One-year resource utilisation, costs and quality of life in patients with acute respiratory distress syndrome (ARDS): secondary analysis of a randomised controlled trial

Contextual design requirements for decision-support tools involved in weaning patients from mechanical ventilation in intensive care units

Weaning patients from ventilation in intensive care units (ICU) is a complex task. There is a growing desire to build decision-support tools to help clinicians during this process, especially those employing Artificial Intelligence (AI). However, tools built for this purpose should fit within and ideally improve the current work environment, to ensure they can successfully integrate into clinical practice. To do so, it is important to identify areas where decision-support tools may aid clinicians, and associated design requirements for such tools. This study analysed the work context surrounding the weaning process from mechanical ventilation in ICU environments, via cognitive task and work domain analyses. In doing so, both what cognitive processes clinicians perform during weaning, and the constraints and affordances of the work environment itself, were described. This study found a number of weaning process tasks where decision-support tools may prove beneficial, and from these a set of contextual design requirements were created. This work benefits researchers interested in creating human-centred decision-support tools for mechanical ventilation that are sensitive to the wider work system.

A cost-effectiveness analysis of reduced viral transmission with baloxavir marboxil versus oseltamivir or no treatment for seasonal and pandemic influenza management in the United Kingdom

BACKGROUND

Baloxavir marboxil is an oral, single-dose, cap-dependent endonuclease inhibitor that reduces the duration of influenza symptoms and rapidly stops viral shedding. We developed a susceptible, exposed, infected, recovered (SEIR) model to inform a cost-effectiveness model (CEM) of baloxavir versus oseltamivir or no antiviral treatment in the UK.

RESEARCH DESIGN AND METHODS

The SEIR model estimated the attack rates among otherwise healthy and high-risk individuals in seasonal and pandemic settings. The CEM assumed that a proportion of infected patients would receive antiviral treatment. Results were reported at the population level (per 10,000 at risk of infection).

RESULTS

The SEIR model estimated greater reductions in infections with baloxavir. In a seasonal setting, baloxavir provided incremental cost-effectiveness ratios (ICERs) of £1884 per quality-adjusted life-year (QALY) gained versus oseltamivir and a dominant cost-effectiveness position versus no antiviral treatment in the total population; ICERs of £2574/QALY versus oseltamivir and £128/QALY versus no antiviral treatment were seen in the high-risk population. Baloxavir was also cost-effective versus oseltamivir or no antiviral treatment and reduced population-level health system occupancy concerns during a pandemic.

CONCLUSION

Baloxavir treatment resulted in the fewest influenza cases and was cost-effective versus oseltamivir or no antiviral treatment from a UK National Health Service perspective.

Improved pediatric ICU mortality prediction for respiratory diseases: machine learning and data subdivision insights

The growing concern of pediatric mortality demands heightened preparedness in clinical settings, especially within intensive care units (ICUs). As respiratory-related admissions account for a substantial portion of pediatric illnesses, there is a pressing need to predict ICU mortality in these cases. This study based on data from 1188 patients, addresses this imperative using machine learning techniques and investigating different class balancing methods for pediatric ICU mortality prediction. This study employs the publicly accessible "Paediatric Intensive Care database" to train, validate, and test a machine learning model for predicting pediatric patient mortality. Features were ranked using three machine learning feature selection techniques, namely Random Forest, Extra Trees, and XGBoost, resulting in the selection of 16 critical features from a total of 105 features. Ten machine learning models and ensemble techniques are used to make accurate mortality predictions. To tackle the inherent class imbalance in the dataset, we applied a unique data partitioning technique to enhance the model's alignment with the data distribution. The CatBoost machine learning model achieved an area under the curve (AUC) of 72.22%, while the stacking ensemble model yielded an AUC of 60.59% for mortality prediction. The proposed subdivision technique, on the other hand, provides a significant improvement in performance metrics, with an AUC of 85.2% and an accuracy of 89.32%. These findings emphasize the potential of machine learning in enhancing pediatric mortality prediction and inform strategies for improved ICU readiness.

Treatment of COVID-19-associated ARDS with umbilical cord-derived mesenchymal stromal cells in the STROMA-CoV-2 multicenter randomized double-blind trial: long-term safety, respiratory function, and quality of life

BACKGROUND

The STROMA-CoV-2 study was a French phase 2b, multicenter, double-blind, randomized, placebo-controlled clinical trial that did not identify a significant efficacy of umbilical cord-derived mesenchymal stromal cells in patients with SARS-CoV-2-induced acute respiratory distress syndrome. Safety on day 28 was found to be good. The aim of our extended study was to assess the 6- and 12-month safety of UC-MSCs administration in the STROMA-CoV-2 cohort.

METHODS

A detailed multi-domain assessment was conducted at 6 and 12 months following hospital discharge focusing on adverse events, lung computed tomography-scan, pulmonary and muscular functional status, and quality of life in the STROMA-CoV-2 cohort including SARS-CoV-2-related early (< 96 h) mild-to-severe acute respiratory distress syndrome.

RESULTS

Between April 2020 and October 2020, 47 patients were enrolled, of whom 19 completed a 1-year follow-up. There were no significant differences in any endpoints or adverse effects between the UC-MSCs and placebo groups at the 6- and 12-month assessments. Ground-glass opacities persisted at 1 year in 5 patients (26.3%). Furthermore, diffusing capacity for carbon monoxide remained altered over 1 year, although no patient required oxygen or non-invasive ventilatory support. Quality of life revealed declines in mental, emotional and physical health throughout the follow-up period, and the six-minute walking distance remained slightly impaired at the 1-year patient assessment.

CONCLUSIONS

This study suggests a favorable safety profile for the use of intravenous UC-MSCs in the context of the first French wave of SARS-CoV-2-related moderate-to-severe acute respiratory distress syndrome, with no adverse effects observed at 1 year.

COVID-19 does not influence functional status after ARDS therapy

RATIONALE

Health-related quality of life after surviving acute respiratory distress syndrome has come into focus in recent years, especially during the coronavirus disease 2019 pandemic.

OBJECTIVES

A total of 144 patients with acute respiratory distress syndrome caused by COVID-19 or of other origin were recruited in a randomized multicenter trial.

METHODS

Clinical data during intensive care treatment and data up to 180 days after study inclusion were collected. Changes in the Sequential Organ Failure Assessment score were used to quantify disease severity. Disability was assessed using the Barthel index on days 1, 28, 90, and 180.

MEASUREMENTS

Mortality rate and morbidity after 180 days were compared between patients with and without COVID-19. Independent risk factors associated with high disability were identified using a binary logistic regression.

