External validation of a smartphone app model to predict the need for massive transfusion using five different definitions

E. I. Hodgman, Michael W Cripps, M. J. Mina, E. M. Bulger, M. A. Schreiber, K. J. Brasel, M. J. Cohen, P. Muskat, J. G. Myers, L. H. Alarcon, M. H. Rahbar, J. B. Holcomb, B. A. Cotton, E. E. Fox, D. J. Del Junco, C. E. Wade, Herbert Phelan

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

BACKGROUND Previously, a model to predict massive transfusion protocol (MTP) (activation) was derived using a single-institution data set. The PRospective, Observational, Multicenter, Major Trauma Transfusion database was used to externally validate this model's ability to predict both MTP activation and massive transfusion (MT) administration using multiple MT definitions. METHODS The app model was used to calculate the predicted probability of MTP activation or MT delivery. The five definitions of MT used were: (1) 10 units packed red blood cells (PRBCs) in 24 hours, (2) Resuscitation Intensity score ≥ 4, (3) critical administration threshold, (4) 4 units PRBCs in 4 hours; and (5) 6 units PRBCs in 6 hours. Receiver operating curves were plotted to compare the predicted probability of MT with observed outcomes. RESULTS Of 1,245 patients in the data set, 297 (24%) met definition 1, 570 (47%) met definition 2, 364 (33%) met definition 3, 599 met definition 4 (49.1%), and 395 met definition 5 (32.4%). Regardless of the outcome (MTP activation or MT administration), the predictive ability of the app model was consistent: when predicting activation of the MTP, the area under the curve for the model was 0.694 and when predicting MT administration, the area under the curve ranged from 0.695 to 0.711. CONCLUSION Regardless of the definition of MT used, the app model demonstrates moderate ability to predict the need for MT in an external, homogenous population. Importantly, the app allows the model to be iteratively recalibrated ("machine learning") and thus could improve its predictive capability as additional data are accrued. LEVEL OF EVIDENCE Diagnostic test study/Prognostic study, level III.

Original languageEnglish (US)
Pages (from-to)397-402
Number of pages6
JournalJournal of Trauma and Acute Care Surgery
Volume84
Issue number2
DOIs
StatePublished - Feb 1 2018

Fingerprint

Erythrocytes
Area Under Curve
Routine Diagnostic Tests
Resuscitation
Databases
Wounds and Injuries
Population
Smartphone
Datasets
Machine Learning

Keywords

  • Massive transfusion
  • prediction model
  • smartphone application
  • trauma

ASJC Scopus subject areas

  • Surgery
  • Critical Care and Intensive Care Medicine

Cite this

External validation of a smartphone app model to predict the need for massive transfusion using five different definitions. / Hodgman, E. I.; Cripps, Michael W; Mina, M. J.; Bulger, E. M.; Schreiber, M. A.; Brasel, K. J.; Cohen, M. J.; Muskat, P.; Myers, J. G.; Alarcon, L. H.; Rahbar, M. H.; Holcomb, J. B.; Cotton, B. A.; Fox, E. E.; Del Junco, D. J.; Wade, C. E.; Phelan, Herbert.

In: Journal of Trauma and Acute Care Surgery, Vol. 84, No. 2, 01.02.2018, p. 397-402.

Research output: Contribution to journalArticle

Hodgman, EI, Cripps, MW, Mina, MJ, Bulger, EM, Schreiber, MA, Brasel, KJ, Cohen, MJ, Muskat, P, Myers, JG, Alarcon, LH, Rahbar, MH, Holcomb, JB, Cotton, BA, Fox, EE, Del Junco, DJ, Wade, CE & Phelan, H 2018, 'External validation of a smartphone app model to predict the need for massive transfusion using five different definitions', Journal of Trauma and Acute Care Surgery, vol. 84, no. 2, pp. 397-402. https://doi.org/10.1097/TA.0000000000001756
Hodgman, E. I. ; Cripps, Michael W ; Mina, M. J. ; Bulger, E. M. ; Schreiber, M. A. ; Brasel, K. J. ; Cohen, M. J. ; Muskat, P. ; Myers, J. G. ; Alarcon, L. H. ; Rahbar, M. H. ; Holcomb, J. B. ; Cotton, B. A. ; Fox, E. E. ; Del Junco, D. J. ; Wade, C. E. ; Phelan, Herbert. / External validation of a smartphone app model to predict the need for massive transfusion using five different definitions. In: Journal of Trauma and Acute Care Surgery. 2018 ; Vol. 84, No. 2. pp. 397-402.
@article{da064569de664ccf976efcb47b127c61,
title = "External validation of a smartphone app model to predict the need for massive transfusion using five different definitions",
abstract = "BACKGROUND Previously, a model to predict massive transfusion protocol (MTP) (activation) was derived using a single-institution data set. The PRospective, Observational, Multicenter, Major Trauma Transfusion database was used to externally validate this model's ability to predict both MTP activation and massive transfusion (MT) administration using multiple MT definitions. METHODS The app model was used to calculate the predicted probability of MTP activation or MT delivery. The five definitions of MT used were: (1) 10 units packed red blood cells (PRBCs) in 24 hours, (2) Resuscitation Intensity score ≥ 4, (3) critical administration threshold, (4) 4 units PRBCs in 4 hours; and (5) 6 units PRBCs in 6 hours. Receiver operating curves were plotted to compare the predicted probability of MT with observed outcomes. RESULTS Of 1,245 patients in the data set, 297 (24{\%}) met definition 1, 570 (47{\%}) met definition 2, 364 (33{\%}) met definition 3, 599 met definition 4 (49.1{\%}), and 395 met definition 5 (32.4{\%}). Regardless of the outcome (MTP activation or MT administration), the predictive ability of the app model was consistent: when predicting activation of the MTP, the area under the curve for the model was 0.694 and when predicting MT administration, the area under the curve ranged from 0.695 to 0.711. CONCLUSION Regardless of the definition of MT used, the app model demonstrates moderate ability to predict the need for MT in an external, homogenous population. Importantly, the app allows the model to be iteratively recalibrated ({"}machine learning{"}) and thus could improve its predictive capability as additional data are accrued. LEVEL OF EVIDENCE Diagnostic test study/Prognostic study, level III.",
keywords = "Massive transfusion, prediction model, smartphone application, trauma",
author = "Hodgman, {E. I.} and Cripps, {Michael W} and Mina, {M. J.} and Bulger, {E. M.} and Schreiber, {M. A.} and Brasel, {K. J.} and Cohen, {M. J.} and P. Muskat and Myers, {J. G.} and Alarcon, {L. H.} and Rahbar, {M. H.} and Holcomb, {J. B.} and Cotton, {B. A.} and Fox, {E. E.} and {Del Junco}, {D. J.} and Wade, {C. E.} and Herbert Phelan",
year = "2018",
month = "2",
day = "1",
doi = "10.1097/TA.0000000000001756",
language = "English (US)",
volume = "84",
pages = "397--402",
journal = "Journal of Trauma and Acute Care Surgery",
issn = "2163-0755",
publisher = "Lippincott Williams and Wilkins",
number = "2",

