4 Which intensive care analytics are a worthwhile investment for developers? An early health technology assessment

Introduction In intensive care (IC), monitoring data are continuously being collected. Unfortunately, this data is often not interpretable for clinicians and therefore goes unused. While advanced analytics can process this data to support improvements in patient care, developers ideally invest in those analytics that lead to the greatest health benefits or cost reductions. We explored the potential cost-effectiveness of three IC analytics. Subjects and methods We assessed the cost-effectiveness of three applications: the early identification of catheter-related bloodstream infection (CRBSI), improved nutrition monitoring and identification of high driving pressure (HDP) in ventilated patients. The literature was used to populate the decision trees and patient data from a Greek IC were used for the nutrition and HDP estimates. We performed univariate, multivariate, and probabilistic sensitivity analyses. Results Early identification of CRBSI reduced mortality and length of stay, thereby increasing quality-adjusted life-years (QALYs) (6.70 vs. 6.58) and reducing costs (€8183 vs. €9,869); probability of CRBSI and the costs per IC day were key influencers of the ICER and the probability of dominance was 90%. Improved nutrition monitoring could increase nutrition adherence (0.37% vs. 0.03%) and decrease costs by reducing length of stay (€9988 vs. €10,720). For HDP, benefits were limited, resulting in higher costs as well as minimal and highly uncertain health benefits. Conclusions Developers should study the potential impact of the analytics they wish to develop before investing time and money. We illustrate that analytics for early identification of CRBSI and improved nutrition monitoring are worth further investment while identifying HDP may not be. The other factors that influence success besides health and financial benefits such as the competition and data access should be assessed. This research was part of the European AEGLE project and received funding from the European Union’s Horizon 2020 Research and Innovation Programme under grant agreement No. 6 44 906.