Institutional members access full text with Ovid®

Determinants of Mortality After Hospital Discharge in ICU Patients: Literature Review and Dutch Cohort Study*

Brinkman, Sylvia MSc1; Bakhshi-Raiez, Ferishta PhD1; Abu-Hanna, Ameen PhD1; Jonge, Evert de MD, PhD2; de Keizer, Nicolette F. PhD1

doi: 10.1097/CCM.0b013e31827ca4f9
Clinical Investigations

Objectives: First, to conduct a literature review on the long-term mortality of ICU patients and its determinants. Second, to assess the influence of the found determinants at 3, 6, and 12 months mortality after hospital discharge in the Dutch ICU population.

Design: Combination of a literature review to evaluate determinants of long-term mortality and a Dutch cohort study in which the found determinants are applied.

Setting: PubMed and EMBASE were searched on English written articles published between 1966 and 2011. The cohort study was conducted in ICU patients from 81 Dutch mixed ICUs.

Data: A total of 24 articles with a main focus on describing or predicting the case-mix adjusted long-term mortality of the general ICU population were identified. The cohort study consisted of 48,107 ICU patients who were discharged alive from the hospital between January 1, 2007, and October 1, 2010.

Interventions: None.

Measurements and Main Results: The included articles are summarized on patient and study characteristics, methods, results, and determinants used for case-mix adjustment. Additionally, the quality of the included articles was assessed using a checklist for studies on long-term survival. The median mortality rate of the general ICU population 1 year after ICU admission was 24% (range 16% to 44%). The determinants used for case-mix adjustment differed widely between the studies. In the cohort study, we found that age, reason for ICU admission, and comorbidities were associated with all long-term mortality endpoints. However, the magnitude and direction of the influence by these determinants differed for the different endpoints (i.e., 3, 6, and 12 mo after hospital discharge).

Conclusions: The long-term mortality found in the included articles was difficult to compare due to low quality, variation in case-mix, study design, and differences in case-mix adjustment. The most commonly used determinants in the literature were comparable to the most important determinants found in the Dutch cohort study.

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

2Department of Intensive Care, Leiden University Medical Center, Leiden, The Netherlands.

*See also p. 1370.

Supported by the Dutch National Intensive Care Evaluation (NICE) registry.

The authors have not disclosed any potential conflicts of interest.

For information regarding this article, E-mail:

© 2013 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins