Elsevier

Journal of Critical Care

Volume 43, February 2018, Pages 148-155
Journal of Critical Care

Clinical Potpourri
The ability of intensive care unit physicians to estimate long-term prognosis in survivors of critical illness

https://doi.org/10.1016/j.jcrc.2017.09.007Get rights and content

Highlights

  • ICU physician prognosis does not match observed outcomes in 1/3 ICU patients

  • Inaccurate prognoses predominantly due to overoptimistic expectation of outcome

  • Overoptimism associated with comorbidities, admission type and events at admission

Abstract

Purpose

To assess the reliability of physicians' prognoses for intensive care unit (ICU) survivors with respect to long-term survival and health related quality of life (HRQoL).

Methods

We performed an observational cohort-study in a single mixed tertiary ICU in The Netherlands. ICU survivors with a length of stay > 48 h were included. At ICU discharge, one-year prognosis was estimated by physicians using the four-option Sabadell score to record their expectations. The outcome of interest was poor outcome, which was defined as dying within one-year follow-up, or surviving with an EuroQoL5D-3 L index < 0.4.

Results

Among 1399 ICU survivors, 1068 (76%) subjects were expected to have a good outcome; 243 (18%) a poor long-term prognosis; 43 (3%) a poor short-term prognosis, and 45 (3%) to die in hospital (i.e. Sabadell score levels). Poor outcome was observed in 38%, 55%, 86%, and 100% of these groups respectively (concomitant c-index: 0.61). The expected prognosis did not match observed outcome in 365 (36%) patients. This was almost exclusively (99%) due to overoptimism. Physician experience did not affect results.

Conclusions

Prognoses estimated by physicians incorrectly predicted long-term survival and HRQoL in one-third of ICU survivors. Moreover, inaccurate prognoses were generally the result of overoptimistic expectations of outcome.

Introduction

The ICU physician is increasingly involved in decision making concerning follow up and post-ICU treatment of patients who have survived ICU care [1]. To do so, it is important to identify patients with an increased risk of poor outcome at the time of ICU discharge [2], [3]. Currently prognoses at ICU discharge are largely based on the intuitive insight of the treating physicians. Based on their clinical expertise, they incorporate a patient's condition before ICU admission (medical history, functional status, quality of life and social environment) and the events during hospital and ICU stay into a holistic prognosis for the patient. This contrasts with the ‘objective’ multivariable prediction models typically used in ICU research, which incorporate a patient's vital status, age and pre-existing comorbidities at ICU admission. However, these models typically do not incorporate prior functional status or quality of life [4], [5], and are mostly focused on prediction of short-term mortality rather than long-term functional outcomes.

Because of these omissions, researchers have tried to validate the ICU physicians' estimations of the risk of poor ICU outcomes [6], [7]. And they directly compared the prognostic performance of physicians' prognoses to those of statistical models [8], [9], [10]. A systematic review of such studies showed that at ICU admission, physicians were more accurate in discriminating patients who would die in comparison to contemporary statistical models [10]. However, thus far, only in the domain of the neurologically critically ill were the studies focused on predicting functional status as outcome of interest [11]. Studies in the general ICU population focused on survival alone [8], [12]. As a result, it is unknown whether ICU physicians are also accurate at predicting survival in conjunction with quality of life, at the moment a general ICU patient is discharged from the ICU.

So, with the increasing attention for the long-term functional consequences of ICU care in all critically ill patients [13], [14], prognostic estimations made at ICU discharge should accurately reflect both the probability of long-term survival and that of an adequate health related quality of life (HRQoL) [12], [13]. Therefore, the aim of this study was to assess the ability of physicians to accurately prognosticate survivors of critical illness upon ICU discharge with respect to their long-term survival and HRQoL.

Section snippets

Study design, setting and participants

This study was designed as a cohort study using data prospectively collected for the purpose of benchmarking and follow-up data for quality of care evaluation. The study was performed at the ICU of the University Medical Center Utrecht. This ICU's population is a mix of medical-surgical critically ill patients, including those after major cardiac, neurological (trauma, vascular and oncology related), gastro-intestinal and transplant surgery, and most types of medical patients. We included ICU

Study population

Among 1676 unique patients enrolled during the inclusion period having a length of stay over 48 h, 1419 (84.7%) survived their ICU stay. Of these eligible patients, three were excluded because of a missing Sabadell score. Seventeen patients were considered lost to follow-up because they were not registered in the Dutch municipal registry, or because there was no available address. This left 1399 patients to be included in this study (fig. 1).

In the total study population, the median ICU length

Discussion

We investigated to what extent the physician's estimation of a patient's prognosis at ICU discharge was in accordance with observed long-term outcomes, and found that ICU physicians performed only moderately. Moreover, when studying the predictive performance for survival and HRQoL combined, the discriminatory performance was poor. One third of all ICU survivors experienced an outcome which was particularly worse than what was expected by the ICU physician.

Interestingly, in patients whom the

Conclusions

The subjective prognosis estimated by ICU physicians incorrectly predicted long-term survival and HRQoL in one out of three ICU patients, regardless of physician experience. This suggests that ICU physicians are currently unable to perform sufficiently reliable risk stratifications in survivors of critical illness with respect to long-term patient-centered outcomes.

The following are the supplementary data related to this article.

Ethical Approval and Consent to participate

The institutional review board (IRB) of the University Medical Center Utrecht approved the study protocol and waived the need for informed consent when working with anonymised patient and follow-up data (UMC Utrecht IRB protocol number 10/006).