MAIN RESULTS

The SOFA score at day 5 was an independent risk factor for high disability in both groups, and score dynamic within the first 5 days significantly impacted disability in the non-COVID group. Mortality after 180 days and impairment measured by the Barthel index did not differ between patients with and without COVID-19.

CONCLUSIONS

Resolution of organ dysfunction within the first 5 days significantly impacts long-term morbidity. Acute respiratory distress syndrome caused by COVID-19 was not associated with increased mortality or morbidity.

Quality of Life in COVID-Related ARDS Patients One Year after Intensive Care Discharge (Odissea Study): A Multicenter Observational Study

BACKGROUND

Investigating the health-related quality of life (HRQoL) after intensive care unit (ICU) discharge is necessary to identify possible modifiable risk factors. The primary aim of this study was to investigate the HRQoL in COVID-19 critically ill patients one year after ICU discharge.

METHODS

In this multicenter prospective observational study, COVID-19 patients admitted to nine ICUs from 1 March 2020 to 28 February 2021 in Italy were enrolled. One year after ICU discharge, patients were required to fill in short-form health survey 36 (SF-36) and impact of event-revised (IES-R) questionnaire. A multivariate linear or logistic regression analysis to search for factors associated with a lower HRQoL and post-traumatic stress disorded (PTSD) were carried out, respectively.

RESULTS

Among 1003 patients screened, 343 (median age 63 years [57-70]) were enrolled. Mechanical ventilation lasted for a median of 10 days [2-20]. Physical functioning (PF 85 [60-95]), physical role (PR 75 [0-100]), emotional role (RE 100 [33-100]), bodily pain (BP 77.5 [45-100]), social functioning (SF 75 [50-100]), general health (GH 55 [35-72]), vitality (VT 55 [40-70]), mental health (MH 68 [52-84]) and health change (HC 50 [25-75]) describe the SF-36 items. A median physical component summary (PCS) and mental component summary (MCS) scores were 45.9 (36.5-53.5) and 51.7 (48.8-54.3), respectively, considering 50 as the normal value of the healthy general population. In all, 109 patients (31.8%) tested positive for post-traumatic stress disorder, also reporting a significantly worse HRQoL in all SF-36 domains. The female gender, history of cardiovascular disease, liver disease and length of hospital stay negatively affected the HRQoL. Weight at follow-up was a risk factor for PTSD (OR 1.02, p = 0.03).

CONCLUSIONS

The HRQoL in COVID-19 ARDS (C-ARDS) patients was reduced regarding the PCS, while the median MCS value was slightly above normal. Some risk factors for a lower HRQoL have been identified, the presence of PTSD is one of them. Further research is warranted to better identify the possible factors affecting the HRQoL in C-ARDS.

COVID-19 and the forgone health benefits of elective operations

BACKGROUND AND AIM

The first SARS-CoV-2 pandemic wave in Germany involved a tradeoff between saving the lives of COVID-19 patients by providing sufficient intensive care unit (ICU) capacity and foregoing the health benefits of elective procedures. This study aims to quantify this tradeoff.

METHODS

The analysis is conducted at both the individual and population levels. The analysis calculates quality-adjusted life years (QALYs) to facilitate a comparison between the health gains from saving the lives of COVID-19 patients in the ICU and the health losses associated with postponing operative procedures. The QALYs gained from saving the lives of COVID-19 patients are calculated based on both the real-world ICU admissions and deaths averted from flattening the first wave. Scenario analysis was used to account for variation in input factors.

RESULTS

At the individual level, the resource-adjusted QALY gain of saving one COVID-19 life is predicted to be 3 to 15 times larger than the QALY loss of deferring one operation (the average multiplier is 9). The real-world QALY gain at the population level is estimated to fall within the range of the QALY loss due to delayed procedures. The modeled QALY gain by flattening the first wave is 3 to 31 times larger than the QALY loss due to delayed procedures (the average multiplier is 17).

CONCLUSION

During the first wave of the pandemic, the resource-adjusted health gain from treating one COVID-19 patient in the ICU was found to be much larger than the health loss from deferring one operation. At the population level, flattening the first wave led to a much larger health gain than the health loss from delaying operative procedures.

Functional, imaging, and respiratory evaluation (FIRE) of patients post-hospitalization for COVID-19: protocol for a pilot observational study

Introduction

We describe a protocol for FIRE CORAL, an observational cohort study that examines the recovery from COVID-19 disease following acute hospitalization with an emphasis on functional, imaging, and respiratory evaluation.

Methods and analysis

FIRE CORAL is a multicenter prospective cohort study of participants recovering from COVID-19 disease with in-person follow-up for functional and pulmonary phenotyping conducted by the National Heart, Lung and Blood Institute (NHLBI) Prevention and Early Treatment of Acute Lung Injury (PETAL) Network. FIRE CORAL will include a subset of participants enrolled in Biology and Longitudinal Epidemiology of PETAL COVID-19 Observational Study (BLUE CORAL), an NHLBI-funded prospective cohort study describing the clinical characteristics, treatments, biology, and outcomes of hospitalized patients with COVID-19 across the PETAL Network. FIRE CORAL consists of a battery of in-person assessments objectively measuring pulmonary function, abnormalities on lung imaging, physical functional status, and biospecimen analyses. Participants will attend and perform initial in-person testing at 3 to 9 months after hospitalization. The primary objective of the study is to determine the feasibility of longitudinal assessments investigating multiple domains of recovery from COVID-19. Secondarily, we will perform descriptive statistics, including the prevalence and characterization of abnormalities on pulmonary function, chest imaging, and functional status. We will also identify potential clinical and biologic factors that predict recovery or the occurrence of persistent impairment of pulmonary function, chest imaging, and functional status.

Ethics and dissemination

FIRE CORAL is approved via the Vanderbilt University central institutional review board (IRB) and via reliance agreement with the site IRBs. Results will be disseminated via the writing group for the protocol committee and reviewed by the PETAL Network publications committee prior to publication. Data obtained via the study will subsequently be made publicly available via NHLBI’s biorepository.

