload Hi-res image Download (PPT)View Large Image Figure ViewerDownload Hi-res image Download (PPT)