National Trends and Variation of Functional Status Deterioration in the Medically Critically Ill

Development and validation of early prediction models for new-onset functional impairment in patients after being transferred from the ICU

A significant number of intensive care unit (ICU) survivors experience new-onset functional impairments that impede their activities of daily living (ADL). Currently, no effective assessment tools are available to identify these high-risk patients. This study aims to develop an interpretable machine learning (ML) model for predicting the onset of functional impairment in critically ill patients. Data for this study were sourced from a comprehensive hospital in China, focusing on adult patients admitted to the ICU from August 2022 to August 2023 without prior functional impairments. A least absolute shrinkage and selection operator (LASSO) model was utilized to select predictors for inclusion in the model. Four models, logistic regression, support vector machine (SVM), random forest (RF), and extreme gradient boosting (XGBoost), were constructed and validated. Model performance was assessed using the area under the curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV). Additionally, the DALEX package was employed to enhance the interpretability of the final models. The study ultimately included 1,380 patients, with 684 (49.6%) exhibiting new-onset functional impairment on the seventh day after leaving the ICU. Among the four models evaluated, the SVM model demonstrated the best performance, with an AUC of 0.909, accuracy of 0.838, sensitivity of 0.902, specificity of 0.772, PPV of 0.802, and NPV of 0.886. ML models are reliable tools for predicting new-onset functional impairments in critically ill patients. Notably, the SVM model emerged as the most effective, enabling early identification of patients at high risk and facilitating the implementation of timely interventions to improve ADL.

Predictors of Functional Decline Among Critically Ill Surgical Patients: A National Analysis

INTRODUCTION

Functional decline is associated with critical illness, though this relationship in surgical patients is unclear. This study aims to characterize functional decline after intensive care unit (ICU) admission among surgical patients.

METHODS

We performed a retrospective analysis of surgical patients admitted to the ICU in the Cerner Acute Physiology and Chronic Health Evaluation database, which includes 236 hospitals, from 2007 to 2017. Patients with and without functional decline were compared. Predictors of decline were modeled.

RESULTS

A total of 52,838 patients were included; 19,310 (36.5%) experienced a functional decline. Median ages of the decline and nondecline groups were 69 (interquartile range 59-78) and 63 (interquartile range 52-72) years, respectively (P < 0.01). The nondecline group had a larger proportion of males (59.1% versus 55.3% in the decline group, P < 0.01). After controlling for sociodemographic covariates, comorbidities, and disease severity upon ICU admission, patients undergoing pulmonary (odds ratio [OR] 6.54, 95% confidence interval [CI] 2.67-16.02), musculoskeletal (OR 4.13, CI 3.51-4.87), neurological (OR 2.67, CI 2.39-2.98), gastrointestinal (OR 1.61, CI 1.38-1.88), and skin and soft tissue (OR 1.35, CI 1.08-1.68) compared to cardiovascular surgeries had increased odds of decline.

CONCLUSIONS

More than one in three critically ill surgical patients experienced a functional decline. Pulmonary, musculoskeletal, and neurological procedures conferred the greatest risk. Additional resources should be targeted toward the rehabilitation of these patients.

Impact of sepsis on Eastern Cooperative Oncology Group performance status among fully ambulatory patients: a prospective nationwide multicenter cohort

Background

Investigations of the impact of sepsis on the Eastern Cooperative Oncology Group performance status (ECOG PS) of fully ambulatory patients are scarce.

Methods

This is a retrospective analysis of prospectively collected nationwide data on septic patients recruited from 19 hospitals of the Korean Sepsis Alliance between August 2019 and December 2020. Adult septic patients with good ECOG PS (i.e., 0 or 1) before sepsis were enrolled in this study. The change in ECOG PS and the prevalence of disability (ECOG PS ≥2) at hospital discharge were recorded.

Results

Of the 4,145 septic patients, 1,735 (41.9%) patients who had ECOG PS of 0 or 1 before sepsis and eventually survived to discharge were selected. After treatment for sepsis, the ECOG PS deteriorated in 514 (29.6%) patients; 376 (21.7%) patients had poor ECOG PS (i.e., ≥2) at hospital discharge. The proportion of patients with poor ECOG PS at hospital discharge increased with increases in the initial sequential organ failure assessment (SOFA) score and lactate level. Furthermore, poor ECOG PS at hospital discharge was found in young patients (aged <65 years, 17.4%), those with no history of cancer (18.2%) or with low comorbidities [Charlson comorbidity index (CCI) ≤2; 13.6%], and those without septic shock (19.9%). In multivariable analysis, age, solid cancer, immunocompromised condition, SOFA score, mechanical ventilation, and use of inappropriate empirical antibiotics (odds ratio: 1.786; 95% confidence interval: 1.151-2.771) were significant risk factors for poor ECOG PS.

