Hostname: page-component-76fb5796d-zzh7m Total loading time: 0 Render date: 2024-04-28T06:55:02.377Z Has data issue: false hasContentIssue false
  • English
  • Français

A methodological comparison of risk scores versus decision trees for predicting drug-resistant infections: A case study using extended-spectrum beta-lactamase (ESBL) bacteremia

Published online by Cambridge University Press:  04 March 2019

Katherine E. Goodman Open the ORCID record for Katherine E. Goodman [Opens in a new window] ,
Justin Lessler ,
Anthony D. Harris ,
Aaron M. Milstone  and
Pranita D. Tamma
Katherine E. Goodman*
Affiliation:
Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland
Justin Lessler
Affiliation:
Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland
Anthony D. Harris
Affiliation:
Department of Epidemiology and Public Health, University of Maryland School of Medicine, Baltimore, Maryland
Aaron M. Milstone
Affiliation:
Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland Division of Infectious Diseases, Department of Pediatrics, Johns Hopkins University School of Medicine, Baltimore, Maryland
Pranita D. Tamma
Affiliation:
Division of Infectious Diseases, Department of Pediatrics, Johns Hopkins University School of Medicine, Baltimore, Maryland
*
Author for correspondence: Katherine E. Goodman, Email: kgoodma7@jhu.edu and Pranita D. Tamma, Email: ptamma1@jhmi.edu
Get access Rights & Permissions [Opens in a new window]

Abstract

Background:

Timely identification of multidrug-resistant gram-negative infections remains an epidemiological challenge. Statistical models for predicting drug resistance can offer utility where rapid diagnostics are unavailable or resource-impractical. Logistic regression–derived risk scores are common in the healthcare epidemiology literature. Machine learning–derived decision trees are an alternative approach for developing decision support tools. Our group previously reported on a decision tree for predicting ESBL bloodstream infections. Our objective in the current study was to develop a risk score from the same ESBL dataset to compare these 2 methods and to offer general guiding principles for using each approach.

Methods:

Using a dataset of 1,288 patients with Escherichia coli or Klebsiella spp bacteremia, we generated a risk score to predict the likelihood that a bacteremic patient was infected with an ESBL-producer. We evaluated discrimination (original and cross-validated models) using receiver operating characteristic curves and C statistics. We compared risk score and decision tree performance, and we reviewed their practical and methodological attributes.

Results:

In total, 194 patients (15%) were infected with ESBL-producing bacteremia. The clinical risk score included 14 variables, compared to the 5 decision-tree variables. The positive and negative predictive values of the risk score and decision tree were similar (>90%), but the C statistic of the risk score (0.87) was 10% higher.

Conclusions:

A decision tree and risk score performed similarly for predicting ESBL infection. The decision tree was more user-friendly, with fewer variables for the end user, whereas the risk score offered higher discrimination and greater flexibility for adjusting sensitivity and specificity.

Type
Original Article
Information
Infection Control & Hospital Epidemiology , Volume 40 , Issue 4 , April 2019 , pp. 400 - 407
Copyright
© 2019 by The Society for Healthcare Epidemiology of America. All rights reserved. 

