The impact of missing trauma data on predicting massive transfusion

BACKGROUND Missing data are inherent in clinical research and may be especially problematic for trauma studies. This study describes a sensitivity analysis to evaluate the impact of missing data on clinical risk prediction algorithms. Three blood transfusion prediction models were evaluated using an observational trauma data set with valid missing data. METHODS The PRospective Observational Multicenter Major 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. MT was received by 297 patients (24%). Missing percentage ranged from 2.2% (heart rate) to 45% (respiratory rate). Proportions of complete cases used in the MT prediction models ranged from 41% to 88%. All models demonstrated similar correct classification percentages using complete case analysis and multiple imputation. In the sensitivity analysis, correct classification upper-lower bound ranges per model were 4%, 10%, and 12%. Predictive accuracy for all models using PROMMTT data was 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.

[1]  S. Wilson,et al.  Improved survival following massive transfusion in patients who have undergone trauma. , 1999, Archives of surgery.

[2]  R. Wilson,et al.  Eight years of experience with massive blood transfusions. , 1971, The Journal of trauma.

[3]  O. Åkerblom,et al.  Cumulative risks of early fresh frozen plasma, cryoprecipitate and platelet transfusion in Europe. , 2006, The Journal of trauma.

[4]  Jose Salinas,et al.  A predictive model for massive transfusion in combat casualty patients. , 2008, The Journal of trauma.

[5]  B. Cotton,et al.  Damage control hematology: the impact of a trauma exsanguination protocol on survival and blood product utilization. , 2008, The Journal of trauma.

[6]  T. O’Keeffe,et al.  A massive transfusion protocol to decrease blood component use and costs. , 2008, Archives of surgery.

[7]  J. Sampalis,et al.  Selection bias found in interpreting analyses with missing data for the prehospital index for trauma. , 2004, Journal of clinical epidemiology.

[8]  C. Wade,et al.  Early predictors of massive transfusion in combat casualties. , 2007, Journal of the American College of Surgeons.

[9]  J. Holcomb,et al.  Emergency department blood transfusion predicts early massive transfusion and early blood component requirement , 2010, Transfusion.

[10]  Roderick J. A. Little.  Regression with Missing X's: A Review , 1992 .

[11]  A. Ho,et al.  Early risk stratification of patients with major trauma requiring massive blood transfusion. , 2011, Resuscitation.

[12]  L. Napolitano.  Cumulative risks of early red blood cell transfusion. , 2006, The Journal of trauma.

[13]  E. Bulger,et al.  The prospective, observational, multicenter, major trauma transfusion (PROMMTT) study: comparative effectiveness of a time-varying treatment with competing risks. , 2013, JAMA surgery.

[14]  John A. Jones,et al.  Association of shock, coagulopathy, and initial vital signs with massive transfusion in combat casualties. , 2010, The Journal of trauma.

[15]  T. Osler,et al.  Impact of Statistical Approaches for Handling Missing Data on Trauma Center Quality , 2009, Annals of Surgery.

[16]  D. Fairclough,et al.  Why are missing quality of life data a problem in clinical trials of cancer therapy? , 1998, Statistics in medicine.

[17]  R. Lefering,et al.  Trauma Associated Severe Hemorrhage (TASH)-Score: probability of mass transfusion as surrogate for life threatening hemorrhage after multiple trauma. , 2006, The Journal of trauma.

[18]  O. Bouamra,et al.  A new approach to outcome prediction in trauma: A comparison with the TRISS model. , 2006, The Journal of trauma.

[19]  P. Royston,et al.  Patrick Royston model with a binary outcome A comparison of imputation techniques for handling missing predictor values in a risk , 2007 .

[20]  L. Blackbourne,et al.  The effect of FFP:RBC ratio on morbidity and mortality in trauma patients based on transfusion prediction score , 2011, Vox sanguinis.

[21]  Lesly A. Dossett,et al.  Early prediction of massive transfusion in trauma: simple as ABC (assessment of blood consumption)? , 2009, The Journal of trauma.

[22]  S. Maclennan,et al.  Risks of fresh frozen plasma and platelets. , 2006, The Journal of trauma.

[23]  Peter Rhee,et al.  Preventable or potentially preventable mortality at a mature trauma center. , 2007, The Journal of trauma.

[24]  Leopoldo C Cancio,et al.  Prediction of mortality and of the need for massive transfusion in casualties arriving at combat support hospitals in Iraq. , 2008, The Journal of trauma.

[25]  Jiajie Zhang,et al.  Coordination and management of multicenter clinical studies in trauma: Experience from the PRospective Observational Multicenter Major Trauma Transfusion (PROMMTT) Study. , 2012, Resuscitation.

[26]  R. Maier,et al.  Patterns of Errors Contributing to Trauma Mortality: Lessons Learned From 2594 Deaths , 2006, Annals of surgery.

[27]  Richard L. George,et al.  The relationship of blood product ratio to mortality: survival benefit or survival bias? , 2009, The Journal of trauma.

[28]  Michael A Dubick,et al.  Damage control resuscitation: directly addressing the early coagulopathy of trauma. , 2007, The Journal of trauma.

[29]  J. Como,et al.  Blood transfusion rates in the care of acute trauma , 2004, Transfusion.

[30]  C. Hillyer,et al.  Transfusion Management of Trauma Patients , 2009, Anesthesia and analgesia.

[31]  Qingxia Chen,et al.  Missing covariate data in medical research: to impute is better than to ignore. , 2010, Journal of clinical epidemiology.

[32]  J L Schafer,et al.  Multiple Imputation for Multivariate Missing-Data Problems: A Data Analyst's Perspective. , 1998, Multivariate behavioral research.

[33]  R. Lefering,et al.  Revalidation and update of the TASH‐Score: a scoring system to predict the probability for massive transfusion as a surrogate for life‐threatening haemorrhage after severe injury , 2011, Vox sanguinis.