Conclusions

One in five septic patients who were fully ambulatory before sepsis were not functionally independent at hospital discharge. Incomplete functional recovery was also seen in a substantial proportion of younger patients, those with low comorbidities, and those without septic shock. However, the adequacy of empirical antibiotics may improve the functional status in such patients.

Predictors of discharge disposition and mortality following hospitalization with SARS-CoV-2 infection

IMPORTANCE

The SARS-CoV-2 pandemic has overwhelmed hospital capacity, prioritizing the need to understand factors associated with type of discharge disposition.

OBJECTIVE

Characterization of disposition associated factors following SARS-CoV-2.

DESIGN

Retrospective study of SARS-CoV-2 positive patients from March 7th, 2020, to May 4th, 2022, requiring hospitalization.

SETTING

Midwest academic health-system.

PARTICIPANTS

Patients above the age 18 years admitted with PCR + SARS-CoV-2.

INTERVENTION

None.

MAIN OUTCOMES

Discharge to home versus PAC (inpatient rehabilitation facility (IRF), skilled-nursing facility (SNF), long-term acute care (LTACH)), or died/hospice while hospitalized (DH).

RESULTS

We identified 62,279 SARS-CoV-2 PCR+ patients; 6,248 required hospitalizations, of whom 4611(73.8%) were discharged home, 985 (15.8%) to PAC and 652 (10.4%) died in hospital (DH). Patients discharged to PAC had a higher median age (75.7 years, IQR: 65.6-85.1) compared to those discharged home (57.0 years, IQR: 38.2-69.9), and had longer mean length of stay (LOS) 14.7 days, SD: 14.0) compared to discharge home (5.8 days, SD: 5.9). Older age (RRR:1.04, 95% CI:1.041-1.055), and higher Elixhauser comorbidity index [EI] (RRR:1.19, 95% CI:1.168-1.218) were associated with higher rate of discharge to PAC versus home. Older age (RRR:1.069, 95% CI:1.060-1.077) and higher EI (RRR:1.09, 95% CI:1.071-1.126) were associated with more frequent DH versus home. Blacks, Asians, and Hispanics were less likely to be discharged to PAC (RRR, 0.64 CI 0.47-0.88), (RRR 0.48 CI 0.34-0.67) and (RRR 0.586 CI 0.352-0.975). Having alpha variant was associated with less frequent PAC discharge versus home (RRR 0.589 CI 0.444-780). The relative risks for DH were lower with a higher platelet count 0.998 (CI 0.99-0.99) and albumin levels 0.342 (CI 0.26-0.45), and higher with increased CRP (RRR 1.006 CI 1.004-1.007) and D-Dimer (RRR 1.070 CI 1.039-1.101). Increased albumin had lower risk to PAC discharge (RRR 0.630 CI 0.497-0.798. An increase in D-Dimer (RRR1.033 CI 1.002-1.064) and CRP (RRR1.002 CI1.001-1.004) was associated with higher risk of PAC discharge. A breakthrough (BT) infection was associated with lower likelihood of DH and PAC.

CONCLUSION

Older age, higher EI, CRP and D-Dimer are associated with PAC and DH discharges following hospitalization with COVID-19 infection. BT infection reduces the likelihood of being discharged to PAC and DH.

Potential Interaction Between Sepsis and Acute Respiratory Distress Syndrome and Effect on the 6-Month Clinical Outcomes: A Preliminary Secondary Analysis of a Prospective Observational Study

BACKGROUND

This study aimed to investigate the effect of the potential interaction between sepsis and acute respiratory distress syndrome (ARDS) on the 6-month clinical outcomes.

METHODS

This secondary analysis of a prospective multicenter observational study included patients who were expected to receive mechanical ventilation for more than 48 h. Patients were stratified based on the incidence of sepsis and further subdivided according to the presence of ARDS. The primary endpoints for patients whose follow-up information was available included mortality (n = 162) and the occurrence of PICS (n = 96) at six months. The diagnosis of PICS was based on any of the following criteria: (1) decrease ≥ 10 points in the physical component score of the 36-item Short Form (SF36) questionnaire; (2) decrease ≥ 10 points in the mental component score of the SF-36; or (3) decline in the Short Memory Questionnaire (SMQ) score and SMQ score < 40 at six months after ICU admission. We conducted multivariate logistic regression analyses to assess the effect of the potential interaction between ARDS and sepsis on the 6-month clinical outcomes.