Strengths and limitations of the study

Strengths:

First US-based multicenter cohort of pulmonary and functional outcomes in patients previously hospitalized for COVID-19 infection

Longitudinal biospecimen measurement allowing for biologic phenotyping of abnormalities

Geographically diverse cohort allowing for a more generalizable understanding of post-COVID pulmonary sequela

Limitations:

Selected cohort given proximity to a participating center

Small cohort which may be underpowered to identify small changes in pulmonary function

BMI1 Silencing Induces Mitochondrial Dysfunction in Lung Epithelial Cells Exposed to Hyperoxia

Acute Lung Injury (ALI), characterized by bilateral pulmonary infiltrates that restrict gas exchange, leads to respiratory failure. It is caused by an innate immune response with white blood cell infiltration of the lungs, release of cytokines, an increase in reactive oxygen species (ROS), oxidative stress, and changes in mitochondrial function. Mitochondrial alterations, changes in respiration, ATP production and the unbalancing fusion and fission processes are key events in ALI pathogenesis and increase mitophagy. Research indicates that BMI1 (B cell-specific Moloney murine leukemia virus integration site 1), a protein of the Polycomb repressive complex 1, is a cell cycle and survival regulator that plays a role in mitochondrial function. BMI1-silenced cultured lung epithelial cells were exposed to hyperoxia to determine the role of BMI1 in mitochondrial metabolism. Its expression significantly decreases in human lung epithelial cells (H441) following hyperoxic insult, as determined by western blot, Qrt-PCR, and functional analysis. This decrease correlates with an increase in mitophagy proteins, PINK1, Parkin, and DJ1; an increase in the expression of tumor suppressor PTEN; changes in the expression of mitochondrial biomarkers; and decreases in the oxygen consumption rate (OCR) and tricarboxylic acid enzyme activity. Our bioinformatics analysis suggested that the BMI1 multifunctionality is determined by its high level of intrinsic disorder that defines the ability of this protein to bind to numerous cellular partners. These results demonstrate a close relationship between BMI1 expression and mitochondrial health in hyperoxia-induced acute lung injury (HALI) and indicate that BMI1 is a potential therapeutic target to treat ALI and Acute Respiratory Distress Syndrome.

Healthcare Utilization and Costs in Sepsis Survivors in Germany-Secondary Analysis of a Prospective Cohort Study

Background: Survivors of sepsis often face long-term sequelae after intensive care treatment. Compared to the period of hospitalization, little is known about the ambulatory healthcare utilization in sepsis patients. The study evaluated healthcare utilization and associated costs of sepsis care including allied health professions after initial hospitalization. Methods: Secondary analysis was performed on data in 210 sepsis patients prospectively enrolled from nine intensive care study centers across Germany. Data was collected via structured surveys among their Primary care (Family-) physicians (PCPs) within the first month after discharge from ICU (baseline) and again at 6, 12 and 24 months after discharge, each relating to the period following the last survey. Costs were assessed by standardized cost unit rates from a health care system’s perspective. Changes in healthcare utilization and costs over time were calculated using the Wilcoxon rank-sum test. Results: Of the 210 patients enrolled, 146 (69.5%) patients completed the 24 months follow-up. In total, 109 patients were hospitalized within the first 6 months post-intensive care. Mean total direct costs per patient at 0–6 months were €17,531 (median: €6047), at 7–12 months €9029 (median: €3312), and at 13–24 months €18,703 (median: €12,828). The largest contributor to the total direct costs within the first 6 months was re-hospitalizations (€13,787 (median: €2965). After this first half year, we observed a significant decline in inpatient care costs for re-hospitalizations (p ≤ 0.001). PCPs were visited by more than 95% of patients over 24 months. Conclusions: Sepsis survivors have high health care utilization. Hospital readmissions are frequent and costly. Highest costs and hospitalizations were observed in more than half of patients within the first six months post-intensive care. Among all outpatient care providers, PCPs were consulted most frequently. Clinical impact: Sepsis survivors have a high healthcare utilization and related costs which persist after discharge from hospital. Within outpatient care, possible needs of sepsis survivors as physiotherapy or psychotherapy seem not to be met appropriately. Development of sepsis aftercare programs for early detection and treatment of complications should be prioritized.

Benefits, risks, and cost-effectiveness of COVID-19 self-tests from a consumer's perspective

BACKGROUND

The aim of this study is to quantify the health benefits, risks, and cost-effectiveness of COVID-19 self-tests from a consumer's perspective in Germany.

METHODS

The analysis is based on a modelling approach using secondary data. The clinical endpoints considered in this analysis are avoided SARS-CoV-2 infections and secondary severe clinical events (death, intensive care unit (ICU) admission, and long COVID syndrome). The study determines the number of self-tests that need to be conducted under a 7-day incidence of 75 per 100,000 population to prevent one infection or severe clinical event. Furthermore, the study calculates the cost of testing per avoided clinical event and quality-adjusted life year (QALY) gained from a consumer perspective.

RESULTS

Disregarding the rate of unreported COVID-19 cases, 4556 self-tests need to be conducted (over 12 years) in order to avoid one undesirable event (death, intensive care unit stay, or long COVID syndrome). Ninety percent of infections are not avoided among direct contacts but along the chain of infection. The costs per quality-adjusted life year gained from a consumer's perspective are €5870. This ratio is particularly sensitive to the 7-day incidence, effective reproduction number, and the age of contacts.

CONCLUSIONS

The benefits of self-testing in the general population at a 7-day incidence rate of 75 per 100,000 appear to be minor. Nevertheless, cost-effectiveness may still be acceptable in the presence of higher-risk contacts given the low costs of self-test kits in Germany.

Continuous Monitoring of Respiratory Rate with Wearable Sensor in Patients Admitted to Hospital with Pneumonia Compared with Intermittent Nurse-Led Monitoring in the United Kingdom: A Cost-Utility Analysis

BACKGROUND

Respiratory rate (RR) is one of the most important physiologic measures for predicting patients' deterioration of clinical condition and final prognosis. In several studies, RR has been the most important predictor of patients' prognoses.

OBJECTIVES

The objective of this study was to conduct a cost-utility analysis to estimate the cost and effectiveness of automatic respiratory rate monitoring (ARRM) with a non-invasive sensor (RespiraSense™) plus intermittent nurse-led RR monitoring (ARRM strategy) compared with intermittent nurse-led RR monitoring (IM strategy) in patients admitted to hospital in the UK with pneumonia.

METHODS

A decision analytic model was developed based on a hypothetical cohort of patients who were admitted to hospital with pneumonia. After admission, the patients could be monitored with either ARRM or IM strategies. The outcomes of interest included total costs and total effectiveness of each strategy, including length of stay (LoS) in hospital, LoS in intensive care unit, quality-adjusted life-years (QALYs), deaths, and incremental cost per QALY gained. An incremental cost of £20,000 or less per QALY gained was considered cost effective. A lifetime time horizon (38 years) was used to capture the long-term benefits. Probabilistic and deterministic sensitivity analyses were performed.

