Variation in severity-adjusted resource use and outcome in intensive care units

EPOS-OHCA: Early Predictors of Outcome and Survival after non-traumatic Out-of-Hospital Cardiac Arrest

Background

Post-cardiac arrest syndrome (PCAS) after out-of-hospital cardiac arrest (OHCA) poses significant challenges due to its complex pathomechanisms involving inflammation, ischemia, and reperfusion injury. The identification of early available prognostic indicators is essential for optimizing therapeutic decisions and improving patient outcomes.

Methods

In this retrospective single-center study, we analyzed real-world data from 463 OHCA patients with either prehospital or in-hospital return of spontaneous circulation (ROSC), treated at the Cardiac Arrest Center of the University Hospital of Marburg (MCAC) from January 2018 to December 2022. We evaluated demographic, prehospital, and clinical variables, including initial rhythms, resuscitation details, and early laboratory results. Statistical analyses included logistic regression to identify predictors of survival and neurological outcomes.

Results

Overall, 46.9% (n = 217) of patients survived to discharge, with 70.1% (n = 152) achieving favorable neurological status (CPC 1 or 2). Age, initial shockable rhythm, resuscitation time to return of spontaneous circulation (ROSC), and early laboratory parameters like lactate, C-reactive protein, and glomerular filtration rate were identified as independent and combined Early Predictors of Outcome and Survival (EPOS), with high significant predictive value for survival (AUC 0.86 [95% CI 0.82-0.89]) and favorable neurological outcome (AUC 0.84 [95% CI 0.80-0.88]).

Conclusion

Integration of EPOS into clinical procedures may significantly improve clinical decision making and thus patient prognosis in the early time-crucial period after OHCA. However, further validation in other patient cohorts is needed.

Effect of mortality prediction models on resource use benchmarking of intensive care units

PURPOSE

Intensive care requires extensive resources. The ICUs' resource use can be compared using standardized resource use ratios (SRURs). We assessed the effect of mortality prediction models on the SRURs.

MATERIALS AND METHODS

We compared SRURs using different mortality prediction models: the recent Finnish Intensive Care Consortium (FICC) model and the SAPS-II model (n = 68,914 admissions). We allocated the resources to severity of illness strata using deciles of predicted mortality. In each risk and year stratum, we calculated the expected resource use per survivor from our modelling approaches using length of ICU stay and Therapeutic Intervention Scoring System (TISS) points.

RESULTS

Resource use per survivor increased from one length of stay (LOS) day and around 50 TISS points in the first decile to 10 LOS-days and 450 TISS in the tenth decile for both risk scoring systems. The FICC model predicted mortality risk accurately whereas the SAPS-II grossly overestimated the risk of death. Despite this, SRURs were practically identical and consistent.

CONCLUSIONS

SRURs provide a robust tool for benchmarking resource use within and between ICUs. SRURs can be used for benchmarking even if recently calibrated risk scores for the specific population are not available.

Eleven years impact of a stepwise educational program on healthcare associated infections and antibiotics consumption in an intensive care unit of a tertiary hospital in Brazil

BACKGROUND

Hospital acquired infections (HAI) and liberal use of broad-spectrum antibiotics are common in intensive care unit(ICU)s of low-middle income countries. We investigated the long-term association of a stepwise multifaceted educational program with the incidence of HAIs and antibiotics use in a Brazilian ICU. We also evaluated the program's cost impact.

METHODS

We retrieved data from a prospective daily collected database of a twelve bedrooms ICU, all admitted patients within a period of eleven years were enrolled.

FINDINGS

From 03/15/2007 to 09/11/2019, we admitted 3059 patients where 2406 (79%) survived the ICU stay. Median age was 51 years-old, and median SAPS3 was 53. The initial density of catheter related blood infection (4.3 events / 1000 patients-day), urinary tract infection (9.2 event / 1000 patients-day) and ventilator associated pneumonia (54.9 events / 1000 patients-day) felt during the observed period to (0.35 events / 1000 patients-day), (0 events / 1000 patients-day), and (1.5 events / 1000 patients-day) respectively. The days of antibiotic therapy also decreased from 797.9 days of therapy / 1000 patients day to 292.3 days of therapy / 1000 patients day. The total cost per patient also decreased. The adjusted mortality rate was steady during the studied period from 23.2% to 22.9%.

INTERPRETATION

A stepwise multifaceted educational program is an effective way to reduce hospital-associated infections, improve the rational use of antibiotics, and reduce costs. This impact occurred in a long term, and is probably consistent.

Implementing an ICU registry in Ethiopia-Implications for critical care quality improvement

PURPOSE

Intensive care units (ICUs) in low- and middle-income countries have high mortality rates, and clinical data are needed to guide quality improvement (QI) efforts. This study utilizes data from a validated ICU registry specially developed for resource-limited settings to identify evidence-based QI priorities for ICUs in Ethiopia.

