Bland-Altman analysis for determining the coincidence between two methods for measuring emergency department crowding: visual analogue scale versus national emergency department overcrowding scale / 中华急诊医学杂志

ABSTRACT

Objective To establish a practicable and easily mastered emergency department overcrowding (EDO) assessment system suitable for domestic ED setting by determining the validity and usefulness of the national emergency department over-crowding study (NEDOCS) tool in a 2000-bed tertiary care academic institution in China in comparison with visual analogue scale (VAS) in order to address the worldwide grave concern of EDO.Methods In a period of 6 months,data of subjective and objective EDO assessed simultaneously three times a day (1:00,9:00,17:00) were collected.The data were analyzed by using Bland-altmann method and Kappa test to determine the coincidence between VAS and NEDOCS assessments.Results The VAS-p value evaluated by physicians was significantly lower than VAS-n value evaluated by nurses [(6.49 ± 1.82) vs.(7.12 ± 1.78),P ＜ 0.01].The reliability analysis showed that Kappa value was as low as 0.112 (P ＜0.01) suggesting there was a great discrepancy between VAS-p value and VAS-n value.VAS-m (average value of VAS-p and VAS-n) was taken for comparing different evaluation systems.The significant correlation was found between the VAS-m and NEDOCS (r =0.7l4,P ＜0.01).However,the Bland-Altman plot showed the 95％ limit of coincidence was in an extensive range (-32.47 to 71.42) suggesting discrepancy existed between two methods.Conclusions The present study suggested there was a significant discrepancy between the two subjective assessments of ED crowding (VAS-p vs.VAS-n).There might be a fatal flaw existed in the assumptions of the original VAS method for EDO measurement.Using the Bland-Altman plot analysis,the results showed that NEDOCS did not authentically reflect the staff' s sense of overcrowding in the ED.It is very important and urgent to establish an objective and effective EDO evaluation system for ED management.