RESULTS

Total costs of patient care in ARRM and IM strategies were £1986.9 million and £2079.4 million, respectively. Total incremental QALYs lived were 3548 higher in the intervention arm (ARRM), meaning that the ARRM strategy was dominant (i.e., less costly [£92.6 million less] and more effective). The results were stable in probabilistic and most of the deterministic sensitivity analyses. Results from threshold analysis indicated that a minimum of 7 and 10% improvement in percentage of early detection of respiratory compromise is required for ARRM to become cost effective and cost saving, respectively.

CONCLUSIONS

Our results indicate that ARRM using RespiraSense, in addition to intermittent nurse-led monitoring of RR, in patients admitted to the hospital with pneumonia could be a cost-saving and cost-effective intervention if the minimum clinical thresholds are met.

A model framework for projecting the prevalence and impact of Long-COVID in the UK

The objective of this paper is to model lost Quality Adjusted Life Years (QALYs) from symptoms arising from COVID-19 disease in the UK population, including symptoms of ‘long-COVID’. The scope includes QALYs lost to symptoms, but not deaths, due to acute COVID-19 and long-COVID. The prevalence of symptomatic COVID-19, encompassing acute symptoms and long-COVID symptoms, was modelled using a decay function. Permanent injury as a result of COVID-19 infection, was modelled as a fixed prevalence. Both parts were combined to calculate QALY loss due to COVID-19 symptoms. Assuming a 60% final attack rate for SARS-CoV-2 infection in the population, we modelled 299,730 QALYs lost within 1 year of infection (90% due to symptomatic COVID-19 and 10% permanent injury) and 557,764 QALYs lost within 10 years of infection (49% due to symptomatic COVID-19 and 51% due to permanent injury). The UK Government willingness-to-pay to avoid these QALY losses would be £17.9 billion and £32.2 billion, respectively. Additionally, 90,143 people were subject to permanent injury from COVID-19 (0.14% of the population). Given the ongoing development in information in this area, we present a model framework for calculating the health economic impacts of symptoms following SARS-CoV-2 infection. This model framework can aid in quantifying the adverse health impact of COVID-19, long-COVID and permanent injury following COVID-19 in society and assist the proactive management of risk posed to health. Further research is needed using standardised measures of patient reported outcomes relevant to long-COVID and applied at a population level.

One-Year Multidisciplinary Follow-Up of Patients With COVID-19 Requiring Invasive Mechanical Ventilation

Objectives

Patients with COVID-19 frequently develop acute respiratory distress syndrome (ARDS) requiring intensive care unit (ICU) admission. Data on long-term survival of these patients are lacking. The authors investigated 1-year survival, quality of life, and functional recovery of patients with COVID-19 ARDS requiring invasive mechanical ventilation.

Design

Prospective observational study.

Setting

Tertiary-care university hospital.

Participants

All patients with COVID-19 ARDS receiving invasive mechanical ventilation and discharged alive from hospital.

Interventions

Patients were contacted by phone after 1 year. Functional, cognitive, and psychological outcomes were explored through a questionnaire and assessed using validated scales. Patients were offered the possibility to undergo a follow-up chest computed tomography (CT) scan.

Measurements and Main Results

The study included all adult (age ≥18 years) patients with COVID-19–related ARDS admitted to an ICU of the authors’ institution between February 25, 2020, and April 27, 2020, who received at least 1 day of invasive mechanical ventilation (IMV). Of 116 patients who received IMV, 61 (52.6%) survived to hospital discharge. These survivors were assessed 1 year after discharge and 56 completed a battery of tests of cognition, activities of daily living, and interaction with family members. They had overall good functional recovery, with >80% reporting good recovery and no difficulties in usual activities. A total of 52 (93%) of patients had no dyspnea at rest. Severe anxiety/depression was reported by 5 (8.9%) patients. Comparing 2-month and 1-year data, the authors observed the most significant improvements in the areas of working status and exertional dyspnea. One-year chest CT scans were available for 36 patients; fibrotic-like changes were present in 4 patients.

Conclusions

All patients who survived the acute phase of COVID-19 and were discharged from the hospital were alive at the 1-year follow up, and the vast majority of them had good overall recovery and quality of life.

The Cost of ARDS: A Systematic Review

BACKGROUND

ARDS is an inflammatory condition of the lungs and is a common condition in adult ICUs. The resources required and costs of care for patients with ARDS are significant because of the severity of the illness and extended ICU lengths of stay.

RESEARCH QUESTION

What are the costs associated with ARDS?

STUDY DESIGN AND METHODS

We systematically searched the literature through April 29, 2021, for articles relevant to ARDS and costs. MEDLINE, Embase, Central, and EconLit databases were searched, and articles that reported on cost data from an original publication in adult patients with ARDS were included. Two authors independently assessed articles for inclusion and extracted data elements related to costs, methodology, health-care system type, economic perspective, and clinical data. Publication quality was assessed using a modified version of the Quality of Health Economic Studies Instrument.

RESULTS

Four thousand six hundred sixty-three publications were found, of which 110 were included for full-text review (κ = 0.72). A total of 22 publications (49,483 patients) were suitable for data extraction. The publications represented a broad range of health-care systems, economic perspectives, costing methodology, and time frames. Mean inpatient costs ranged from $8,476 (2021 US dollars [USD]) to $547,974 (2021 USD) and were highest in publications of lower quality and in American health systems and were associated with trauma cohorts. Outpatient costs were highest in publications with higher readmission rates, longer durations of follow-up, and in American health systems.

INTERPRETATION

A wide range of costing data is available for ARDS. A comprehensive synthesis of this literature frames the reasons for this and allows estimates to reflect the context in which they were assessed. This information will be of value to researchers and administrators interested in the economics of caring for patients with ARDS.

TRIAL REGISTRY

PROSPERO; No.: CRD42020192487.

Predicting Duration of Mechanical Ventilation in Acute Respiratory Distress Syndrome Using Supervised Machine Learning

Background: Acute respiratory distress syndrome (ARDS) is an intense inflammatory process of the lungs. Most ARDS patients require mechanical ventilation (MV). Few studies have investigated the prediction of MV duration over time. We aimed at characterizing the best early scenario during the first two days in the intensive care unit (ICU) to predict MV duration after ARDS onset using supervised machine learning (ML) approaches. Methods: For model description, we extracted data from the first 3 ICU days after ARDS diagnosis from patients included in the publicly available MIMIC-III database. Disease progression was tracked along those 3 ICU days to assess lung severity according to Berlin criteria. Three robust supervised ML techniques were implemented using Python 3.7 (Light Gradient Boosting Machine (LightGBM); Random Forest (RF); and eXtreme Gradient Boosting (XGBoost)) for predicting MV duration. For external validation, we used the publicly available multicenter database eICU. Results: A total of 2466 and 5153 patients in MIMIC-III and eICU databases, respectively, received MV for >48 h. Median MV duration of extracted patients was 6.5 days (IQR 4.4–9.8 days) in MIMIC-III and 5.0 days (IQR 3.0–9.0 days) in eICU. LightGBM was the best model in predicting MV duration after ARDS onset in MIMIC-III with a root mean square error (RMSE) of 6.10–6.41 days, and it was externally validated in eICU with RMSE of 5.87–6.08 days. The best early prediction model was obtained with data captured in the 2nd day. Conclusions: Supervised ML can make early and accurate predictions of MV duration in ARDS after onset over time across ICUs. Supervised ML models might have important implications for optimizing ICU resource utilization and high acute cost reduction of MV.

