Secondary Logo

Journal Logo

Institutional members access full text with Ovid®

Which Models Can I Use to Predict Adult ICU Length of Stay? A Systematic Review*

Verburg, Ilona Willempje Maria MSc1; Atashi, Alireza MSc2,3; Eslami, Saeid PhD1,4,5; Holman, Rebecca PhD1,6; Abu-Hanna, Ameen PhD1; de Jonge, Everet MD7; Peek, Niels PhD8; de Keizer, Nicolette Fransisca PhD1

doi: 10.1097/CCM.0000000000002054
Online Review Article
Buy
SDC

Objective: We systematically reviewed models to predict adult ICU length of stay.

Data Sources: We searched the Ovid EMBASE and MEDLINE databases for studies on the development or validation of ICU length of stay prediction models.

Study Selection: We identified 11 studies describing the development of 31 prediction models and three describing external validation of one of these models.

Data Extraction: Clinicians use ICU length of stay predictions for planning ICU capacity, identifying unexpectedly long ICU length of stay, and benchmarking ICUs. We required the model variables to have been published and for the models to be free of organizational characteristics and to produce accurate predictions, as assessed by R2 across patients for planning and identifying unexpectedly long ICU length of stay and across ICUs for benchmarking, with low calibration bias. We assessed the reporting quality using the Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies.

Data Synthesis: The number of admissions ranged from 253 to 178,503. Median ICU length of stay was between 2 and 6.9 days. Two studies had not published model variables and three included organizational characteristics. None of the models produced predictions with low bias. The R2 was 0.05–0.28 across patients and 0.01–0.64 across ICUs. The reporting scores ranged from 49 of 78 to 60 of 78 and the methodologic scores from 12 of 22 to 16 of 22.

Conclusion: No models completely satisfy our requirements for planning, identifying unexpectedly long ICU length of stay, or for benchmarking purposes. Physicians using these models to predict ICU length of stay should interpret them with reservation.

1Department of Medical Informatics, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands.

2Student Research Committee, Mashhad University of Medical Sciences, Mashhad, Iran.

3Cancer Informatics Department, Breast Cancer Research Center, ACECR, Tehran, Iran.

4Pharmaceutical Research Center, School of Pharmacy, Mashhad University of Medical Sciences, Mashhad, Iran.

5Department of Medical Informatics, Faculty of Medicine, Mashhad University of Medical Sciences, Mashhad, Iran.

6Clinical Research Unit, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands.

7Department of Intensive Care, Leiden University Medical Center, Leiden, The Netherlands.

8Health e-Research Centre, Division of Imaging, Informatics and Data Sciences, School of Health Sciences, The University of Manchester, Manchester, United Kingdom.

*See also p. 379.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (http://journals.lww.com/ccmjournal).

Dr. Verburg received grant support (The National Intensive Care Evaluation [NICE] foundation pays the department of Medical Informatics for maintaining the national database, providing feedback reports, and doing analyses. She is an employee of the department of Medical Informatics). Her institution received support for participation in review activities. Dr. Holman’s institution received grant support (The NICE foundation pays the department of Medical Informatics for maintaining the national database, providing feedback reports, and doing analyses. She is an employee of the department of Medical Informatics) and received support for participation in review activities. She disclosed employment (She is also employed by the Clinical Research Unit of the Academic Medical Center, Amsterdam, The Netherlands). Dr. de Keizer’s institution received grant support from the NICE foundation (The department of medical informatics receive funding for the maintenance of the Dutch NICE registry. This literature review is related to a project on benchmarking length of stay), served as a board member for the NICE foundation (She is a member of the NICE board), and is employed by the NICE foundation. The remaining authors have disclosed that they do not have any potential conflicts of interest.

For information regarding this article, E-mail: i.w.verburg@amc.uva.nl

Copyright © 2017 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.