Joseph Agor1, Osman Y Özaltın2, Julie S Ivy3, Muge Capan4, Ryan Arnold5, Santiago Romero6. 1. School of Mechanical, Industrial and Manufacturing Engineering, Oregon State University, Corvallis, OR 97331-6001, United States. 2. Edward P. Fitts Department of Industrial and Systems Engineering, North Carolina State University, 400 Daniels Hall, Raleigh, NC 27695-7906, United States. Electronic address: oyozalti@ncsu.edu. 3. Edward P. Fitts Department of Industrial and Systems Engineering, North Carolina State University, 400 Daniels Hall, Raleigh, NC 27695-7906, United States. 4. Decision Sciences & MIS Department, LeBow College of Business, Drexel University, Philadelphia, PA 19104, United States. 5. Department of Emergency Medicine and School of Medicine, Drexel University, Philadelphia, PA 19102, United States. 6. Mayo Clinic Rochester, Center for Innovation, Rochester, MN 55905, United States.
Abstract
OBJECTIVE: We aim to investigate the hypothesis that using information about which variables are missing along with appropriate imputation improves the performance of severity of illness scoring systems used to predict critical patient outcomes. STUDY DESIGN AND SETTING: We quantify the impact of missing and imputed variables on the performance of prediction models used in the development of a sepsis-related severity of illness scoring system. Electronic health records (EHR) data were compiled from Christiana Care Health System (CCHS) on 119,968 adult patients hospitalized between July 2013 and December 2015. Two outcomes of interest were considered for prediction: (1) first transfer to intensive care unit (ICU) and (2) in-hospital mortality. Five different prediction models were employed. Indicators were utilized in these prediction models to identify when variables were missing and imputed. RESULTS: We observed statistically significant gains in prediction performance when moving from models that did not indicate missing information to those that did. Moreover, this increase was higher in models that use summary variables as predictors compared to those that use all variables. CONCLUSION: When developing prediction models using longitudinal EHR data, researchers should explore the incorporation of indicators for missing variables along with appropriate imputation.
OBJECTIVE: We aim to investigate the hypothesis that using information about which variables are missing along with appropriate imputation improves the performance of severity of illness scoring systems used to predict critical patient outcomes. STUDY DESIGN AND SETTING: We quantify the impact of missing and imputed variables on the performance of prediction models used in the development of a sepsis-related severity of illness scoring system. Electronic health records (EHR) data were compiled from Christiana Care Health System (CCHS) on 119,968 adult patients hospitalized between July 2013 and December 2015. Two outcomes of interest were considered for prediction: (1) first transfer to intensive care unit (ICU) and (2) in-hospital mortality. Five different prediction models were employed. Indicators were utilized in these prediction models to identify when variables were missing and imputed. RESULTS: We observed statistically significant gains in prediction performance when moving from models that did not indicate missing information to those that did. Moreover, this increase was higher in models that use summary variables as predictors compared to those that use all variables. CONCLUSION: When developing prediction models using longitudinal EHR data, researchers should explore the incorporation of indicators for missing variables along with appropriate imputation.
Authors: Titus A P de Hond; Wout J Hamelink; Mark C H de Groot; Imo E Hoefer; Jan Jelrik Oosterheert; Saskia Haitjema; Karin A H Kaasjager Journal: PLoS One Date: 2022-07-11 Impact factor: 3.752
Authors: Barbara E Jones; Jian Ying; Mckenna R Nevers; Patrick R Alba; Olga V Patterson; Kelly S Peterson; Elizabeth Rutter; Matthew A Christensen; Sarah Stern; Makoto M Jones; Adi Gundlapalli; Nathan C Dean; Matthew C Samore; Tome Greene Journal: J Gen Intern Med Date: 2022-03-09 Impact factor: 5.128
Authors: Ayis Pyrros; Adam Eugene Flanders; Jorge Mario Rodríguez-Fernández; Andrew Chen; Patrick Cole; Daniel Wenzke; Eric Hart; Samuel Harford; Jeanne Horowitz; Paul Nikolaidis; Nadir Muzaffar; Viveka Boddipalli; Jai Nebhrajani; Nasir Siddiqui; Melinda Willis; Houshang Darabi; Oluwasanmi Koyejo; William Galanter Journal: Acad Radiol Date: 2021-05-21 Impact factor: 3.173