Hostname: page-component-586b7cd67f-vdxz6 Total loading time: 0 Render date: 2024-11-29T04:55:42.911Z Has data issue: false hasContentIssue false
  • English
  • Français

Machine learning for the prediction of antimicrobial stewardship intervention in hospitalized patients receiving broad-spectrum agents

Published online by Cambridge University Press:  18 June 2020

Rachel J. Bystritsky Open the ORCID record for Rachel J. Bystritsky [Opens in a new window] ,
Alex Beltran ,
Albert T. Young ,
Andrew Wong ,
Xiao Hu  and
Sarah B. Doernberg
Rachel J. Bystritsky*
Affiliation:
Department of Medicine, Infectious Diseases, University of California–San Francisco, San Francisco, California
Alex Beltran
Affiliation:
Department of Bioengineering, University of California–San Francisco, San Francisco, California
Albert T. Young
Affiliation:
School of Medicine, University of California–San Francisco, San Francisco, California
Andrew Wong
Affiliation:
School of Medicine, University of California–San Francisco, San Francisco, California
Xiao Hu
Affiliation:
Department of Bioengineering, University of California–San Francisco, San Francisco, California
Sarah B. Doernberg
Affiliation:
Department of Medicine, Infectious Diseases, University of California–San Francisco, San Francisco, California
*
Author for correspondence: Rachel J. Bystritsky, E-mail: [email protected]
Get access Rights & Permissions [Opens in a new window]

Abstract

Objective:

A significant proportion of inpatient antimicrobial prescriptions are inappropriate. Post-prescription review with feedback has been shown to be an effective means of reducing inappropriate antimicrobial use. However, implementation is resource intensive. Our aim was to evaluate the performance of traditional statistical models and machine-learning models designed to predict which patients receiving broad-spectrum antibiotics require a stewardship intervention.

Methods:

We performed a single-center retrospective cohort study of inpatients who received an antimicrobial tracked by the antimicrobial stewardship program. Data were extracted from the electronic medical record and were used to develop logistic regression and boosted-tree models to predict whether antibiotic therapy required stewardship intervention on any given day as compared to the criterion standard of note left by the antimicrobial stewardship team in the patient’s chart. We measured the performance of these models using area under the receiver operating characteristic curves (AUROC), and we evaluated it using a hold-out validation cohort.

Results:

Both the logistic regression and boosted-tree models demonstrated fair discriminatory power with AUROCs of 0.73 (95% confidence interval [CI], 0.69–0.77) and 0.75 (95% CI, 0.72–0.79), respectively (P = .07). Both models demonstrated good calibration. The number of patients that would need to be reviewed to identify 1 patient who required stewardship intervention was high for both models (41.7–45.5 for models tuned to a sensitivity of 85%).

Conclusions:

Complex models can be developed to predict which patients require a stewardship intervention. However, further work is required to develop models with adequate discriminatory power to be applicable to real-world antimicrobial stewardship practice.

Type
Original Article
Information
Infection Control & Hospital Epidemiology , Volume 41 , Issue 9 , September 2020 , pp. 1022 - 1027
Check for updates
Copyright
© 2020 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

