This was a prospective observational study using performance indicators to assess hospital incident command groups’ decision-making and performance
Study setting
Six consecutive tabletop simulation exercises at six separate major hospitals (A-E. tables 1 and 2) were conducted during the fall of 2016 in the region of Stockholm, Sweden during the period of October 2016 to December 2016. All six simulations were antagonistic scenarios, were planned and carried out by the regional hospital disaster preparedness coordinators in Stockholm, Sweden. In an effort to make the simulations as realistic as possible, information to participants prior to the simulations, was limited. Participants were only informed of the date and approximate time. The type if incident was withheld. The extent of the simulations varied, i.e. some simulation exercises included other parts of the hospital, while others focused solely on the HICG. In both instances, the HICG had access to all units and representatives per disaster plans, facilitating similar conditions for evaluation of the HICG.
The designated hospital incident command groups, which are activated in accordance to the hospital disaster management plans, were the study subjects. Information concerning the nature of the respective incidents was withheld from participants prior to the exercise. The duration of each simulation ranged from 2 hours and 13 minutes to 6 hours and 52 minutes.
Data collection
Data collection was based on observation and included variables as required by the DiMI (3). The observers (JM and AR) were present in the hospital incident command room throughout the entire duration of each simulation. Written documentation and logfiles from the HICGs were obtained after completion of the simulations in order to ensure accurate documentation.
The DiMI consists of 22 measurable indicators divided into two groups of 11 indicators with 11 measuring decision-making skills and 11 measuring structural procedure skills. Time standards for indicators were reached through expert consensus (17, 18), The indicators reflecting decision-making skills consist of six reactive and five proactive decision-making indicators (16). Reactive decision-making indicators are characterized by decisions that may make an immediate impact on initial hospital response while proactive decision-making indicators are anticipatory in nature, characterized by decisions affecting prolonged response. Each indicator was scored on a scale from 0-2. A value of 0 indicates that the standard for the indicator was not completed. A value of 1 indicates that the standard for the indicator is partially completed or not completed within the predefined required time frame. A value of 2 indicates that the standard for the indicator was completed correctly and within the predefined required time frame.
Data analysis
Data from all simulations was first imported to Microsoft Excel for Mac version 16.33 and analyzed using descriptive and inferential statistics.
Individual indicators were analyzed using ANOVA and DUNN post hoc analysis. Differences in means for decision-making and staff procedure skills were assessed using one-way ANOVA and Kruskal-Wallis Test. Pearson’s correlation was used to assess the association between decision-making and staff procedure skills. Due to the data being rank-order data as well as a lack of assumption concerning the distribution of data, a Spearman’s rho correlation coefficient was computed to measure the degree of association between the different groups of indicators, i.e. decision-making and staff procedure skills and subgroups of decision-making skills.
A ρ value <0.05 was considered significant.
Data analysis was conducted using JASP version 0.9.2(JASP Team 2018) and Statistical Package for Social Sciences (SPSS) version 25 (IBM SPSS Statistics North Castle, New York, USA).