The impact of missing trauma data on predicting massive transfusion

Amber W. Trickey, Erin E. Fox, Deborah J. Del Junco, Jing Ning, John B. Holcomb, Karen Brasel, Mitchell J. Cohen, Martin Schreiber, Eileen M. Bulger, Herb A. Phelan, Louis H. Alarcon, John G. Myers, Peter Muskat, Bryan A. Cotton, Charles E. Wade, Mohammad H. Rahbar

    Research output: Contribution to journalArticle

    13 Citations (Scopus)

    Abstract

    BACKGROUND: Missing data are inherent in clinical research andmay be especially problematic for trauma studies. This study describes a sensitivity analysis to evaluate the impact ofmissing data on clinical risk prediction algorithms.Three blood transfusion predictionmodelswere evaluated using an observational trauma data set with valid missing data. METHODS: The PRospectiveObservationalMulticenterMajor Trauma Transfusion (PROMMTT) study included patients requiring one or more unit of red blood cells at 10 participating US Level I trauma centers from July 2009 to October 2010. Physiologic, laboratory, and treatment data were collected prospectively up to 24 hours after hospital admission. Subjects who received 10 or more units of red blood cells within 24 hours of admission were classified as massive transfusion (MT) patients. Correct classification percentages for three MT prediction models were evaluated using complete case analysis and multiple imputation. A sensitivity analysis for missing data was conducted to determine the upper and lower bounds for correct classification percentages. RESULTS: PROMMTT study enrolled 1,245 subjects.MTwas received by 297 patients (24%). Missing percentage ranged from 2.2% (heart rate) to 45%(respiratory rate). Proportions of complete cases used in theMTpredictionmodels ranged from 41%to 88%.Allmodels demonstrated similar correct classification percentages using complete case analysis and multiple imputation. In the sensitivity analysis, correct classification upper-lower bound ranges permodelwere 4%,10%, and 12%. Predictive accuracy for allmodels usingPROMMTTdatawas lower than reported in the original data sets. CONCLUSION: Evaluating the accuracy clinical prediction models with missing data can be misleading, especially with many predictor variables and moderate levels of missingness per variable. The proposed sensitivity analysis describes the influence of missing data on risk prediction algorithms. Reporting upper-lower bounds for percent correct classification may be more informative than multiple imputation, which provided similar results to complete case analysis in this study.

    Original languageEnglish (US)
    JournalJournal of Trauma and Acute Care Surgery
    Volume75
    Issue number1 SUPPL1
    DOIs
    StatePublished - 2013

    Fingerprint

    Wounds and Injuries
    Erythrocytes
    Trauma Centers
    Respiratory Rate
    Blood Transfusion
    Heart Rate
    Research
    Datasets
    Therapeutics

    Keywords

    • Incomplete data
    • Massive transfusion
    • PROMMTT
    • Trauma

    ASJC Scopus subject areas

    • Critical Care and Intensive Care Medicine
    • Surgery

    Cite this

    Trickey, A. W., Fox, E. E., Del Junco, D. J., Ning, J., Holcomb, J. B., Brasel, K., ... Rahbar, M. H. (2013). The impact of missing trauma data on predicting massive transfusion. Journal of Trauma and Acute Care Surgery, 75(1 SUPPL1). https://doi.org/10.1097/TA.0b013e3182914530

    The impact of missing trauma data on predicting massive transfusion. / Trickey, Amber W.; Fox, Erin E.; Del Junco, Deborah J.; Ning, Jing; Holcomb, John B.; Brasel, Karen; Cohen, Mitchell J.; Schreiber, Martin; Bulger, Eileen M.; Phelan, Herb A.; Alarcon, Louis H.; Myers, John G.; Muskat, Peter; Cotton, Bryan A.; Wade, Charles E.; Rahbar, Mohammad H.

    In: Journal of Trauma and Acute Care Surgery, Vol. 75, No. 1 SUPPL1, 2013.