Hecker, MT, Aron, DC, Patel, NP, Lehmann, MK, Donskey, CJ. Unnecessary use of antimicrobials in hospitalized patients: current patterns of misuse with an emphasis on the antianaerobic spectrum of activity. Arch Intern Med 2003;163:972–978.CrossRefGoogle ScholarPubMed
Cosgrove, SE, Seo, SK, Bolon, MK, et al; CDC Prevention Epicenter Program. Evaluation of postprescription review and feedback as a method of promoting rational antimicrobial use: a multicenter intervention. Infect Control Hosp Epidemiol 2012;33:374–380.CrossRefGoogle ScholarPubMed
Barlam, TF, Cosgrove, SE, Abbo, LM, et al.Implementing an antibiotic stewardship program: guidelines by the Infectious Diseases Society of America and the Society for Healthcare Epidemiology of America. Clin Infect Dis 2016;62(10):e51–e77.CrossRefGoogle Scholar
Honda, H, Murakami, S, Tagashira, Y, et al.Efficacy of a postprescription review of broad-spectrum antimicrobial agents with feedback: a 4-year experience of antimicrobial stewardship at a tertiary care center. Open Forum Infect Dis 2018;5(12):ofy314.CrossRefGoogle Scholar
Tamma, PD, Avdic, E, Keenan, JF, et al.What is the more effective antibiotic stewardship intervention: preprescription authorization or postprescription review with feedback? Clin Infect Dis 2017;64:537–543.Google ScholarPubMed
Lesprit, P, et al.Postprescription review improves in-hospital antibiotic use: a multicenter randomized controlled trial. Clin Microbiol Infect 2015;21(2):180.e1–e7. Epub 2014 Oct 14.CrossRefGoogle ScholarPubMed
Davey, P, et al.Interventions to improve antibiotic prescribing practices for hospital inpatients. Cochrane Database Syst Rev 2017;2:CD003543.Google ScholarPubMed
Doernberg, SB, Abbo, LM, Burdette, SD, et al.Essential resources and strategies for antibiotic stewardship programs in the acute care setting. Clin Infect Dis 2018;67:1168–1174.CrossRefGoogle ScholarPubMed
Catho, G, De Kraker, M, Waldispul Suter, B, et al.Study protocol for a multicentre, cluster randomized, superiority trial evaluating the impact of computerized decision support, audit and feedback on antibiotic use: the COMPuterized Antibiotic Stewardship Study (COMPASS). BMJ Open 2018;8(6):e022666.CrossRefGoogle Scholar
Deo, RC. Machine Learning in Medicine. Circulation 2015;132:1920–1930.CrossRefGoogle ScholarPubMed
Huang, H, Ding, N, Yang, T, et al.Cross-sectional whole-genome sequencing and epidemiological study of multidrug-resistant Mycobacterium tuberculosis in China. Clin Infect Dis 2018; ciy883.Google Scholar
Escobar, GJ, Baker, JM, Kipnis, P, et al.Prediction of recurrent Clostridium difficile infection using comprehensive electronic medical records in an integrated healthcare delivery system. Infect Control Hosp Epidemiol 2017;38:1196–1203.CrossRefGoogle Scholar
Brier, GW. Verification of forecasts expressed in terms of probability. Mon Weather Rev 1950;78:1e3.2.0.CO;2>CrossRefGoogle Scholar
Cohen, ME, Ko, CY, Bilimoria, KY, et al.Optimizing ACS NSQIP modeling for evaluation of surgical quality and risk: patient risk adjustment, procedure mix adjustment, shrinkage adjustment, and surgical focus. J Am Coll Surg 2013;217:336–346.CrossRefGoogle ScholarPubMed
Deitterich, TG. Approximate statistical tests for comparing supervised classification learning algorithms. Neural Comput 1998;10:1895–1923.CrossRefGoogle Scholar
Watson, DS, Krutzinna, J, Griffiths, CE, McInnes, IB, Barnes, MR, Floridi, L. Clinical applications of machine learning algorithms: beyond the black box. BMJ 2019;364:I886.CrossRefGoogle ScholarPubMed
Vu, MT, Adalı, T, Ba, D, et al.A shared vision for machine learning in neuroscience. J Neurosci 2018;38:1601–1607.CrossRefGoogle ScholarPubMed
Deng, J, Yuan, Q, Mamitsuka, H, Zhu, S. Drug E-Rank: Predicting Drug-Target Interactions by Learning to Rank. Methods Mol Biol 2018;1807:195–202.CrossRefGoogle Scholar
Supplementary material: File

Bystritsky et al. supplementary material

Bystritsky et al. supplementary material

Download Bystritsky et al. supplementary material(File)
File 19.1 KB