RESUMO
Objectives@#Predictive models for critical events in the intensive care unit (ICU) might help providers anticipate patient deterioration. At the heart of predictive model development lies the ability to accurately label significant events, thereby facilitating the use of machine learning and similar strategies. We conducted this study to establish the validity of an automated system for tagging respiratory and hemodynamic deterioration by comparing automatic tags to tagging by expert reviewers. @*Methods@#This retrospective cohort study included 72,650 unique patient stays collected from Electronic Medical Records of the University of Massachusetts’ eICU. An enriched subgroup of stays was manually tagged by expert reviewers. The tags generated by the reviewers were compared to those generated by an automated system. @*Results@#The automated system was able to rapidly and efficiently tag the complete database utilizing available clinical data. The overall agreement rate between the automated system and the clinicians for respiratory and hemodynamic deterioration tags was 89.4% and 87.1%, respectively. The automatic system did not add substantial variability beyond that seen among the reviewers. @*Conclusions@#We demonstrated that a simple rule-based tagging system could provide a rapid and accurate tool for mass tagging of a compound database. These types of tagging systems may replace human reviewers and save considerable resources when trying to create a validated, labeled database used to train artificial intelligence algorithms. The ability to harness the power of artificial intelligence depends on efficient clinical validation of targeted conditions; hence, these systems and the methodology used to validate them are crucial.
RESUMO
Objectives@#Predictive models for critical events in the intensive care unit (ICU) might help providers anticipate patient deterioration. At the heart of predictive model development lies the ability to accurately label significant events, thereby facilitating the use of machine learning and similar strategies. We conducted this study to establish the validity of an automated system for tagging respiratory and hemodynamic deterioration by comparing automatic tags to tagging by expert reviewers. @*Methods@#This retrospective cohort study included 72,650 unique patient stays collected from Electronic Medical Records of the University of Massachusetts’ eICU. An enriched subgroup of stays was manually tagged by expert reviewers. The tags generated by the reviewers were compared to those generated by an automated system. @*Results@#The automated system was able to rapidly and efficiently tag the complete database utilizing available clinical data. The overall agreement rate between the automated system and the clinicians for respiratory and hemodynamic deterioration tags was 89.4% and 87.1%, respectively. The automatic system did not add substantial variability beyond that seen among the reviewers. @*Conclusions@#We demonstrated that a simple rule-based tagging system could provide a rapid and accurate tool for mass tagging of a compound database. These types of tagging systems may replace human reviewers and save considerable resources when trying to create a validated, labeled database used to train artificial intelligence algorithms. The ability to harness the power of artificial intelligence depends on efficient clinical validation of targeted conditions; hence, these systems and the methodology used to validate them are crucial.
RESUMO
BackgroundThe COVID-19 pandemic has major ramifications for global health and the economy, with growing concerns about economic recession and implications for mental health. Here we investigated the associations between COVID-19 pandemic-related income loss with financial strain and mental health trajectories over a 1-month course. MethodsTwo independent studies were conducted in the U.S and in Israel at the beginning of the outbreak (March-April 2020, T1; N = 4 171) and at a 1-month follow-up (T2; N = 1 559). Mixed-effects models were applied to assess associations among COVID-19-related income loss, financial strain, and pandemic-related worries about health, with anxiety and depression, controlling for multiple covariates including pre-COVID-19 income. FindingsIn both studies, income loss and financial strain were associated with greater depressive symptoms at T1, above and beyond T1 anxiety, worries about health, and pre-COVID-19 income. Worsening of income loss was associated with exacerbation of depression at T2 in both studies. Worsening of subjective financial strain was associated with exacerbation of depression at T2 in one study (US). InterpretationIncome loss and financial strain were uniquely associated with depressive symptoms and the exacerbation of symptoms over time, above and beyond pandemic-related anxiety. Considering the painful dilemma of lockdown versus reopening, with the tradeoff between public health and economic wellbeing, our findings provide evidence that the economic impact of COVID-19 has negative implications for mental health. FundingThis study was supported by grants from the National Institute of Mental Health, the US-Israel Binational Science Foundation, Foundation Dora and Kirsh Foundation.