RESULTS

The mortality in the ARDS sub-group was higher than that in the non-ARDS subgroup [47% (7/15) versus 21% (18/85)] in the non-sepsis group. However, the mortality in the ARDS and non-ARDS subgroups was similar in the sepsis group. Multivariate logistic regression analyses revealed that ARDS was significantly associated with mortality in the non-sepsis group (adjusted OR: 5.25; 95% CI: 1.45-19.09; p = .012), but not in the sepsis group (P for interaction = .087). Multivariate logistic regression analyses showed ARDS was not associated with PICS occurrence in the non-sepsis and sepsis groups (P-value for the interaction = .039).

CONCLUSIONS

This hypothesis-generating study suggested that the effect of ARDS on the 6-month outcomes depended on the presence or absence of sepsis.

TRIAL REGISTRATION

Not applicable.

Development and validation of early prediction models for new-onset functional impairment at hospital discharge of ICU admission

PURPOSE

We aimed to develop and validate models for predicting new-onset functional impairment after intensive care unit (ICU) admission with predictors routinely collected within 2 days of admission.

METHODS

In this multi-center retrospective cohort study of acute care hospitals in Japan, we identified adult patients who were admitted to the ICU with independent activities of daily living before hospitalization and survived for at least 2 days from April 2014 to October 2020. The primary outcome was functional impairment defined as Barthel Index ≤ 60 at hospital discharge. In the internal validation dataset (April 2014 to March 2019), using routinely collected 94 candidate predictors within 2 days of ICU admission, we trained and tuned the six conventional and machine-learning models with repeated random sub-sampling cross-validation. We computed the variable importance of each predictor to the models. In the temporal validation dataset (April 2019 to October 2020), we measured the performance of these models.

RESULTS

We identified 19,846 eligible patients. Functional impairment at discharge was developed in 33% of patients (n = 6488/19,846). In the temporal validation dataset, all six models showed good discrimination ability with areas under the curve above 0.86, and the differences among the six models were negligible. Variable importance revealed newly detected early predictors, including worsened neurologic conditions and catabolism biomarkers such as decreased serum albumin and increased blood urea nitrogen.

CONCLUSION

We successfully developed early prediction models of new-onset functional impairment after ICU admission that achieved high performance using only data routinely collected within 2 days of ICU admission.

Predictors of Post-Acute Sequelae of COVID-19 Development and Rehabilitation: A Retrospective Study

Objective

To examine the frequency of postacute sequelae of SARS-CoV-2 (PASC) and the factors associated with rehabilitation utilization in a large adult population with PASC.

Design

Retrospective study.

Setting

Midwest hospital health system.

Participants

19,792 patients with COVID-19 from March 10, 2020, to January 17, 2021.

Intervention

Not applicable.

Main Outcome Measures

Descriptive analyses were conducted across the entire cohort along with an adult subgroup analysis. A logistic regression was performed to assess factors associated with PASC development and rehabilitation utilization.

Results

In an analysis of 19,792 patients, the frequency of PASC was 42.8% in the adult population. Patients with PASC compared with those without had a higher utilization of rehabilitation services (8.6% vs 3.8%, P<.001). Risk factors for rehabilitation utilization in patients with PASC included younger age (odds ratio [OR], 0.99; 95% confidence interval [CI], 0.98-1.00; P=.01). In addition to several comorbidities and demographics factors, risk factors for rehabilitation utilization solely in the inpatient population included male sex (OR, 1.24; 95% CI, 1.02-1.50; P=.03) with patients on angiotensin-converting-enzyme inhibitors or angiotensin-receptor blockers 3 months prior to COVID-19 infections having a decreased risk of needing rehabilitation (OR, 0.80; 95% CI, 0.64-0.99; P=.04).

Conclusions

Patients with PASC had higher rehabilitation utilization. We identified several clinical and demographic factors associated with the development of PASC and rehabilitation utilization.

Key Data Elements for Longitudinal Tracking of Physical Function: A Modified Delphi Consensus Study

OBJECTIVE

Physical function is associated with important outcomes, yet there is often a lack of continuity in routine assessment. The purpose of this study was to determine data elements and instruments for longitudinal measurement of physical function in routine care among patients transitioning from acute care hospital setting to home with home health care.

METHODS

A 4-round modified Delphi process was conducted with 13 participants with expertise in physical therapy, health care administration, health services research, physiatry/medicine, and health informatics. Three anonymous rounds identified important and feasible data elements. A fourth in-person round finalized the recommended list of individual data elements. Next, 2 focus groups independently provided additional perspectives from other stakeholders.