    Research output: Contribution to journalArticle

    Trickey, AW, Fox, EE, Del Junco, DJ, Ning, J, Holcomb, JB, Brasel, K, Cohen, MJ, Schreiber, M, Bulger, EM, Phelan, HA, Alarcon, LH, Myers, JG, Muskat, P, Cotton, BA, Wade, CE & Rahbar, MH 2013, 'The impact of missing trauma data on predicting massive transfusion', Journal of Trauma and Acute Care Surgery, vol. 75, no. 1 SUPPL1. https://doi.org/10.1097/TA.0b013e3182914530
    Trickey, Amber W. ; Fox, Erin E. ; Del Junco, Deborah J. ; Ning, Jing ; Holcomb, John B. ; Brasel, Karen ; Cohen, Mitchell J. ; Schreiber, Martin ; Bulger, Eileen M. ; Phelan, Herb A. ; Alarcon, Louis H. ; Myers, John G. ; Muskat, Peter ; Cotton, Bryan A. ; Wade, Charles E. ; Rahbar, Mohammad H. / The impact of missing trauma data on predicting massive transfusion. In: Journal of Trauma and Acute Care Surgery. 2013 ; Vol. 75, No. 1 SUPPL1.
    @article{978675856b7f4f4eb9c463ff44e187f0,
    title = "The impact of missing trauma data on predicting massive transfusion",
    abstract = "BACKGROUND: Missing data are inherent in clinical research andmay be especially problematic for trauma studies. This study describes a sensitivity analysis to evaluate the impact ofmissing data on clinical risk prediction algorithms.Three blood transfusion predictionmodelswere evaluated using an observational trauma data set with valid missing data. METHODS: The PRospectiveObservationalMulticenterMajor Trauma Transfusion (PROMMTT) study included patients requiring one or more unit of red blood cells at 10 participating US Level I trauma centers from July 2009 to October 2010. Physiologic, laboratory, and treatment data were collected prospectively up to 24 hours after hospital admission. Subjects who received 10 or more units of red blood cells within 24 hours of admission were classified as massive transfusion (MT) patients. Correct classification percentages for three MT prediction models were evaluated using complete case analysis and multiple imputation. A sensitivity analysis for missing data was conducted to determine the upper and lower bounds for correct classification percentages. RESULTS: PROMMTT study enrolled 1,245 subjects.MTwas received by 297 patients (24{\%}). Missing percentage ranged from 2.2{\%} (heart rate) to 45{\%}(respiratory rate). Proportions of complete cases used in theMTpredictionmodels ranged from 41{\%}to 88{\%}.Allmodels demonstrated similar correct classification percentages using complete case analysis and multiple imputation. In the sensitivity analysis, correct classification upper-lower bound ranges permodelwere 4{\%},10{\%}, and 12{\%}. Predictive accuracy for allmodels usingPROMMTTdatawas lower than reported in the original data sets. CONCLUSION: Evaluating the accuracy clinical prediction models with missing data can be misleading, especially with many predictor variables and moderate levels of missingness per variable. The proposed sensitivity analysis describes the influence of missing data on risk prediction algorithms. Reporting upper-lower bounds for percent correct classification may be more informative than multiple imputation, which provided similar results to complete case analysis in this study.",
    keywords = "Incomplete data, Massive transfusion, PROMMTT, Trauma",
    author = "Trickey, {Amber W.} and Fox, {Erin E.} and {Del Junco}, {Deborah J.} and Jing Ning and Holcomb, {John B.} and Karen Brasel and Cohen, {Mitchell J.} and Martin Schreiber and Bulger, {Eileen M.} and Phelan, {Herb A.} and Alarcon, {Louis H.} and Myers, {John G.} and Peter Muskat and Cotton, {Bryan A.} and Wade, {Charles E.} and Rahbar, {Mohammad H.}",
    year = "2013",
    doi = "10.1097/TA.0b013e3182914530",
    language = "English (US)",
    volume = "75",
    journal = "Journal of Trauma and Acute Care Surgery",
    issn = "2163-0755",
    publisher = "Lippincott Williams and Wilkins",
    number = "1 SUPPL1",

    }

    TY - JOUR

    T1 - The impact of missing trauma data on predicting massive transfusion

    AU - Trickey, Amber W.

    AU - Fox, Erin E.

    AU - Del Junco, Deborah J.

    AU - Ning, Jing

    AU - Holcomb, John B.

    AU - Brasel, Karen

    AU - Cohen, Mitchell J.

    AU - Schreiber, Martin

    AU - Bulger, Eileen M.

    AU - Phelan, Herb A.

    AU - Alarcon, Louis H.

    AU - Myers, John G.

    AU - Muskat, Peter

    AU - Cotton, Bryan A.

    AU - Wade, Charles E.