Roles of Early Mobilization Program in Preventing Muscle Weakness and Decreasing Psychiatric Disorders in Patients with Coronavirus Disease 2019 Pneumonia: A Retrospective Observational Cohort Study

Although many coronavirus 2019 patients have experienced persistent symptoms and a long-term decline in quality of life after discharge, the details of these persistent symptoms and the effect of early rehabilitation are still unclear. We conducted a single-center, retrospective observational study to investigate the prevalence of persistent symptoms three months after discharge from the intensive care unit by checking the medical records. All patients received an early mobilization program. Four out of 13 patients (31%) had postintensive care syndrome. No patients had muscle weakness, and 11 patients (85%) returned to their previous work. However, psychiatric disorder, such as anxiety (23%) and posttraumatic stress disorder (15%), were observed. Eleven patients claimed persistent symptoms, including fatigue and numbness in the extremities. Our results suggest that the implementation of an early rehabilitation program plays some role in preventing muscle weakness and that decreasing psychiatric disorders should be a next target of patient care in the intensive care unit.

Prediction of weaning from mechanical ventilation using Convolutional Neural Networks

Weaning from mechanical ventilation covers the process of liberating the patient from mechanical support and removing the associated endotracheal tube. The management of weaning from mechanical ventilation comprises a significant proportion of the care of critically ill intubated patients in Intensive Care Units (ICUs). Both prolonged dependence on mechanical ventilation and premature extubation expose patients to an increased risk of complications and increased health care costs. This work aims to develop a decision support model using routinely-recorded patient information to predict extubation readiness. In order to do so, we have deployed Convolutional Neural Networks (CNN) to predict the most appropriate treatment action in the next hour for a given patient state, using historical ICU data extracted from MIMIC-III. The model achieved 86% accuracy and 0.94 area under the receiver operating characteristic curve (AUC-ROC). We also performed feature importance analysis for the CNN model and interpreted these features using the DeepLIFT method. The results of the feature importance assessment show that the CNN model makes predictions using clinically meaningful and appropriate features. Finally, we implemented counterfactual explanations for the CNN model. This can help clinicians understand what feature changes for a particular patient would lead to a desirable outcome, i.e. readiness to extubate.

A preliminary cost-effectiveness analysis of lung protective ventilation with extra corporeal carbon dioxide removal (ECCO2R) in the management of acute respiratory distress syndrome (ARDS)

Background

Mechanical ventilation (MV) is the cornerstone in the management of the acute respiratory distress syndrome (ARDS). Recent research suggests that decreasing the intensity of MV using lung protective ventilation (LPV) with lower tidal volume (Vt) and driving pressure (∆P) could improve survival. Extra-corporal CO₂ removal (ECCO₂R) precisely enables LPV by allowing lower Vt, ∆P and mechanical power while maintaining PaCO₂ within a physiologic range. This study evaluates the potential cost-effectiveness of ECCO₂R-enabled LPV in France.

Methods

We modelled the distribution over time of ventilated ARDS patients across 3 health-states (alive & ventilated, alive & weaned from ventilation, dead). We compared the outcomes of 3 strategies: MV (no ECCO₂R), LPV (ECCO₂R when PaCO₂ > 55 mmHg) and Ultra-LPV (ECCO₂R for all). Patients characteristics, ventilation settings, survival and lengths of stay were derived from a large ARDS epidemiology study. Survival benefits associated with lower ∆P were taken from the analysis of more than 3000 patients enrolled in 9 randomized trials. Health outcomes were expressed in quality-adjusted life years (QALYs). Incremental cost-effectiveness ratios (ICERs) were computed with both Day 60 cost and Lifetime cost.

Results

Both LPV and ULPV as enabled by ECCO2R provided favorable results at Day 60 as compared to MV. Survival rates were increased with the protective strategies, notably with ULPV that provided even more manifest benefits as compared to MV. LPV and ULPV produced +0.162 and + 0.627 incremental QALYs as compared to MV, respectively. LPV and ULPV costs were augmented because of their survival benefits. Nonetheless, ICERs of LPV and ULPV vs. MV were all well below the €50,000 threshold. ULPV also presented with favorable ICERs as compared to LPV (i.e. less than €25,000/QALY).

Conclusions

ECCO₂R-enabled LPV strategies might provide cost-effective survival benefit. Additional data from interventional and observational studies are needed to support this preliminary model-based analysis.

One year after ICU admission for severe community-acquired pneumonia of bacterial, viral or unidentified etiology. What are the outcomes?

INTRODUCTION

Multiplex polymerase chain reaction (mPCR) for respiratory virus testing is increasingly used in community-acquired pneumonia (CAP), however data on one-year outcome in intensive care unit (ICU) patients with reference to the causative pathogen are scarce.

MATERIALS AND METHODS

We performed a single-center retrospective study in 123 ICU patients who had undergone respiratory virus testing for CAP by mPCR and with known one-year survival status. Functional status including dyspnea (mMRC score), autonomy (ADL Katz score) and need for new home-care ventilatory support was assessed at a one-year post-ICU follow-up. Mortality rates and functional status were compared in patients with CAP of a bacterial, viral or unidentified etiology one year after ICU admission.

RESULTS

The bacterial, viral and unidentified groups included 19 (15.4%), 37 (30.1%), and 67 (54.5%) patients, respectively. In multivariate analysis, one-year mortality in the bacterial group was higher compared to the viral group (HR 2.92, 95% CI 1.71-7.28, p = 0.02) and tended to be higher compared to the unidentified etiology group (p = 0.06); but no difference was found between the viral and the unidentified etiology group (p = 0.43). In 64/83 one-year survivors with a post-ICU follow-up consultation, there were no differences in mMRC score, ADL Katz score and new home-care ventilatory support between the groups (p = 0.52, p = 0.37, p = 0.24, respectively). Severe dyspnea (mMRC score = 4 or death), severe autonomy deficiencies (ADL Katz score ≤ 2 or death), and major adverse respiratory events (new home-care ventilatory support or death) were observed in 52/104 (50.0%), 47/104 (45.2%), and 65/104 (62.5%) patients, respectively; with no difference between the bacterial, viral and unidentified group: p = 0.58, p = 0.06, p = 0.61, respectively.

