Preoperative Prediction of Postoperative Infections Using Machine Learning and Electronic Health Record Data.
Journal
Annals of surgery
ISSN: 1528-1140
Titre abrégé: Ann Surg
Pays: United States
ID NLM: 0372354
Informations de publication
Date de publication:
01 Apr 2024
01 Apr 2024
Historique:
medline:
11
3
2024
pubmed:
27
9
2023
entrez:
27
9
2023
Statut:
ppublish
Résumé
To estimate preoperative risk of postoperative infections using structured electronic health record (EHR) data. Surveillance and reporting of postoperative infections is primarily done through costly, labor-intensive manual chart reviews on a small sample of patients. Automated methods using statistical models applied to postoperative EHR data have shown promise to augment manual review as they can cover all operations in a timely manner. However, there are no specific models for risk-adjusting infectious complication rates using EHR data. Preoperative EHR data from 30,639 patients (2013-2019) were linked to the American College of Surgeons National Surgical Quality Improvement Program preoperative data and postoperative infection outcomes data from 5 hospitals in the University of Colorado Health System. EHR data included diagnoses, procedures, operative variables, patient characteristics, and medications. Lasso and the knockoff filter were used to perform controlled variable selection. Outcomes included surgical site infection, urinary tract infection, sepsis/septic shock, and pneumonia up to 30 days postoperatively. Among >15,000 candidate predictors, 7 were chosen for the surgical site infection model and 6 for each of the urinary tract infection, sepsis, and pneumonia models. Important variables included preoperative presence of the specific outcome, wound classification, comorbidities, and American Society of Anesthesiologists physical status classification. The area under the receiver operating characteristic curve for each model ranged from 0.73 to 0.89. Parsimonious preoperative models for predicting postoperative infection risk using EHR data were developed and showed comparable performance to existing American College of Surgeons National Surgical Quality Improvement Program risk models that use manual chart review. These models can be used to estimate risk-adjusted postoperative infection rates applied to large volumes of EHR data in a timely manner.
Sections du résumé
OBJECTIVE
OBJECTIVE
To estimate preoperative risk of postoperative infections using structured electronic health record (EHR) data.
BACKGROUND
BACKGROUND
Surveillance and reporting of postoperative infections is primarily done through costly, labor-intensive manual chart reviews on a small sample of patients. Automated methods using statistical models applied to postoperative EHR data have shown promise to augment manual review as they can cover all operations in a timely manner. However, there are no specific models for risk-adjusting infectious complication rates using EHR data.
METHODS
METHODS
Preoperative EHR data from 30,639 patients (2013-2019) were linked to the American College of Surgeons National Surgical Quality Improvement Program preoperative data and postoperative infection outcomes data from 5 hospitals in the University of Colorado Health System. EHR data included diagnoses, procedures, operative variables, patient characteristics, and medications. Lasso and the knockoff filter were used to perform controlled variable selection. Outcomes included surgical site infection, urinary tract infection, sepsis/septic shock, and pneumonia up to 30 days postoperatively.
RESULTS
RESULTS
Among >15,000 candidate predictors, 7 were chosen for the surgical site infection model and 6 for each of the urinary tract infection, sepsis, and pneumonia models. Important variables included preoperative presence of the specific outcome, wound classification, comorbidities, and American Society of Anesthesiologists physical status classification. The area under the receiver operating characteristic curve for each model ranged from 0.73 to 0.89.
CONCLUSIONS
CONCLUSIONS
Parsimonious preoperative models for predicting postoperative infection risk using EHR data were developed and showed comparable performance to existing American College of Surgeons National Surgical Quality Improvement Program risk models that use manual chart review. These models can be used to estimate risk-adjusted postoperative infection rates applied to large volumes of EHR data in a timely manner.
Identifiants
pubmed: 37753703
doi: 10.1097/SLA.0000000000006106
pii: 00000658-990000000-00654
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
720-726Subventions
Organisme : NLM NIH HHS
ID : L30 LM014131
Pays : United States
Organisme : NHLBI NIH HHS
ID : L40 HL154485
Pays : United States
Organisme : AHRQ HHS
ID : R01 HS027417
Pays : United States
Informations de copyright
Copyright © 2023 Wolters Kluwer Health, Inc. All rights reserved.
