Supporting the detection of patient deterioration: Observation chart design affects the recognition of abnormal vital signs

Aim: To evaluate the effect of observation chart design on the ability of health professionals and novice chart users to recognise patient deterioration. Methods: Participants were 45 health professionals (doctors and nurses) and 46 novices. Each participant completed 48 trials in which they viewed realistic patient observations recorded on six hospital observation charts of differing design quality. Each chart design was used on eight trials, four times with normal data and four times with abnormal data. On each trial, the participant's task was to identify any abnormal observation or else to indicate that all observations were normal (based on normal physiological ranges given in the instructions). The main outcome measures were participants' error rates and response times for detecting abnormal observations on each of the six charts. Results: There was a significant effect of chart type on error rates (p < 0.001), but health professionals made the same number of errors as novices (p = 0.43). Chart type also had a significant effect on response times (p < 0.001). Health professionals responded faster overall than novices (p = 0.006); however, a significant interaction between chart type and participant group (p = 0.02) indicated that the health professionals' advantage was confined to the two most rudimentary charts. No significant differences were found between doctors and nurses on either measure. Conclusions: Our findings suggest that observation chart design has a substantial impact on the decision accuracy and response times of both health professionals and novices in recognising abnormal patient observations. (C) 2012 Elsevier Ireland Ltd. All rights reserved.