Introduction

Intensive care is generally regarded as expensive, and as a result beds are limited. This has raised serious questions about rationing when there are insufficient beds for all those referred. However, the evidence for the cost effectiveness of intensive care is weak and the work that does exist usually assumes that those who are not admitted do not survive, which is not always the case. Randomised studies of the effectiveness of intensive care are difficult to justify on ethical grounds; therefore, this observational study examined the cost effectiveness of ICU admission by comparing patients who were accepted into ICU after ICU triage to those who were not accepted, while attempting to adjust such comparison for confounding factors.

Methods

This multi-centre observational cohort study involved 11 hospitals in 7 EU countries and was designed to assess the cost effectiveness of admission to intensive care after ICU triage. A total of 7,659 consecutive patients referred to the intensive care unit (ICU) were divided into those accepted for admission and those not accepted. The two groups were compared in terms of cost and mortality using multilevel regression models to account for differences across centres, and after adjusting for age, Karnofsky score and indication for ICU admission. The analyses were also stratified by categories of Simplified Acute Physiology Score (SAPS) II predicted mortality (< 5%, 5% to 40% and >40%). Cost effectiveness was evaluated as cost per life saved and cost per life-year saved.

Results

Admission to ICU produced a relative reduction in mortality risk, expressed as odds ratio, of 0.70 (0.52 to 0.94) at 28 days. When stratified by predicted mortality, the odds ratio was 1.49 (0.79 to 2.81), 0.7 (0.51 to 0.97) and 0.55 (0.37 to 0.83) for <5%, 5% to 40% and >40% predicted mortality, respectively. Average cost per life saved for all patients was $103,771 (€82,358) and cost per life-year saved was $7,065 (€5,607). These figures decreased substantially for patients with predicted mortality higher than 40%, $60,046 (€47,656) and $4,088 (€3,244), respectively. Results were very similar when considering three-month mortality. Sensitivity analyses performed to assess the robustness of the results provided findings similar to the main analyses.

Conclusions

Not only does ICU appear to produce an improvement in survival, but the cost per life saved falls for patients with greater severity of illness. This suggests that intensive care is similarly cost effective to other therapies that are generally regarded as essential.

Background

Two acute respiratory distress syndrome (ARDS) subphenotypes (hyperinflammatory and hypoinflammatory) with distinct clinical and biological features and differential treatment responses have been identified using latent class analysis (LCA) in seven individual cohorts. To facilitate bedside identification of subphenotypes, clinical classifier models using readily available clinical variables have been described in four randomised controlled trials. We aimed to assess the performance of these models in observational cohorts of ARDS.

Methods

In this observational, multicohort, retrospective study, we validated two machine-learning clinical classifier models for assigning ARDS subphenotypes in two observational cohorts of patients with ARDS: Early Assessment of Renal and Lung Injury (EARLI; n=335) and Validating Acute Lung Injury Markers for Diagnosis (VALID; n=452), with LCA-derived subphenotypes as the gold standard. The primary model comprised only vital signs and laboratory variables, and the secondary model comprised all predictors in the primary model, with the addition of ventilatory variables and demographics. Model performance was assessed by calculating the area under the receiver operating characteristic curve (AUC) and calibration plots, and assigning subphenotypes using a probability cutoff value of 0·5 to determine sensitivity, specificity, and accuracy of the assignments. We also assessed the performance of the primary model in EARLI using data automatically extracted from an electronic health record (EHR; EHR-derived EARLI cohort). In Large Observational Study to Understand the Global Impact of Severe Acute Respiratory Failure (LUNG SAFE; n=2813), a multinational, observational ARDS cohort, we applied a custom classifier model (with fewer variables than the primary model) to determine the prognostic value of the subphenotypes and tested their interaction with the positive end-expiratory pressure (PEEP) strategy, with 90-day mortality as the dependent variable.

Findings

The primary clinical classifier model had an area under receiver operating characteristic curve (AUC) of 0·92 (95% CI 0·90-0·95) in EARLI and 0·88 (0·84-0·91) in VALID. Performance of the primary model was similar when using exclusively EHR-derived predictors compared with manually curated predictors (AUC=0·88 [95% CI 0·81-0·94] vs 0·92 [0·88-0·97]). In LUNG SAFE, 90-day mortality was higher in patients assigned the hyperinflammatory subphenotype than in those with the hypoinflammatory phenotype (414 [57%] of 725 vs 694 [33%] of 2088; p<0·0001). There was a significant treatment interaction with PEEP strategy and ARDS subphenotype (p=0·041), with lower 90-day mortality in the high PEEP group of patients with the hyperinflammatory subphenotype (hyperinflammatory subphenotype: 169 [54%] of 313 patients in the high PEEP group vs 127 [62%] of 205 patients in the low PEEP group; hypoinflammatory subphenotype: 231 [34%] of 675 patients in the high PEEP group vs 233 [32%] of 734 patients in the low PEEP group).

Interpretation

Classifier models using clinical variables alone can accurately assign ARDS subphenotypes in observational cohorts. Application of these models can provide valuable prognostic information and could inform management strategies for personalised treatment, including application of PEEP, once prospectively validated.

Funding

US National Institutes of Health and European Society of Intensive Care Medicine.