}

TY - JOUR

T1 - External validation of a smartphone app model to predict the need for massive transfusion using five different definitions

AU - Hodgman, E. I.

AU - Cripps, Michael W

AU - Mina, M. J.

AU - Bulger, E. M.

AU - Schreiber, M. A.

AU - Brasel, K. J.

AU - Cohen, M. J.

AU - Muskat, P.

AU - Myers, J. G.

AU - Alarcon, L. H.

AU - Rahbar, M. H.

AU - Holcomb, J. B.

AU - Cotton, B. A.

AU - Fox, E. E.

AU - Del Junco, D. J.

AU - Wade, C. E.

AU - Phelan, Herbert

PY - 2018/2/1

Y1 - 2018/2/1

N2 - BACKGROUND Previously, a model to predict massive transfusion protocol (MTP) (activation) was derived using a single-institution data set. The PRospective, Observational, Multicenter, Major Trauma Transfusion database was used to externally validate this model's ability to predict both MTP activation and massive transfusion (MT) administration using multiple MT definitions. METHODS The app model was used to calculate the predicted probability of MTP activation or MT delivery. The five definitions of MT used were: (1) 10 units packed red blood cells (PRBCs) in 24 hours, (2) Resuscitation Intensity score ≥ 4, (3) critical administration threshold, (4) 4 units PRBCs in 4 hours; and (5) 6 units PRBCs in 6 hours. Receiver operating curves were plotted to compare the predicted probability of MT with observed outcomes. RESULTS Of 1,245 patients in the data set, 297 (24%) met definition 1, 570 (47%) met definition 2, 364 (33%) met definition 3, 599 met definition 4 (49.1%), and 395 met definition 5 (32.4%). Regardless of the outcome (MTP activation or MT administration), the predictive ability of the app model was consistent: when predicting activation of the MTP, the area under the curve for the model was 0.694 and when predicting MT administration, the area under the curve ranged from 0.695 to 0.711. CONCLUSION Regardless of the definition of MT used, the app model demonstrates moderate ability to predict the need for MT in an external, homogenous population. Importantly, the app allows the model to be iteratively recalibrated ("machine learning") and thus could improve its predictive capability as additional data are accrued. LEVEL OF EVIDENCE Diagnostic test study/Prognostic study, level III.

AB - BACKGROUND Previously, a model to predict massive transfusion protocol (MTP) (activation) was derived using a single-institution data set. The PRospective, Observational, Multicenter, Major Trauma Transfusion database was used to externally validate this model's ability to predict both MTP activation and massive transfusion (MT) administration using multiple MT definitions. METHODS The app model was used to calculate the predicted probability of MTP activation or MT delivery. The five definitions of MT used were: (1) 10 units packed red blood cells (PRBCs) in 24 hours, (2) Resuscitation Intensity score ≥ 4, (3) critical administration threshold, (4) 4 units PRBCs in 4 hours; and (5) 6 units PRBCs in 6 hours. Receiver operating curves were plotted to compare the predicted probability of MT with observed outcomes. RESULTS Of 1,245 patients in the data set, 297 (24%) met definition 1, 570 (47%) met definition 2, 364 (33%) met definition 3, 599 met definition 4 (49.1%), and 395 met definition 5 (32.4%). Regardless of the outcome (MTP activation or MT administration), the predictive ability of the app model was consistent: when predicting activation of the MTP, the area under the curve for the model was 0.694 and when predicting MT administration, the area under the curve ranged from 0.695 to 0.711. CONCLUSION Regardless of the definition of MT used, the app model demonstrates moderate ability to predict the need for MT in an external, homogenous population. Importantly, the app allows the model to be iteratively recalibrated ("machine learning") and thus could improve its predictive capability as additional data are accrued. LEVEL OF EVIDENCE Diagnostic test study/Prognostic study, level III.

KW - Massive transfusion

KW - prediction model

KW - smartphone application

KW - trauma

UR - http://www.scopus.com/inward/record.url?scp=85041627095&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85041627095&partnerID=8YFLogxK

U2 - 10.1097/TA.0000000000001756

DO - 10.1097/TA.0000000000001756

M3 - Article

VL - 84

SP - 397

EP - 402

JO - Journal of Trauma and Acute Care Surgery

JF - Journal of Trauma and Acute Care Surgery

SN - 2163-0755

IS - 2

ER -