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

Livorsi, DJ, Chorazy, ML, Schweizer, ML, et al. A systematic review of the epidemiology of carbapenem-resistant Enterobacteriaceae in the United States. Antimicrob Resist Infect Control. 2018;7:55.CrossRefGoogle ScholarPubMed
McDanel, J, Schweizer, M, Crabb, V, et al. Incidence of extended-spectrum beta-lactamase (ESBL)-producing Escherichia coli and Klebsiella infections in the United States: a systematic literature review. Infect Control Hosp Epidemiol 2017;38:1209–1215.CrossRefGoogle ScholarPubMed
Micek, ST, Hampton, N, Kollef, M. Risk factors and outcomes for ineffective empiric treatment of sepsis caused by gram-negative pathogens: stratification by onset of infection. Antimicrob Agents Chemother 2018;62(1): e01577–17.CrossRefGoogle Scholar
Zhang, D, Micek, ST, Kollef, MH. Time to appropriate antibiotic therapy is an independent determinant of postinfection ICU and hospital lengths of stay in patients with sepsis. Crit Care Med 2015;43:2133–2140.CrossRefGoogle ScholarPubMed
Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing, 28th ed. Supplement M100S. Wayne, PA: CLSI; 2018.Google Scholar
Ledeboer, NA, Lopansri, BK, Dhiman, N, et al. Identification of gram-negative bacteria and genetic resistance determinants from positive blood culture broths by use of the verigene gram-negative blood culture multiplex microarray-based molecular assay. J Clin Microbiol 2015;53:2460–2472.CrossRefGoogle ScholarPubMed
Ward, C, Stocker, K, Begum, J, Wade, P, Ebrahimsa, U, Goldenberg, SD. Performance evaluation of the Verigene (Nanosphere) and FilmArray (BioFire) molecular assays for identification of causative organisms in bacterial bloodstream infections. Eur J Clin Microbiol Infect Dis 2015;34:487–496.CrossRefGoogle ScholarPubMed
Goodman, KE, Lessler, J, Cosgrove, SE, et al. A clinical decision tree to predict whether a bacteremic patient is infected with an extended-spectrum beta-lactamase–producing organism. Clin Infect Dis 2016;63:896–903.10.1093/cid/ciw425CrossRefGoogle ScholarPubMed
Antimicrobial resistant phenotype definitions. Centers for Disease Control and Prevention webstie. https://www.cdc.gov/nhsn/pdfs/ps-analysis-resources/phenotype_definitions.pdf. Published 2016. Accessed January 15, 2019.Google Scholar
Kantele, A, Laaveri, T, Mero, S, et al. Antimicrobials increase travelers’ risk of colonization by extended-spectrum beta-lactamase–producing Enterobacteriaceae. Clin Infect Dis 2015;60:837–846.CrossRefGoogle Scholar
Ostholm-Balkhed, A, Tarnberg, M, Nilsson, M, et al. Travel-associated faecal colonization with ESBL-producing Enterobacteriaceae: incidence and risk factors. J Antimicrob Chemother 2013;68:2144–2153.CrossRefGoogle ScholarPubMed
Duda, RO, Hart, PE, Stork, DG. Pattern Classification, 2nd ed. New York: Wiley-Interscience; 2001.Google Scholar
Breiman, L, Friedman, J, Stone, C, Olshen, R. Classification and Regression Trees. Boca Raton, FL: CRC/Chapman & Hall; 1984.Google Scholar
Tseng, WP, Chen, YC, Yang, BJ, et al. Predicting multidrug-resistant gram-negative bacterial colonization and associated infection on hospital admission. Infect Control Hosp Epidemiol 2017;38:1216–1225.CrossRefGoogle ScholarPubMed
Rottier, WC, van Werkhoven, CH, Bamberg, YRP, et al. Development of diagnostic prediction tools for bacteraemia caused by third-generation cephalosporin-resistant enterobacteria in suspected bacterial infections: a nested case-control study. Clin Microbiol Infect 2018;24: 1315–1321.CrossRefGoogle ScholarPubMed
Augustine, MR, Testerman, TL, Justo, JA, et al. Clinical risk score for prediction of extended-spectrum beta-lactamase–producing enterobacteriaceae in bloodstream isolates. Infect Control Hosp Epidemiol 2017;38:266–272.CrossRefGoogle ScholarPubMed
Kawamoto, K, Houlihan, CA, Balas, EA, Lobach, DF. Improving clinical practice using clinical decision support systems: a systematic review of trials to identify features critical to success. BMJ 2005;330:765.CrossRefGoogle Scholar
Peduzzi, P, Concato, J, Kemper, E, Holford, TR, Feinstein, AR. A simulation study of the number of events per variable in logistic regression analysis. J Clin Epidemiol 1996;49:1373–1379.CrossRefGoogle ScholarPubMed
Vittinghoff, E, McCulloch, CE. Relaxing the rule of ten events per variable in logistic and Cox regression. Am J Epidemiol 2007;165:710–718.CrossRefGoogle ScholarPubMed
Dietterich, T. Overfitting and undercomputing in machine learning. ACM Comput Surv 1995;27:326–327.CrossRefGoogle Scholar
Chen, X, Ishwaran, H. Pathway hunting by random survival forests. Bioinformatics 2013;29:99–105.10.1093/bioinformatics/bts643CrossRefGoogle ScholarPubMed
Strobl, C, Malley, J, Tutz, G. An introduction to recursive partitioning: rationale, application, and characteristics of classification and regression trees, bagging, and random forests. Psychol Method 2009;14:323–348.CrossRefGoogle ScholarPubMed
Drummond, C, Holte, RC. Exploiting the cost (in)sensitivity of decision tree splitting criteria. Proceedings of the Seventeenth International Conference on Machine Learning. Stanford, CA; 2000.Google Scholar
Tibshirani, R. Regression shrinkage and selection via the lasso. J Roy Stat Soc B 1996;58:267–288.Google Scholar
van der Laan, MJ, Polley, EC, Hubbard, AE. Super learner. Stat Appl Genet Mol Biol 2007;6:article25. Epub 2007 Sep 16.CrossRefGoogle ScholarPubMed
Song, YY, Lu, Y. Decision tree methods: applications for classification and prediction. Shanghai Arch Psychiatr 2015;27:130–135.Google ScholarPubMed
Akaike, H. A new look at the statistical model identification. IEEE Trans Automat Control 1974;19:716–723.CrossRefGoogle Scholar