    AU - Rahbar, Mohammad H.

    PY - 2013

    Y1 - 2013

    N2 - BACKGROUND: Missing data are inherent in clinical research andmay be especially problematic for trauma studies. This study describes a sensitivity analysis to evaluate the impact ofmissing data on clinical risk prediction algorithms.Three blood transfusion predictionmodelswere evaluated using an observational trauma data set with valid missing data. METHODS: The PRospectiveObservationalMulticenterMajor Trauma Transfusion (PROMMTT) study included patients requiring one or more unit of red blood cells at 10 participating US Level I trauma centers from July 2009 to October 2010. Physiologic, laboratory, and treatment data were collected prospectively up to 24 hours after hospital admission. Subjects who received 10 or more units of red blood cells within 24 hours of admission were classified as massive transfusion (MT) patients. Correct classification percentages for three MT prediction models were evaluated using complete case analysis and multiple imputation. A sensitivity analysis for missing data was conducted to determine the upper and lower bounds for correct classification percentages. RESULTS: PROMMTT study enrolled 1,245 subjects.MTwas received by 297 patients (24%). Missing percentage ranged from 2.2% (heart rate) to 45%(respiratory rate). Proportions of complete cases used in theMTpredictionmodels ranged from 41%to 88%.Allmodels demonstrated similar correct classification percentages using complete case analysis and multiple imputation. In the sensitivity analysis, correct classification upper-lower bound ranges permodelwere 4%,10%, and 12%. Predictive accuracy for allmodels usingPROMMTTdatawas lower than reported in the original data sets. CONCLUSION: Evaluating the accuracy clinical prediction models with missing data can be misleading, especially with many predictor variables and moderate levels of missingness per variable. The proposed sensitivity analysis describes the influence of missing data on risk prediction algorithms. Reporting upper-lower bounds for percent correct classification may be more informative than multiple imputation, which provided similar results to complete case analysis in this study.

    AB - BACKGROUND: Missing data are inherent in clinical research andmay be especially problematic for trauma studies. This study describes a sensitivity analysis to evaluate the impact ofmissing data on clinical risk prediction algorithms.Three blood transfusion predictionmodelswere evaluated using an observational trauma data set with valid missing data. METHODS: The PRospectiveObservationalMulticenterMajor Trauma Transfusion (PROMMTT) study included patients requiring one or more unit of red blood cells at 10 participating US Level I trauma centers from July 2009 to October 2010. Physiologic, laboratory, and treatment data were collected prospectively up to 24 hours after hospital admission. Subjects who received 10 or more units of red blood cells within 24 hours of admission were classified as massive transfusion (MT) patients. Correct classification percentages for three MT prediction models were evaluated using complete case analysis and multiple imputation. A sensitivity analysis for missing data was conducted to determine the upper and lower bounds for correct classification percentages. RESULTS: PROMMTT study enrolled 1,245 subjects.MTwas received by 297 patients (24%). Missing percentage ranged from 2.2% (heart rate) to 45%(respiratory rate). Proportions of complete cases used in theMTpredictionmodels ranged from 41%to 88%.Allmodels demonstrated similar correct classification percentages using complete case analysis and multiple imputation. In the sensitivity analysis, correct classification upper-lower bound ranges permodelwere 4%,10%, and 12%. Predictive accuracy for allmodels usingPROMMTTdatawas lower than reported in the original data sets. CONCLUSION: Evaluating the accuracy clinical prediction models with missing data can be misleading, especially with many predictor variables and moderate levels of missingness per variable. The proposed sensitivity analysis describes the influence of missing data on risk prediction algorithms. Reporting upper-lower bounds for percent correct classification may be more informative than multiple imputation, which provided similar results to complete case analysis in this study.

    KW - Incomplete data

    KW - Massive transfusion

    KW - PROMMTT

    KW - Trauma

    UR - http://www.scopus.com/inward/record.url?scp=84880416692&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84880416692&partnerID=8YFLogxK

    U2 - 10.1097/TA.0b013e3182914530

    DO - 10.1097/TA.0b013e3182914530

    M3 - Article

    C2 - 23778514

    AN - SCOPUS:84880416692

    VL - 75

    JO - Journal of Trauma and Acute Care Surgery

    JF - Journal of Trauma and Acute Care Surgery

    SN - 2163-0755

    IS - 1 SUPPL1

    ER -