CONCLUSIONS

CAP of bacterial origin had a poorer outcome than CAP of viral or unidentified origin. At one-year, impairment of functional status was frequently observed, with no difference according to the etiology.

[Health services research in intensive care medicine in Germany : Status quo and future challenges exemplified by acute pulmonary failure]

Health services research (HSR) is a multidisciplinary field of research that describes disease treatment and health care and their framework conditions. In the last 20 years, the HSR aspect became more and more the clinical focus of intensive care medicine. Under this aspect HSR investigates the use of clinical measures and their impact on patient outcome under routine intensive care medical conditions. This article provides an overview of the current state of HSR in intensive care medicine in Germany using the example of acute respiratory distress syndrome (ARDS). The ARDS still represents a clinical disease with high intra-hospital mortality (30-60%) despite progress in intensive care medicine. Survivors of ARDS have substantial long-term limitations on physical and mental health. The treatment of ARDS patients is tedious, laborious for intensive care unit staff and complex. Despite evident treatment recommendations, these are only insufficiently implemented in the clinical routine. With the help of quality indicators, benchmarking, certification and peer review procedures, the quality of intensive care treatment in the clinical routine can be documented and improved. An important role in HSR is patient safety and focusing on the outcome with evaluation of the patient's will. As part of the establishment of the innovation fund for HSR, promising intensive medical care projects have been promoted to improve the quality of care and the quality of long-term outcome for intensive care patients. An important focus lies on the identification of factors that improve long-term quality of life after intensive care. The expansion of registries and telemedicine in intensive care offers the opportunity to bundle and share experiences more effectively and thereby establish (guideline-based) treatment recommendations faster in the clinical practice.

Approaches to Addressing Post-Intensive Care Syndrome among Intensive Care Unit Survivors. A Narrative Review

Critical illness can be lethal and devastating to survivors. Improvements in acute care have increased the number of intensive care unit (ICU) survivors. These survivors confront a range of new or worsened health states that collectively are commonly denominated post-intensive care syndrome (PICS). These problems include physical, cognitive, psychological, and existential aspects, among others. Burgeoning interest in improving long-term outcomes for ICU survivors has driven an array of potential interventions to improve outcomes associated with PICS. To date, the most promising interventions appear to relate to very early physical rehabilitation. Late interventions within aftercare and recovery clinics have yielded mixed results, although experience in heart failure programs suggests the possibility that very early case management interventions may help improve intermediate-term outcomes, including mortality and hospital readmission. Predictive models have tended to underperform, complicating study design and clinical referral. The complexity of the health states associated with PICS suggests that careful and rigorous evaluation of multidisciplinary, multimodality interventions-tied to the specific conditions of interest-will be required to address these important problems.

Cost-Effectiveness of Parenteral Nutrition Containing ω-3 Fatty Acids in Hospitalized Adult Patients From 5 European Countries and the US

Background

ω‐3 Fatty acid (FA)–containing parenteral nutrition (PN) is associated with improvements in patient outcomes and with reductions in hospital length of stay (HLOS) vs standard PN regimens (containing non–ω‐3 FA lipid emulsions). We present a cost‐effectiveness analysis of ω‐3 FA–containing PN vs standard PN in 5 European countries (France, Germany, Italy, Spain, UK) and the US.

Methods

This pharmacoeconomic model was based on estimates of ω‐3 efficacy reported in a recent meta‐analysis and data from country‐specific sources. It utilized a probabilistic discrete event simulation model to compare ω‐3 FA–containing PN with standard PN in a population of critically ill and general ward patients. The influence of model parameters was evaluated using probabilistic and deterministic sensitivity analyses.

Results

Overall costs were reduced with ω‐3 FA–containing PN in all 6 countries compared with standard PN, ranging from €1741 (±€1284) in Italy to €5576 (±€4193) in the US. Expenses for infections and HLOS were lower in all countries for ω‐3 FA–containing PN vs standard PN, with the largest cost differences for both in the US (infection: €825 ± €4001; HLOS: €4879 ± €1208) and the smallest savings in the UK for infections and in Spain for HLOS (€63 ± €426 and €1636 ± €372, respectively).

Conclusion

This cost‐effectiveness analysis in 6 countries demonstrates that the superior clinical efficacy of ω‐3 FA–containing PN translates into significant decreases in mean treatment cost, rendering it an attractive cost‐saving alternative to standard PN across different healthcare systems.

Acute respiratory distress syndrome: how do patients fare after the intensive care unit?

Patients with acute respiratory distress syndrome require ventilation strategies that have been shown to be important for reducing short-term mortality, such as protective ventilation and prone position ventilation. However, patients who survive have a prolonged stay in both the intensive care unit and the hospital, and they experience a reduction in overall satisfaction with life (independence, acceptance and positive outlook) as well as decreased mental health (including anxiety, depression and posttraumatic stress disorder symptoms), physical health (impaired physical state and activities of daily living; fatigue and muscle weakness), social health and the ability to participate in social activities (including relationships with friends and family, hobbies and social gatherings).

Influence of quality of intensive care on quality of life/return to work in survivors of the acute respiratory distress syndrome: prospective observational patient cohort study (DACAPO)

Background

Significant long-term reduction in health-related quality of life (HRQoL) is often observed in survivors of the acute respiratory distress syndrome (ARDS), and return to work (RtW) is limited. There is a paucity of data regarding the relationship between the quality of care (QoC) in the intensive care unit (ICU) and both HRQoL and RtW in ARDS survivors. Therefore, the aim of our study was to investigate associations between indicators of QoC and HRQoL and RtW in a cohort of survivors of ARDS.

Methods

To determine the influence of QoC on HRQoL and RtW 1 year after ICU-discharge, ARDS patients were recruited into a prospective multi-centre patient cohort study and followed up regularly after discharge. Patients were asked to complete self-report questionnaires on HRQoL (Short Form 12 physical component scale (PCS) and mental component scale (MCS)) and RtW. Indicators of QoC pertaining to volume, structural and process quality, and general characteristics were recorded on ICU level. Associations between QoC indicators and HrQoL and RtW were investigated by multivariable linear and Cox regression modelling, respectively. B values and hazard ratios (HRs) are reported with corresponding 95% confidence intervals (CIs).

