Nishant Sahni1, Roshan Tourani2, Donald Sullivan3, Gyorgy Simon2. 1. Division of General Internal Medicine, University of Minnesota, Delaware Street SE, MMC 741, Minneapolis, MN, USA. nishant.sahni@gmail.com. 2. Institute of Health Informatics, University of Minnesota, Minneapolis, MN, USA. 3. Oregon Health Sciences University, Portland, OR, USA.
Abstract
BACKGROUND: Predicting death in a cohort of clinically diverse, multi-condition hospitalized patients is difficult. This frequently hinders timely serious illness care conversations. Prognostic models that can determine 6-month death risk at the time of hospital admission can improve access to serious illness care conversations. OBJECTIVE: The objective is to determine if the demographic, vital sign, and laboratory data from the first 48 h of a hospitalization can be used to accurately quantify 6-month mortality risk. DESIGN: This is a retrospective study using electronic medical record data linked with the state death registry. PARTICIPANTS: Participants were 158,323 hospitalized patients within a 6-hospital network over a 6-year period. MAIN MEASURES: Main measures are the following: the first set of vital signs, complete blood count, basic and complete metabolic panel, serum lactate, pro-BNP, troponin-I, INR, aPTT, demographic information, and associated ICD codes. The outcome of interest was death within 6 months. KEY RESULTS: Model performance was measured on the validation dataset. A random forest model-mini serious illness algorithm-used 8 variables from the initial 48 h of hospitalization and predicted death within 6 months with an AUC of 0.92 (0.91-0.93). Red cell distribution width was the most important prognostic variable. min-SIA (mini serious illness algorithm) was very well calibrated and estimated the probability of death to within 10% of the actual value. The discriminative ability of the min-SIA was significantly better than historical estimates of clinician performance. CONCLUSION: min-SIA algorithm can identify patients at high risk of 6-month mortality at the time of hospital admission. It can be used to improved access to timely, serious illness care conversations in high-risk patients.
BACKGROUND: Predicting death in a cohort of clinically diverse, multi-condition hospitalized patients is difficult. This frequently hinders timely serious illness care conversations. Prognostic models that can determine 6-month death risk at the time of hospital admission can improve access to serious illness care conversations. OBJECTIVE: The objective is to determine if the demographic, vital sign, and laboratory data from the first 48 h of a hospitalization can be used to accurately quantify 6-month mortality risk. DESIGN: This is a retrospective study using electronic medical record data linked with the state death registry. PARTICIPANTS: Participants were 158,323 hospitalized patients within a 6-hospital network over a 6-year period. MAIN MEASURES: Main measures are the following: the first set of vital signs, complete blood count, basic and complete metabolic panel, serum lactate, pro-BNP, troponin-I, INR, aPTT, demographic information, and associated ICD codes. The outcome of interest was death within 6 months. KEY RESULTS: Model performance was measured on the validation dataset. A random forest model-mini serious illness algorithm-used 8 variables from the initial 48 h of hospitalization and predicted death within 6 months with an AUC of 0.92 (0.91-0.93). Red cell distribution width was the most important prognostic variable. min-SIA (mini serious illness algorithm) was very well calibrated and estimated the probability of death to within 10% of the actual value. The discriminative ability of the min-SIA was significantly better than historical estimates of clinician performance. CONCLUSION: min-SIA algorithm can identify patients at high risk of 6-month mortality at the time of hospital admission. It can be used to improved access to timely, serious illness care conversations in high-risk patients.
Entities:
Keywords:
data mining; hospital outcomes; palliative care; predictive models
Authors: J Randall Curtis; Lois Downey; Anthony L Back; Elizabeth L Nielsen; Sudiptho Paul; Alexandria Z Lahdya; Patsy D Treece; Priscilla Armstrong; Ronald Peck; Ruth A Engelberg Journal: JAMA Intern Med Date: 2018-07-01 Impact factor: 21.873
Authors: John J You; James Downar; Robert A Fowler; François Lamontagne; Irene W Y Ma; Dev Jayaraman; Jennifer Kryworuchko; Patricia H Strachan; Roy Ilan; Aman P Nijjar; John Neary; John Shik; Kevin Brazil; Amen Patel; Kim Wiebe; Martin Albert; Anita Palepu; Elysée Nouvet; Amanda Roze des Ordons; Nishan Sharma; Amane Abdul-Razzak; Xuran Jiang; Andrew Day; Daren K Heyland Journal: JAMA Intern Med Date: 2015-04 Impact factor: 21.873
Authors: Michael E Detsky; Michael O Harhay; Dominique F Bayard; Aaron M Delman; Anna E Buehler; Saida A Kent; Isabella V Ciuffetelli; Elizabeth Cooney; Nicole B Gabler; Sarah J Ratcliffe; Mark E Mikkelsen; Scott D Halpern Journal: JAMA Date: 2017-06-06 Impact factor: 56.272
Authors: Rachelle Bernacki; Mathilde Hutchings; Judith Vick; Grant Smith; Joanna Paladino; Stuart Lipsitz; Atul A Gawande; Susan D Block Journal: BMJ Open Date: 2015-10-06 Impact factor: 2.692