MATERIALS AND METHODS

A retrospective cohort analysis of data from two tertiary referral hospital ICUs in Addis Ababa, Ethiopia from July 2021-June 2022 was conducted to describe casemix, complications and outcomes and identify features associated with ICU mortality.

RESULTS

Among 496 patients, ICU mortality was 35.3%. The most common reasons for ICU admission were respiratory failure (24.0%), major head injury (17.5%) and sepsis/septic shock (13.3%). Complications occurred in 41.0% of patients. ICU mortality was higher among patients with respiratory failure (46.2%), sepsis (66.7%) and vasopressor requirements (70.5%), those admitted from the hospital ward (64.7%), and those experiencing major complications in the ICU (62.3%).

CONCLUSIONS

In this study, ICU mortality was high, and complications were common and associated with increased mortality. ICU registries are invaluable tools to understand local casemix and clinical outcomes, especially in resource-limited settings. These findings provide a foundation for QI efforts and a baseline to evaluate their impact.

The Association Between Prepandemic ICU Performance and Mortality Variation in COVID-19: A Multicenter Cohort Study of 35,619 Critically Ill Patients

BACKGROUND

During the COVID-19 pandemic, ICUs remained under stress and observed elevated mortality rates and high variations of outcomes. A knowledge gap exists regarding whether an ICU performing best during nonpandemic times would still perform better when under high pressure compared with the least performing ICUs.

RESEARCH QUESTION

Does prepandemic ICU performance explain the risk-adjusted mortality variability for critically ill patients with COVID-19?

STUDY DESIGN AND METHODS

This study examined a cohort of adults with real-time polymerase chain reaction-confirmed COVID-19 admitted to 156 ICUs in 35 hospitals from February 16, 2020, through December 31, 2021, in Brazil. We evaluated crude and adjusted in-hospital mortality variability of patients with COVID-19 in the ICU during the pandemic. Association of baseline (prepandemic) ICU performance and in-hospital mortality was examined using a variable life-adjusted display (VLAD) during the pandemic and a multivariable mixed regression model adjusted by clinical characteristics, interaction of performance with the year of admission, and mechanical ventilation at admission.

RESULTS

Thirty-five thousand six hundred nineteen patients with confirmed COVID-19 were evaluated. The median age was 52 years, median Simplified Acute Physiology Score 3 was 42, and 18% underwent invasive mechanical ventilation. In-hospital mortality was 13% and 54% for those receiving invasive mechanical ventilation. Adjusted in-hospital mortality ranged from 3.6% to 63.2%. VLAD in the most efficient ICUs was higher than the overall median in 18% of weeks, whereas VLAD was 62% and 84% in the underachieving and least efficient groups, respectively. The least efficient baseline ICU performance group was associated independently with increased mortality (OR, 2.30; 95% CI, 1.45-3.62) after adjusting for patient characteristics, disease severity, and pandemic surge.

INTERPRETATION

ICUs caring for patients with COVID-19 presented substantial variation in risk-adjusted mortality. ICUs with better baseline (prepandemic) performance showed reduced mortality and less variability. Our findings suggest that achieving ICU efficiency by targeting improvement in organizational aspects of ICUs may impact outcomes, and therefore should be a part of the preparedness for future pandemics.

The Influence of Potential Organ Donors on Standardized Mortality Ratios and ICU Benchmarking

OBJECTIVES

The standardized mortality ratio (SMR) is a common metric to benchmark ICUs. However, SMR may be artificially distorted by the admission of potential organ donors (POD), who have nearly 100% mortality, although risk prediction models may not identify them as high-risk patients. We aimed to evaluate the impact of PODs on SMR.

DESIGN

Retrospective registry-based multicenter study.

SETTING

Twenty ICUs in Finland, Estonia, and Switzerland in 2015-2017.

PATIENTS

Sixty thousand forty-seven ICU patients.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

We used a previously validated mortality risk model to calculate the SMRs. We investigated the impact of PODs on the overall SMR, individual ICU SMR and ICU benchmarking. Of the 60,047 patients admitted to the ICUs, 514 (0.9%) were PODs, and 477 (93%) of them died. POD deaths accounted for 7% of the total 6738 in-hospital deaths. POD admission rates varied from 0.5 to 18.3 per 1000 admissions across ICUs. The risk prediction model predicted a 39% in-hospital mortality for PODs, but the observed mortality was 93%. The ratio of the SMR of the cohort without PODs to the SMR of the cohort with PODs was 0.96 (95% CI, 0.93-0.99). Benchmarking results changed in 70% of ICUs after excluding PODs.