Results

877 (of initially 1225 enrolled) people with ARDS formed the DACAPO survivor cohort, 396 were finally followed up to 1 year after discharge. The twelve-month survivors were characterized by a reduced HRQoL with a greater impairment in the physical component (Md 41.2 IQR [34–52]) compared to the mental component (Md 47.3 IQR [33–57]). Overall, 50% of the patients returned to work. The proportion of ventilated ICU patients showed significant negative associations with both 12 months PCS (B = − 11.22, CI −20.71; − 1,74) and RtW (HR = 0,18, CI 0,04;0,80). All other QoC indicators were not significantly related to outcome.

Conclusions

Associations between ICU QoC and long-term HrQoL and RtW were weak and largely non-significant. Residual confounding by case mix, treatment variables before or during ICU stay and variables pertaining to the post intensive care period (e.g. rehabilitation) cannot be ruled out.

Trial registration

Clinicaltrials.govNCT02637011.

(December 22, 2015, retrospectively registered)

The Potential Health Care Costs And Resource Use Associated With COVID-19 In The United States

With the coronavirus disease 2019 (COVID-19) pandemic, one of the major concerns is the direct medical cost and resource use burden imposed on the US health care system. We developed a Monte Carlo simulation model that represented the US population and what could happen to each person who got infected. We estimated resource use and direct medical costs per symptomatic infection and at the national level, with various "attack rates" (infection rates), to understand the potential economic benefits of reducing the burden of the disease. A single symptomatic COVID-19 case could incur a median direct medical cost of $3,045 during the course of the infection alone. If 80 percent of the US population were to get infected, the result could be a median of 44.6 million hospitalizations, 10.7 million intensive care unit (ICU) admissions, 6.5 million patients requiring a ventilator, 249.5 million hospital bed days, and $654.0 billion in direct medical costs over the course of the pandemic. If 20 percent of the US population were to get infected, there could be a median of 11.2 million hospitalizations, 2.7 million ICU admissions, 1.6 million patients requiring a ventilator, 62.3 million hospital bed days, and $163.4 billion in direct medical costs over the course of the pandemic.

Association of Economic Status and Mortality in Patients with Acute Respiratory Distress Syndrome

The high cost of treatment for acute respiratory distress syndrome (ARDS) is a concern for healthcare systems, while the impact of patients’ socio-economic status on the risk of ARDS-associated mortality remains controversial. This study investigated associations between patients’ income at the time of ARDS diagnosis and ARDS-specific mortality rate after treatment initiation. Data from records provided by the National Health Insurance Service of South Korea were used. Adult patients admitted for ARDS treatment from 2013 to 2017 were included in the study. Patients’ income in the year of diagnosis was evaluated. A total of 14,600 ARDS cases were included in the analysis. The 30-day and 1-year mortality rates were 48.6% and 70.3%, respectively. In multivariable Cox regression model, we compared income quartiles, showing that compared to income strata Q1, the Q2 (p = 0.719), Q3 (p = 0.946), and Q4 (p = 0.542) groups of income level did not affect the risk of 30-day mortality, respectively. Additionally, compared to income strata Q1, the Q2 (p = 0.762), Q3 (p = 0.420), and Q4 (p = 0.189) strata did not affect the risk of 1-year mortality. Patient income at the time of ARDS diagnosis did not affect the risk of 30-day or 1-year mortality in the present study based on South Korea’s health insurance data.

The association of intensive care with utilization and costs of outpatient healthcare services and quality of life

Background

Little is known about outpatient health services use following critical illness and intensive care. We examined the association of intensive care with outpatient consultations and quality of life in a population-based sample.

Methods

Cross-sectional analysis of data from 6,686 participants of the Study of Health in Pomerania (SHIP), which consists of two independent population-based cohorts. Statistical modeling was done using Poisson regression, negative binomial and generalized linear models for consultations, and a fractional response model for quality of life (EQ-5D-3L index value), with results expressed as prevalence ratios (PR) or percent change (PC). Entropy balancing was used to adjust for observed confounding.

Results

ICU treatment in the previous year was reported by 139 of 6,686 (2,1%) participants, and was associated with a higher probability (PR 1.05 [CI:1.03;1.07]), number (PC +58.0% [CI:22.8;103.2]) and costs (PC +64.1% [CI:32.0;103.9]) of annual outpatient consultations, as well as with a higher number of medications (PC +37.8% [CI:17.7;61.5]). Participants with ICU treatment were more likely to visit a specialist (PR 1.13 [CI:1.09; 1.16]), specifically internal medicine (PR 1.67 [CI:1.45;1.92]), surgery (PR 2.42 [CI:1.92;3.05]), psychiatry (PR 2.25 [CI:1.30;3.90]), and orthopedics (PR 1.54 [CI:1.11;2.14]). There was no significant effect regarding general practitioner consultations. ICU treatment was also associated with lower health-related quality of life (EQ-5D index value: PC -13.7% [CI:-27.0;-0.3]). Furthermore, quality of life was inversely associated with outpatient consultations in the previous month, more so for participants with ICU treatment.

Conclusions

Our findings suggest that ICU treatment is associated with an increased utilization of outpatient specialist services, higher medication intake, and impaired quality of life.

Extracorporeal membrane oxygenation as rescue therapy for severe hypoxemic respiratory failure

Extracorporeal membrane oxygenation (ECMO) has been used for more than 50 years as salvage therapy for patients with severe cardiopulmonary failure refractory to conventional treatment. ECMO was first used in the 1960s to treat hypoxemic respiratory failure in newborns. On the basis of its success in that population, ECMO began to be used in the early 1970s to treat adult hypoxemic respiratory failure. However, outcomes for adults were, somewhat perplexingly, quite poor. By the 1980s, use of ECMO for severe hypoxemia was rare outside of the pediatric population. ECMO technology, however, continued to evolve and improve. Multiple case reports and small series describing ECMO use as rescue for adults with severe hypoxemia from various lung pathologies have appeared in the literature over the past three decades. Adult respiratory distress syndrome (ARDS) is often the final common pathway of various pathologies affecting adults and causing hypoxemic respiratory failure. It is prevalent in intensive care units throughout the world and has, since it was first described in 1967, carried a high mortality. No specific therapy for ARDS has been found, and current care is supportive, primarily by mechanical ventilation. Results from recent randomized controlled trials, however, suggest that ECMO may have a place in the treatment of these patients. This article reviews these studies and recommends adding severe ARDS to the list of established indications for ECMO in patients with hypoxemic respiratory failure.