Déclaration de conflit d'intérêts
The authors report no conflicts of interest.
Références
Aasen DM, Bronsert MR, Rozeboom PD, et al. Relationships between predischarge and postdischarge infectious complications, length of stay, and unplanned readmissions in the ACS NSQIP database. Surgery. 2021;169:325–332.
Herwaldt LA, Cullen JJ, Scholz D, et al. A prospective study of outcomes, healthcare resource utilization, and costs associated with postoperative nosocomial infections. Infect Control Hosp Epidemiol. 2006;27:1291–1298.
Branch-Elliman W, Strymish J, Itani KM, et al. Using clinical variables to guide surgical site infection detection: a novel surveillance strategy. Am J Infect Control. 2014;42:1291–1295.
Branch-Elliman W, Strymish J, Kudesia V, et al. Natural language processing for real-time catheter-associated urinary tract infection surveillance: results of a pilot implementation trial. Infect Control Hosp Epidemiol. 2015;36:1004–1010.
Bronsert M, Singh AB, Henderson WG, et al. Identification of postoperative complications using electronic health record data and machine learning. Am J Surg. 2020;220:114–119.
Colborn KL, Bronsert M, Amioka E, et al. Identification of surgical site infections using electronic health record data. Am J Infect Control. 2018;46:1230–1235.
Colborn KL, Bronsert M, Hammermeister K, et al. Identification of urinary tract infections using electronic health record data. Am J Infect Control. 2018;47:371–375.
FitzHenry F, Murff HJ, Matheny ME, et al. Exploring the frontier of electronic health record surveillance: the case of postoperative complications. Med Care. 2013;51:509–516.
Goto M, Ohl ME, Schweizer ML, et al. Accuracy of administrative code data for the surveillance of healthcare-associated infections: a systematic review and meta-analysis. Clin Infect Dis. 2014;58:688–696.
Gundlapalli AV, Divita G, Redd A, et al. Detecting the presence of an indwelling urinary catheter and urinary symptoms in hospitalized patients using natural language processing. J Biomed Inform. 2017;71s:S39–S45.
Hsu HE, Shenoy ES, Kelbaugh D, et al. An electronic surveillance tool for catheter-associated urinary tract infection in intensive care units. Am J Infect Control. 2015;43:592–599.
Hu Z, Simon GJ, Arsoniadis EG, et al. Automated detection of postoperative surgical site infections using supervised methods with electronic health record data. Stud Health Technol Inform. 2015;216:706–710.
Murff HJ, FitzHenry F, Matheny ME, et al. Automated identification of postoperative complications within an electronic medical record using natural language processing. JAMA. 2011;306:848–855.
Colborn KL, Zhuang Y, Dyas AR, et al. Development and validation of models for detection of postoperative infections using structured electronic health records data and machine learning. Surgery. 2023;173:464–471.
Wei WQ, Bastarache LA, Carroll RJ, et al. Evaluating phecodes, clinical classification software, and ICD-9-CM codes for phenome-wide association studies in the electronic health record. PLoS One. 2017;12:e0175508.
Wu P, Gifford A, Meng X, et al. Mapping ICD-10 and ICD-10-CM codes to Phecodes: workflow development and initial evaluation. JMIR Med Inform. 2019;7:e14325.
Charlson M, Szatrowski TP, Peterson J, et al. Validation of a combined comorbidity index. J Clin Epidemiol. 1994;47:1245–1251.
American College of Surgeons. Program ACoSNSQI. User Guide for the 2016 ACS NSQIP Participant Use Data File (PUF). 2016. Accessed December 10, 2021. https://www.facs.org/media/kthpmx3h/nsqip_puf_userguide_2016.pdf
Meguid RA, Bronsert MR, Juarez-Colunga E, et al. Surgical Risk Preoperative Assessment System (SURPAS): II. Parsimonious risk models for postoperative adverse outcomes addressing need for laboratory variables and surgeon specialty-specific models. Ann Surg. 2016;264:10–22.
Moons KG, Altman DG, Reitsma JB, et al. Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis (TRIPOD): explanation and elaboration. Ann Intern Med. 2015;162:W1–W73.
Barber RF, Candes EJ. Controlling the false discovery rate via knockoffs. Ann Stat. 2015;43:2055–2085.
