OBJECTIVES: To describe the modeling techniques used for early prediction of outcome in traumatic brain injury (TBI) and to identify aspects for potential improvements. STUDY DESIGN AND SETTING: We reviewed key methodological aspects of studies published between 1970 and 2005 that proposed a prognostic model for the Glasgow Outcome Scale of TBI based on admission data. RESULTS: We included 31 papers. Twenty-four were single-center studies, and 22 reported on fewer than 500 patients. The median of the number of initially considered predictors was eight, and on average five of these were selected for the prognostic model, generally including age, Glasgow Coma Score (or only motor score), and pupillary reactivity. The most common statistical technique was logistic regression with stepwise selection of predictors. Model performance was often quantified by accuracy rate rather than by more appropriate measures such as the area under the receiver-operating characteristic curve. Model validity was addressed in 15 studies, but mostly used a simple split-sample approach, and external validation was performed in only four studies. CONCLUSION: Although most models agree on the three most important predictors, many were developed on small sample sizes within single centers and hence lack generalizability. Modeling strategies have to be improved, and include external validation.
OBJECTIVES: To describe the modeling techniques used for early prediction of outcome in traumatic brain injury (TBI) and to identify aspects for potential improvements. STUDY DESIGN AND SETTING: We reviewed key methodological aspects of studies published between 1970 and 2005 that proposed a prognostic model for the Glasgow Outcome Scale of TBI based on admission data. RESULTS: We included 31 papers. Twenty-four were single-center studies, and 22 reported on fewer than 500 patients. The median of the number of initially considered predictors was eight, and on average five of these were selected for the prognostic model, generally including age, Glasgow Coma Score (or only motor score), and pupillary reactivity. The most common statistical technique was logistic regression with stepwise selection of predictors. Model performance was often quantified by accuracy rate rather than by more appropriate measures such as the area under the receiver-operating characteristic curve. Model validity was addressed in 15 studies, but mostly used a simple split-sample approach, and external validation was performed in only four studies. CONCLUSION: Although most models agree on the three most important predictors, many were developed on small sample sizes within single centers and hence lack generalizability. Modeling strategies have to be improved, and include external validation.
Authors: Bram Jacobs; Tjemme Beems; Ton M van der Vliet; Arie B van Vugt; Cornelia Hoedemaekers; Janneke Horn; Gaby Franschman; Ian Haitsma; Joukje van der Naalt; Teuntje M J C Andriessen; George F Borm; Pieter E Vos Journal: Neurocrit Care Date: 2013-08 Impact factor: 3.210
Authors: Endre Czeiter; Stefania Mondello; Noemi Kovacs; Janos Sandor; Andrea Gabrielli; Kara Schmid; Frank Tortella; Kevin K W Wang; Ronald L Hayes; Pal Barzo; Erzsebet Ezer; Tamas Doczi; Andras Buki Journal: J Neurotrauma Date: 2012-04-30 Impact factor: 5.269
Authors: Julia Halford; Sean Shen; Kyohei Itamura; Jaclynn Levine; Albert C Chong; Gregg Czerwieniec; Thomas C Glenn; David A Hovda; Paul Vespa; Ross Bullock; W Dalton Dietrich; Stefania Mondello; Joseph A Loo; Ina-Beate Wanner Journal: J Cereb Blood Flow Metab Date: 2017-08-17 Impact factor: 6.200