We did not include fixed costs, which are the costs that do not change when the EEG monitoring strategy changes such as the overall cost of operating the intensive care unit. We used this method because it focuses attention on variable costs, which are the costs that change when more or fewer EEGs are done and when monitoring is for longer or shorter durations.
To calculate total cost, for each component we multiplied the number of units times the cost per unit, and then summed the values for all components. For example, the average hourly cost for technologists or physicians and the average cost per procedure for equipment. We estimated how much of each component is used and assigned a cost to each unit of use. We used a micro-costing method to estimate the direct medical costs involved in EEG monitoring such as equipment, technologists, and physicians. In children having seizures, the probability of identifying seizures was 0.55 with 1 hour of EEG monitoring 0.85 with 24 hours of EEG monitoring and 0.89 with 48 hours of EEG monitoring. Because seizures could begin and end at different times, we used some of the same studies to determine when seizures occurred in relation to monitoring initiation ( Supplemental Table 2). We pooled data from 17 studies which mostly reported the results of clinically indicated EEG monitoring in critically ill children and found that 740 of 2,247 (0.33) children had electrographic seizures ( Supplemental Table 1). The reference lists from identified publications were also reviewed. One author (N.S.A) conducted a systematic review using PubMed keywords EEG monitoring, critical care, pediatric, and seizures to identify papers published in English related to EEG monitoring in critically ill children.
To analyze the model, we needed information on the probability of identifying electrographic seizures with monitoring and the cost of EEG monitoring. The square signifies the choose node, circles signify the chance nodes, and triangles signify the terminal nodes. Depending on the duration of EEG monitoring, the electrographic seizures may or may not be identified. For a given patient, electrographic seizures may or may not occur. The decision maker may choose between the four EEG monitoring strategies (no EEG, 1 hour EEG, 24 hours EEG, or 48 hours EEG). 2012) Since healthcare costs are rising and societal resources are limited, it is important to consider the consequences of implementing these guidelines.( Cassel and Guest 2012, Holloway and Ringel 2011) To provide this information, we modelled four EEG monitoring strategies for identifying electrographic seizures in critically ill children and evaluated their relative cost-effectiveness.ĭecision tree used to represent the relationships among the variables important to monitoring strategy decisions. 2012) Unfortunately, EEG monitoring is costly because it involves expensive equipment and substantial work by technicians and physicians, and thus small changes in the duration of monitoring can have substantial resource impacts.( Gutierrez-Colina, et al. 2006, Williams, Jarrar and Buchhalter 2011) Recent guidelines recommended monitoring at-risk encephalopathic children for 48 hours.( Brophy, et al. 2012, Shahwan, Bailey, Shekerdemian, et al., Tay, et al. 2013, Schreiber, Zelleke, Gaillard, et al. 2014, Piantino, Wainwright, Grimason, et al. 2013, Abend and Dlugos 2007, Abend, Gutierrez-Colina, Topjian, et al. 2011) The incidence of electrographic seizures varies from 10 to 40%, depending in part on the duration of EEG monitoring.( Abend, Arndt, Carpenter, et al. Most electrographic seizures in critically ill children with acute encephalopathy have no clinical correlate so identification requires continuous EEG monitoring.( Abend, et al.
0 kommentar(er)