RESULTS

Response rates were 100% for online rounds 1, 3, and 4 and 92% for round 2. In round 1, 9 domains were identified: physical function, participation, adverse events, behavioral/emotional health, social support, cognition, complexity of illness/disease burden, health care utilization, and demographics. Following the fourth round, 27 individual data elements were recommended. Of these, 20 (74%) are "administrative" and available from most hospital electronic medical records. Additional focus groups confirmed these selections and provided input on standardizing collection methods. A website has been developed to share these results and invite other health care systems to participate in future data sharing of these identified data elements.

CONCLUSION

A modified Delphi consensus process was used to identify critical data elements to track changes in patient physical function in routine care as they transition from acute hospital to home with home health.

IMPACT

Expert consensus on comprehensive and feasible measurement of physical function in routine care provides health care professionals and institutions with guidance in establishing discrete medical records data that can improve patient care, discharge decisions, and future research.

Recent Trends in Admission Diagnosis and Related Mortality in the Medically Critically Ill

PURPOSE

With decades of declining ICU mortality, we hypothesized that the outcomes and distribution of diseases cared for in the ICU have changed and we aimed to further characterize them.

STUDY DESIGN AND METHODS

A retrospective cohort analysis of 287,154 nonsurgical-critically ill adults, from 237 U.S. ICUs, using the manually abstracted Cerner APACHE Outcomes database from 2008 to 2016 was performed. Surgical patients, rare admission diagnoses (<100 occurrences), and low volume hospitals (<100 total admissions) were excluded. Diagnoses were distributed into mutually exclusive organ system/disease-based categories based on admission diagnosis. Multi-level mixed-effects negative binomial regression was used to assess temporal trends in admission, in-hospital mortality, and length of stay (LOS).

RESULTS

The number of ICU admissions remained unchanged (IRR 0.99, 0.98-1.003) while certain organ system/disease groups increased (toxicology [25%], hematologic/oncologic [55%] while others decreased (gastrointestinal [31%], pulmonary [24%]). Overall risk-adjusted in-hospital mortality was unchanged (IRR 0.98, 0.96-1.0004). Risk-adjusted ICU LOS (Estimate -0.06 days/year, -0.07 to -0.04) decreased. Risk-adjusted mortality varied significantly by disease.

CONCLUSION

Risk-adjusted ICU mortality rate did not change over the study period, but there was evidence of shifting disease burden across the critical care population. Our data provides useful information regarding future ICU personnel and resource needs.

Serum Levels of Acylcarnitines and Amino Acids Are Associated with Liberation from Organ Support in Patients with Septic Shock

Sepsis-induced metabolic dysfunction is associated with mortality, but the signatures that differentiate variable clinical outcomes among survivors are unknown. Our aim was to determine the relationship between host metabolism and chronic critical illness (CCI) in patients with septic shock. We analyzed metabolomics data from mechanically ventilated patients with vasopressor-dependent septic shock from the placebo arm of a recently completed clinical trial. Baseline serum metabolites were measured by liquid chromatography-mass spectrometry and ¹H-nuclear magnetic resonance. We conducted a time-to-event analysis censored at 28 days. Specifically, we determined the relationship between metabolites and time to extubation and freedom from vasopressors using a competing risk survival model, with death as a competing risk. We also compared metabolite concentrations between CCI patients, defined as intensive care unit level of care ≥ 14 days, and those with rapid recovery. Elevations in two acylcarnitines and four amino acids were related to the freedom from organ support (subdistributional hazard ratio < 1 and false discovery rate < 0.05). Proline, glycine, glutamine, and methionine were also elevated in patients who developed CCI. Our work highlights the need for further testing of metabolomics to identify patients at risk of CCI and to elucidate potential mechanisms that contribute to its etiology.

Improved Prediction of Older Adult Discharge After Trauma Using a Novel Machine Learning Paradigm

BACKGROUND

The ability to reliably predict outcomes after trauma in older adults (age ≥ 65 y) is critical for clinical decision making. Using novel machine-learning techniques, we sought to design a nonlinear, competing risks paradigm for prediction of older adult discharge disposition following injury.

MATERIALS AND METHODS

The National Trauma Databank (NTDB) was used to identify patients 65+ y between 2007 and 2014. Training was performed on an enriched cohort of diverse patients. Factors included age, comorbidities, length of stay, and physiologic parameters to predict in-hospital mortality and discharge disposition (home versus skilled nursing/long-term care facility). Length of stay and discharge status were analyzed via competing risks survival analysis with Bayesian additive regression trees and a multinomial mixed model.