Candes E, Fan Y, Janson L, et al. Panning for gold: ‘model-X’ knockoffs for high dimensional controlled variable selection. J Royal Stat Soc Series B. 2018;80:551–577.
Dyas AR, Zhuang Y, Meguid RA, et al. Development and validation of a model for surveillance of postoperative bleeding complications using structured electronic health records data. Surgery. 2022;172:1728–1732.
Barber RF, Candes E, Janson L, et al. knockoff: the knockoff filter for controlled variable selection. Version R package version 0.3.3. 2020. https://cran.r-project.org/web/packages/knockoff/index.html
Friedman J, Hastie T, Tibshirani R. Regularization paths for generalized linear models via coordinate descent. J Stat Softw. 2010;33:1–22.
Robin X, Turck N, Hainard A, et al. pROC: an open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinformatics. 2011;12:77.
Gasparini A. Comorbidity: an R package for computing comorbidity scores. J Open Source Software. 2018;3:648.
Meguid RA, Bronsert MR, Juarez-Colunga E, et al. Surgical Risk Preoperative Assessment System (SURPAS): III. Accurate preoperative prediction of 8 adverse outcomes using 8 predictor variables. Ann Surg. 2016;264:23–31.
Corey KM, Kashyap S, Lorenzi E, et al. Development and validation of machine learning models to identify high-risk surgical patients using automatically curated electronic health record data (Pythia): a retrospective, single-site study. PLoS Med. 2018;15:e1002701.
Weller GB, Lovely J, Larson DW, et al. Leveraging electronic health records for predictive modeling of post-surgical complications. Stat Methods Med Res. 2017;27:3271–3285.
Xue B, Li D, Lu C, et al. Use of machine learning to develop and evaluate models using preoperative and intraoperative data to identify risks of postoperative complications. JAMA Netw Open. 2021;4:e212240.
Bihorac A, Ozrazgat-Baslanti T, Ebadi A, et al. My Surgery Risk: development and validation of a machine-learning risk algorithm for major complications and death after surgery. Ann Surg. 2019;269:652–662.
Henderson WG, Bronsert MR, Hammermeister KE, et al. Refining the predictive variables in the “Surgical Risk Preoperative Assessment System” (SURPAS): a descriptive analysis. Patient Saf Surg. 2019;13:28.
Harth KC, Blatnik JA, Anderson JM, et al. Effect of surgical wound classification on biologic graft performance in complex hernia repair: an experimental study. Surgery. 2013;153:481–492.
Mioton LM, Jordan SW, Hanwright PJ, et al. The relationship between preoperative wound classification and postoperative infection: a multi-institutional analysis of 15,289 patients. Arch Plast Surg. 2013;40:522–529.
Onyekwelu I, Yakkanti R, Protzer L, et al. Surgical wound classification and surgical site infections in the orthopaedic patient. J Am Acad Orthop Surg Glob Res Rev. 2017;1:e022.
Bronsert MR, Henderson WG, Colborn KL, et al. Effect of present at time of surgery on unadjusted and risk-adjusted postoperative complication rates. J Am Coll Surg. 2023;236:7–15.
Helkin A, Jain SV, Gruessner A, et al. Impact of ASA score misclassification on NSQIP predicted mortality: a retrospective analysis. Perioper Med (Lond). 2017;6:23.
Gorvetzian JW, Epler KE, Schrader S, et al. Operating room staff and surgeon documentation curriculum improves wound classification accuracy. Heliyon. 2018;4:e00728.
Levy SM, Lally KP, Blakely ML, et al. Surgical wound misclassification: a multicenter evaluation. J Am Coll Surg. 2015;220:323–329.
Singh S, Podila S, Pyon G, et al. An analysis of 3,954 cases to determine surgical wound classification accuracy: does your institution need a Monday morning quarterback. Am J Surg. 2020;220:1115–1118.
Stefanou A, Worden A, Kandagatla P, et al. Surgical wound misclassification to clean from clean-contaminated in common abdominal operations. J Surg Res. 2020;246:131–138.
Davis SE, Lasko TA, Chen G, et al. Calibration drift in regression and machine learning models for acute kidney injury. J Am Med Inform Assoc. 2017;24:1052–1061.