The Cost Effectiveness of Single-Patient-Use Electrocardiograph Cable and Lead Systems in Monitoring for Coronary Artery Bypass Graft Surgery

Background: During admission for coronary artery bypass graft (CABG) surgery patients receive electrocardiograph (ECG) monitoring; for which reusable ECG cable and leads (rECG) are standard. Objective: Evaluate the cost effectiveness of a single-patient-use ECG cable and lead system (spECG). Methods: Review of the Medicare 2011-2014 database followed by a cost-effectiveness model considering a Medicare facility transitioning from rECG ($9 per patient) to spECG ($15). In-hospital ECG monitoring was for ≤8 days. In the model, patients underwent CABG and recovered in the intensive care unit, before transfer to the general ward and discharge. Surgical site infection (SSI) resulted in increased length of stay, readmission, or outpatient care. Health outcomes impacted EQ-5D-measured quality adjusted life years (QALYs). Health and cost outcomes were discounted at 3.5% annually. All costs in 2016 USD. Significance (95% level) was assessed via 2,000 simulations. Results: In 2014, 5.49% of patients had an SSI by 90-days post-surgery, with spECG reducing the odds of an SSI (odds ratio: 0.74, 0.62-0.89). Mean 40-year, per-patient costs to Medicare were $65,497 with rECG and $65,048 with spECG. The $450 saving was significant, with a median (95% credible interval) reduction of $466 ($174 to $989). Cost drivers were days required to treat inpatient SSIs. QALYs increases with spECG were significant but minor (median increase 0.008). Medicare savings may total $40 million per year with use of spECG. Conclusions: Post-operative SSI is a concern for Medicare patients undergoing CABG, and use of spECG is likely to provide cost and patient benefits.

Long-term outcome after the acute respiratory distress syndrome: different from general critical illness?

Purpose of review

To review the current research data on long-term outcome and health-related quality of life in survivors of the acute respiratory distress syndrome (ARDS) and to compare these findings with those from non-ARDS patients surviving critical illness.

Recent findings

Between 6 months and 2 years after discharge from ICU, survivors of ARDS present with substantial impairments of the levels of body function (muscle strength, walking capacity and/or physical activity (physical SF-36 score). In contrast to non-ARDS patients from surgical ICUs, a standardized intensified physical therapy during early course of illness in ARDS patients could not show an improvement of long-term physical function performance. Furthermore, a substantial part of further ARDS patients suffer from depression (26–33%), anxiety (38–44%) or posttraumatic stress disorder (22–24%). In general, the level of functional autonomy and daily life activities was reduced, and in one study, 6 months after ICU-discharge this level was significantly lower in ARDS patients compared with non-ARDS patients. In a recent study, 44% of ARDS survivors were jobless 1 year after critical illness, whereas half of previously employed patients returned to work within 4 months after hospital discharge. General health-related quality of life was significantly reduced compared with a matched population in all studies.

Summary

Surviving ARDS is associated with a long-term substantial reduction in health-related quality of life and such a reduction does not differ from findings in patients surviving other critical illness. In further research, a special attention should be paid to prevention measures of the ‘post intensive care syndrome’ as well as to patient important domains, which might better explain the patient's and families’ demands.

Evaluating the Cost-Effectiveness of Proportional-Assist Ventilation Plus vs. Pressure Support Ventilation in the Intensive Care Unit in Two Countries

Background: Mechanical ventilation is an integral, but expensive, part of the intensive care unit (ICU). Optimal use of mechanical ventilation could save costs and improve patient outcomes. Here, the cost effectiveness of proportional assist ventilation (PAV™ ventilation by Medtronic) is estimated relative to pressure support ventilation (PSV).

Methods: A cohort-level, clinical model was built using data from clinical trials. The model estimates patient-ventilator asynchrony >10%, tracheostomy, ventilator-associated pneumonia, other nosocomial infections, spontaneous breathing trial success, hypoxemia, and death. Cost and quality of life are associated with all events, with cost effectiveness defined as the cost per quality-adjusted life year (QALY) gained in the US and UK.

Results: The mean cost of ICU care was lower with PAV™ than with PSV in the US and UK, but the total cost of care over 40 years was higher due to more patients surviving and incurring future care costs. Reduced time on mechanical ventilation, fewer nosocomial infections, and extended life expectancy with PAV™ drove QALY improvement. The cost per QALY gained with PAV™ was $8,628 and £2,985.

Conclusion: PAV™ improves quality of life and reduces short-term costs. PAV™ is likely to be considered cost-effective over 40-years in the US and UK.

Costs and expected gain in lifetime health from intensive care versus general ward care of 30,712 individual patients: a distribution-weighted cost-effectiveness analysis

BACKGROUND

Clinicians, hospital managers, policy makers, and researchers are concerned about high costs, increased demand, and variation in priorities in the intensive care unit (ICU). The objectives of this modelling study are to describe the extra costs and expected health gains associated with admission to the ICU versus the general ward for 30,712 patients and the variation in cost-effectiveness estimates among subgroups and individuals, and to perform a distribution-weighted economic evaluation incorporating extra weighting to patients with high severity of disease.

METHODS

We used a decision-analytic model that estimates the incremental cost per quality-adjusted life year (QALY) gained (ICER) from ICU admission compared with general ward care using Norwegian registry data from 2008 to 2010. We assigned increasing weights to health gains for those with higher severity of disease, defined as less expected lifetime health if not admitted. The study has inherent uncertainty of findings because a randomized clinical trial comparing patients admitted or rejected to the ICU has never been performed. Uncertainty is explored in probabilistic sensitivity analysis.

RESULTS

The mean cost-effectiveness of ICU admission versus ward care was €11,600/QALY, with 1.6 QALYs gained and an incremental cost of €18,700 per patient. The probability (p) of cost-effectiveness was 95% at a threshold of €22,000/QALY. The mean ICER for medical admissions was €10,700/QALY (p = 97%), €12,300/QALY (p = 93%) for admissions after acute surgery, and €14,700/QALY (p = 84%) after planned surgery. For individualized ICERs, there was a 50% probability that ICU admission was cost-effective for 85% of the patients at a threshold of €64,000/QALY, leaving 15% of the admissions not cost-effective. In the distributional evaluation, 8% of all patients had distribution-weighted ICERs (higher weights to gains for more severe conditions) above €64,000/QALY. High-severity admissions gained the most, and were more cost-effective.

CONCLUSIONS

On average, ICU admission versus general ward care was cost-effective at a threshold of €22,000/QALY (p = 95%). According to the individualized cost-effectiveness information, one in six ICU admissions was not cost-effective at a threshold of €64,000/QALY. Almost half of these admissions that were not cost-effective can be regarded as acceptable when weighted by severity of disease in terms of expected lifetime health. Overall, existing ICU services represent reasonable resource use, but considerable uncertainty becomes evident when disaggregating into individualized results.