CONCLUSIONS

Despite their relatively small overall number, PODs make up a large proportion of ICU patients who die. PODs cause bias in SMRs and in ICU benchmarking. We suggest excluding PODs when benchmarking ICUs with SMR.

Variation in Severity-Adjusted Resource use and Outcome for Neurosurgical Emergencies in the Intensive Care Unit

BACKGROUND

The correlation between the standardized resource use ratio (SRUR) and standardized hospital mortality ratio (SMR) for neurosurgical emergencies is not known. We studied SRUR and SMR and the factors affecting these in patients with traumatic brain injury (TBI), nontraumatic intracerebral hemorrhage (ICH), and subarachnoid hemorrhage (SAH).

METHODS

We extracted data of patients treated in six university hospitals in three countries (2015-2017). Resource use was measured as SRUR based on purchasing power parity-adjusted direct costs and either intensive care unit (ICU) length of stay (costSRURlength of stay) or daily Therapeutic Intervention Scoring System scores (costSRURTherapeutic Intervention Scoring System). Five a priori defined variables reflecting differences in structure and organization between the ICUs were used as explanatory variables in bivariable models, separately for the included neurosurgical diseases.

RESULTS

Out of 28,363 emergency patients treated in six ICUs, 6,162 patients (22%) were admitted with a neurosurgical emergency (41% nontraumatic ICH, 23% SAH, 13% multitrauma TBI, and 23% isolated TBI). The mean costs for neurosurgical admissions were higher than for nonneurosurgical admissions, and the neurosurgical admissions corresponded to 23.6-26.0% of all direct costs related to ICU emergency admissions. A higher physician-to-bed ratio was associated with lower SMRs in the nonneurosurgical admissions but not in the neurosurgical admissions. In patients with nontraumatic ICH, lower costSRURs were associated with higher SMRs. In the bivariable models, independent organization of an ICU was associated with lower costSRURs in patients with nontraumatic ICH and isolated/multitrauma TBI but with higher SMRs in patients with nontraumatic ICH. A higher physician-to-bed ratio was associated with higher costSRURs for patients with SAH. Larger units had higher SMRs for patients with nontraumatic ICH and isolated TBI. None of the ICU-related factors were associated with costSRURs in nonneurosurgical emergency admissions.

CONCLUSIONS

Neurosurgical emergencies constitute a major proportion of all emergency ICU admissions. A lower SRUR was associated with higher SMR in patients with nontraumatic ICH but not for the other diagnoses. Different organizational and structural factors seemed to affect resource use for the neurosurgical patients compared with nonneurosurgical patients. This emphasizes the importance of case-mix adjustment when benchmarking resource use and outcomes.

Modelling of intensive care unit (ICU) length of stay as a quality measure: a problematic exercise

BACKGROUND

Intensive care unit (ICU) length of stay (LOS) and the risk adjusted equivalent (RALOS) have been used as quality metrics. The latter measures entail either ratio or difference formulations or ICU random effects (RE), which have not been previously compared.

METHODS

From calendar year 2016 data of an adult ICU registry-database (Australia & New Zealand Intensive Care Society (ANZICS) CORE), LOS predictive models were established using linear (LMM) and generalised linear (GLMM) mixed models. Model fixed effects quality-metric formulations were estimated as RALOSR for LMM (geometric mean derived from log(ICU LOS)) and GLMM (day) and observed minus expected ICU LOS (OMELOS from GLMM). Metric confidence intervals (95%CI) were estimated by bootstrapping; random effects (RE) were predicted for LMM and GLMM. Forest-plot displays of ranked quality-metric point-estimates (95%CI) were generated for ICU hospital classifications (metropolitan, private, rural/regional, and tertiary). Robust rank confidence sets (point estimate and 95%CI), both marginal (pertaining to a singular ICU) and simultaneous (pertaining to all ICU differences), were established.

RESULTS

The ICU cohort was of 94,361 patients from 125 ICUs (metropolitan 16.9%, private 32.8%, rural/regional 6.4%, tertiary 43.8%). Age (mean, SD) was 61.7 (17.5) years; 58.3% were male; APACHE III severity-of-illness score 54.6 (25.7); ICU annual patient volume 1192 (702) and ICU LOS 3.2 (4.9). There was no concordance of ICU ranked model predictions, GLMM versus LMM, nor for the quality metrics used, RALOSR, OMELOS and site-specific RE for each of the ICU hospital classifications. Furthermore, there was no concordance between ICU ranking confidence sets, marginal and simultaneous for models or quality metrics.

CONCLUSIONS

Inference regarding adjusted ICU LOS was dependent upon the statistical estimator and the quality index used to quantify any LOS differences across ICUs. That is, there was no "one best model"; thus, ICU "performance" is determined by model choice and any rankings thereupon should be circumspect.