RESULTS

The resulting sample size was 47,037 patients. Admission GCS and age were important in predicting mortality and discharge disposition. As GCS decreased, patients were more likely to die (risk ratio increased by average of 1.4 per 2-point drop in GCS, P < 0.001). As GCS decreased, patients were also more likely to be discharged to a skilled nursing or long-term care facility (risk ratio decreased by 0.08 per 2-point decrease in GCS, P< 0.001). The area under curve for prediction of discharge home was improved in the competing risks model 0.73 versus 0.43 in the traditional multinomial mixed model.

CONCLUSIONS

Predicting older adult discharge disposition after trauma is improved using machine learning over traditional regression analysis. We confirmed that a nonlinear, competing risks paradigm enhances prediction on any given hospital day post injury.

Morbidity and Mortality Trends of Pancreatitis: An Observational Study

Background: Pancreatitis accounts for more than $2.5 billion of healthcare costs and remains the most common gastrointestinal (GI) admission. Few contemporary studies have assessed temporal trends of incidence, complications, management, and outcomes for acute pancreatitis in hospitalized patients at the national level. Methods: We used data from one of the largest hospital-based databases available in the United States, the Healthcare Cost and Utilization Project's (HCUP) State Inpatient Database, from 10 states between 2008 and 2015. We included patients with a diagnosis of acute pancreatitis (ICD-9 CM 577.0). Patient- and hospital-level data were used to estimate incidence and inpatient mortality rates. Results: From 80,736,256 hospitalizations, 929,914 (1.15%) cases of acute pancreatitis were identified, 186,226 (20.2%) of which were caused by gallbladder disease). The median age was 53 years (interquartile range [IQR], 41-67) and 50.8% were men. In-hospital mortality was 2.5% and crude mortality rates declined from 2.9% to 2.0% over the study period. Admission year remained significant after adjusting for patient demographics and comorbidities (odds ratio [OR], 0.90; 95% confidence interval [CI], 0.89-0.90; p < 0.001). Gallbladder disease was associated with decreased odds of mortality (OR, 0.60; 95% CI, 0.57-0.62). Median length of stay was four days (IQR, 2-7) and decreased over time. The rates of surgical and endoscopic interventions were highest in 2011 (peak incidence of 16.1% and 9.5%, respectively) and have been decreasing since. Surgical providers were, on average, more likely than medical providers to perform surgery in both those with and without gallbladder disease etiology (gallbladder disease OR, 7.11; 95% CI, 5.46-9.25; non-gallbladder disease OR, 20.50; 95% CI, 16.81-25.01), endoscopy (gallbladder disease OR, 1.22; 95% CI, 0.87-1.72; non-gallbladder disease OR, 1.60; 95% CI, 1.18-2.16), or both (gallbladder disease OR, 7.00; 95% CI, 5.22-9.37; non-gallbladder disease OR, 8.85; 95% CI, 5.61-13.96). Conclusions: The incidence of pancreatitis, from 2008 to 2015, has increased whereas inpatient mortality (i.e., case fatality) has decreased. Understanding temporal trends in outcomes and management along with provider, hospital, and regional variation can better identify areas for future research and collaboration in managing these patients.

Discordant Cardiopulmonary Resuscitation and Code Status at Death

CONTEXT

One fundamental way to honor patient autonomy is to establish and enact their wishes for end-of-life care. Limited research exists regarding adherence with code status.

OBJECTIVES

This study aimed to characterize cardiopulmonary resuscitation (CPR) attempts discordant with documented code status at the time of death in the U.S. and to elucidate potential contributing factors.

METHODS

The Cerner Acute Physiology and Chronic Health Evaluation (APACHE) outcomes database, which includes 237 U.S. hospitals that collect manually abstracted data from all critical care patients, was queried for adults admitted to intensive care units with a documented code status at the time of death from January 2008 to December 2016. The primary outcome was discordant CPR at death. Multivariable logistic regression models were used to identify patient-level and hospital-level associated factors after adjustment for age, hospital, and illness severity (APACHE III score).

RESULTS

A total of 21,537 patients from 56 hospitals were included. Of patients with a do-not-resuscitate code status, 149 (0.8%) received CPR at death, and associated factors included black race, higher APACHE III score, or treatment in small or nonteaching hospitals. Of patients with a full code status, 203 (9.0%) did not receive CPR at death, and associated factors included higher APACHE III score, primary neurologic or trauma diagnosis, or admission in a more recent year.

CONCLUSION

At the time of death, 1.6% of patients received or did not undergo CPR in a manner discordant with their documented code statuses. Race and institutional factors were associated with discordant resuscitation, and addressing these disparities may promote concordant end-of-life care in all patients.