Consent for publication

Not applicable.

Availability of supporting data

A minimal version of this study's research datasets are available from the corresponding author on reasonable request, which also takes into account Dutch Law and good scientific practice for sharing anonymised biomedical patient data.

Competing interests statement

This study was supported by the NutsOhra Foundation, project nr 1404–013, entitled “Prognostics and decision making in prolonged intensive care treatment”. Additionally, all authors were appointed as researchers and/or medical doctors by the University Medical Center Utrecht, the academic hospital where this study was performed. Neither the NutsOhra foundation, nor the hospital had influence on any part of the conducting or writing of this study. On behalf of all authors, the corresponding

Funding

This study was supported by the NutsOhra Foundation, project nr 1404-013, entitled “Prognostics and decision making in prolonged intensive care treatment”.

Authors' contributions

IWS contributed to data collection, carried out the data analysis and drafted the manuscript. IWS and OC conceived the study. OC contributed to data collection. DL participated in the study's design and coordination and helped to draft the manuscript. AS contributed to data collection. JD participated in the drafting of the manuscript. DD participated in the drafting of the manuscript and contributed to data collection. LP participated in the data analysis, advised on the methodological design

Acknowledgements

Not applicable.

References (32)

  • R. Barrera et al.

    Accuracy of predictions of survival at admission to the intensive care unit

    J Crit Care

    (2001)
  • A.R. Donders et al.

    Review: a gentle introduction to imputation of missing values

    J Clin Epidemiol

    (2006)
  • A. Schandl et al.

    Early prediction of new-onset physical disability after intensive care unit stay: a preliminary instrument

    Crit Care

    (2014)
  • M. van der Schaaf et al.

    Functional status after intensive care: a challenge for rehabilitation professionals to improve outcome

    J Rehabil Med

    (2009)
  • D.W. Dowdy et al.

    Quality of life in adult survivors of critical illness: a systematic review of the literature

    Intensive Care Med

    (2005)
  • J.E. Zimmerman et al.

    Acute physiology and chronic health evaluation (APACHE) IV: hospital mortality assessment for today's critically ill patients

    Crit Care Med

    (2006)
  • R.P. Moreno et al.

    SAPS 3—from evaluation of the patient to evaluation of the intensive care unit. Part 2: development of a prognostic model for hospital mortality at ICU admission

    Intensive Care Med

    (2005)
  • W. Meadow et al.

    Foreseeing the future: predicting 6 month outcomes from events in the medical intensive care unit (MICU)

    Am J Resp Crit Care

    (2011)
  • E. Litton et al.

    Comparison of physician prediction with 2 prognostic scoring systems in predicting 2-year mortality after intensive care admission: a linked-data cohort study

    J Crit Care

    (2012)
  • N. Scholz et al.

    Outcome prediction in critical care: physicians' prognoses vs. scoring systems

    Eur J Anaesthesiol

    (2004)
  • T. Sinuff et al.

    Mortality predictions in the intensive care unit: comparing physicians with scoring systems

    Crit Care Med

    (2006)
  • A. Finley Caulfield et al.

    Outcome prediction in mechanically ventilated neurologic patients by junior neurointensivists

    Neurology

    (2010)
  • B. Afessa et al.

    Predicting mortality in intensive care unit survivors using a subjective scoring system

    Crit Care

    (2007)
  • S.G. Oeyen et al.

    Quality of life after intensive care: a systematic review of the literature

    Crit Care Med

    (2010)
  • M.C. Kerckhoffs et al.

    Long-term outcomes of ICU treatment

    Ned Tijdschr Geneeskd

    (2016)
  • J.L. Vincent et al.

    The SOFA (sepsis-related organ failure assessment) score to describe organ dysfunction/failure. On behalf of the working group on sepsis-related problems of the European Society of Intensive Care Medicine

    Intensive Care Med

    (1996)
  • Cited by (23)

    • Using long-term predicted Quality of Life in ICU clinical practice to prepare patients for life post-ICU: A feasibility study

      2022, Journal of Critical Care
      Citation Excerpt :

      Seven patients died during their ICU admission after being identified as expected survivors and a measurement had taken place. This is in accordance with prior literature, which reported that an overoptimistic prognosis did not match the outcome in a third of patients [6]. Clinicians in this study all acknowledged that health deterioration was a part of ICU care, and did not think they should not have included these patients.

    • On predictions in critical care: The individual prognostication fallacy in elderly patients

      2021, Journal of Critical Care
      Citation Excerpt :

      Approximately 1 in 6 patients who were unanimously predicted to die actually survived [15]. Similar problems with the quality of heuristic outcome predictions have been reported in other settings [16,17]. To overcome the flaws of personal judgement and improve the accuracy of prognostication, more structured approaches to the collection and analysis of data have been developed in the past [18].

    • Are providers overconfident in predicting outcome after cardiac arrest?

      2020, Resuscitation
      Citation Excerpt :

      Our results parallel those in general critical care patients as well. One study asked providers to predict long-term prognosis of general ICU patients by asking them to choose from four options (good prognosis to death) and showed that providers were wrong a third of the time.40 Similar to our work, most errors were overly optimistic predictions and providers' experience did not affect results.40

    • Reprint of: Intensive care for old patients

      2020, Journal Europeen des Urgences et de Reanimation
    • Intensive care for old patients

      2020, Bulletin de l'Academie Nationale de Medecine
    View all citing articles on Scopus
    View full text