Forecasting disease trajectories in critical illness: comparison of probabilistic dynamic systems to static models to predict patient status in the intensive care unit.

OBJECTIVE: Conventional prediction models fail to integrate the constantly evolving nature of critical illness. Alternative modelling approaches to study dynamic changes in critical illness progression are needed. We compare static risk prediction models to dynamic probabilistic models in early critical illness. DESIGN: We developed models to simulate disease trajectories of critically ill COVID-19 patients across different disease states. Eighty per cent of cases were randomly assigned to a training and 20% of the cases were used as a validation cohort. Conventional risk prediction models were developed to analyse different disease states for critically ill patients for the first 7 days of intensive care unit (ICU) stay. Daily disease state transitions were modelled using a series of multivariable, multinomial logistic regression models. A probabilistic dynamic systems modelling approach was used to predict disease trajectory over the first 7 days of an ICU admission. Forecast accuracy was assessed and simulated patient clinical trajectories were developed through our algorithm. SETTING AND PARTICIPANTS: We retrospectively studied patients admitted to a Cleveland Clinic Healthcare System in Ohio, for the treatment of COVID-19 from March 2020 to December 2022. RESULTS: 5241 patients were included in the analysis. For ICU days 2-7, the static (conventional) modelling approach, the accuracy of the models steadily decreased as a function of time, with area under the curve (AUC) for each health state below 0.8. But the dynamic forecasting approach improved its ability to predict as a function of time. AUC for the dynamic forecasting approach were all above 0.90 for ICU days 4-7 for all states. CONCLUSION: We demonstrated that modelling critical care outcomes as a dynamic system improved the forecasting accuracy of the disease state. Our model accurately identified different disease conditions and trajectories, with a <10% misclassification rate over the first week of critical illness.

Functional Outcomes, Goals, and Goal Attainment among Chronically Critically Ill Long-Term Acute Care Hospital Patients.

Rationale: Chronically critically ill patients admitted to a long-term acute care hospital (LTACH) setting are a vulnerable population of intensive care unit (ICU) survivors. Little is known of the goals and functional outcomes achieved by patients after rehabilitation in the LTACH setting. Objectives: We sought to examine patient goals and functional outcomes, including swallowing function, among ICU survivors admitted to an LTACH with a tracheostomy. Methods: This was a prospective observational cohort study of chronic critically ill LTACH patients. Results: Fifty older subjects with a median duration of intubation before tracheostomy of 13 days were enrolled. ICU-acquired weakness and cognitive impairment were present in 40 (80%) and 36 (72%) patients, as measured by the Medical Research Council scale and Montreal Cognitive Assessment, respectively. Mental health problems were also common, with 16 (32%) patients experiencing moderate to severe anxiety, 9 (18%) experiencing moderate to severe depression, and 11 (22%) reporting symptoms consistent with post-traumatic stress disorder, according to the Hospital Anxiety and Depression Scale and Post-Traumatic Stress Syndrome 10-Questions Inventory, respectively. Pharyngeal dysfunction, as measured by Fiberoptic Endoscopic Evaluation of Swallow examination, was present in 37 (74%) patients. Patient goals, in decreasing order of frequency, included eating and drinking, speaking, walking, returning home, and toileting. By LTACH discharge, goal attainment was variable, with 97% of those who ranked speaking as important being able to speak and 88% of those who ranked eating and drinking as important being able to eat and drink but with only 21% being able to walk and only 18% being able to self-toilet. Discharge to the home or acute rehabilitation setting, achieved in 52% of the population, was associated with greater strength, as measured by the total Medical Research Council score (P = 0.002) as well as by the European 5-dimension, 5-level health-related quality of life instrument domains of mobility (P = 0.008) and self-care (P = 0.04). Conclusions: Goal attainment during this period of recovery was variable. The ability to speak, eat, and drink, frequently identified as goals by these patients, were achieved, whereas functional goals such as walking were rarely achieved. These findings highlight the importance of identifying patient goals and setting realistic expectations informed by functional assessments when rehabilitating this vulnerable patient population in the LTACH and subsequent post-acute care settings.

Clinicians' Perceptions of Behavioral Economic Strategies to Increase the Use of Lung-Protective Ventilation.

Rationale: Lung-protective ventilation (LPV) improves outcomes in patients with acute respiratory distress syndrome (ARDS) and has also shown benefits in patients without ARDS. Despite this evidence, LPV use remains low.Objectives: To understand clinicians' perceptions of using behavioral economic strategies to improve rates of LPV use.Methods: We conducted semistructured interviews of clinicians across seven intensive care units within a university health system. We purposefully sampled clinicians of different professional backgrounds and experience levels. Each interview included descriptions of three of five strategies grounded in behavioral economic theory designed to facilitate clinicians' use of LPV: 1) an order set autopopulated with LPV settings ("default"), 2) an order set providing a choice between autopopulated LPV settings and open-ended order entry for alternative settings ("active choice"), 3) requirement of written justification if settings other than LPV were ordered or documented ("accountable justification"), 4) automated ARDS identification and clinician prompting ("alert"), and 5) provision of clinicians' and their peers' individual rates of LPV use ("peer comparison"). Descriptions were followed by open-ended questions to elicit perceptions about advantages, disadvantages, and acceptability. Initial interview transcripts were reviewed by two investigators to develop a thematic codebook, which was refined iteratively with the use of constant comparative methods.Results: We completed 17 interviews of physicians, nurse practitioners, and respiratory therapists. Strategies that prepopulated settings (default, active choice, and accountable justification) were perceived as providing benefit by reducing workloads and serving as cognitive prompts. The default and active choice strategies were more acceptable than accountable justification, which was perceived as potentially frustrating due to workflow impedance. The alert strategy was met with concerns about alert accuracy and alarm fatigue. The peer comparison strategy led to concerns about timing and fear of punitive measures. Participants believed that the default and active choice strategies would be highly acceptable, whereas few interviewees thought the alert would be acceptable. The active choice strategy was most consistently identified as potentially highly effective.Conclusions: Behavioral economic strategies have great potential as acceptable and potentially effective strategies to increase the use of LPV.

Radiographical findings in patients with liver cirrhosis and hepatic encephalopathy.

BACKGROUND AND AIMS: Hepatic encephalopathy is a common complication encountered in patients with liver cirrhosis. Hepatic encephalopathy is not reflected in the current liver transplant allocation system. Correlation was sought between hepatic encephalopathy with findings detected on radiographic imaging studies and the patient's clinical profile. METHODS: A retrospective analysis was conducted of patients with cirrhosis, who presented for liver transplant evaluation in 2009 and 2010. Patients with hepatocellular carcinoma, ejection fraction less than 60% and who had a TIPS (transjugular intrahepatic portosystemic shunting) procedure or who did not complete the evaluation were excluded. Statistical analysis was performed and variables found to be significant on univariate analysis (P < 0.05) were analysed by a multivariate logistic regression model. RESULTS: A total of 117 patients met the inclusion criteria and were divided into a hepatic encephalopathy group (n = 58) and a control group (n = 59). Univariate analysis found that a smaller portal vein diameter, smaller liver antero-posterior diameter, liver nodularity and use of diuretics or centrally acting medications showed significant correlation with hepatic encephalopathy. This association was confirmed for smaller portal vein, use of diuretics and centrally acting medications in the multivariate analysis. CONCLUSION: A decrease in portal vein diameter was associated with increased risk of encephalopathy. Identifying patients with smaller portal vein diameter may warrant screening for encephalopathy by more advanced psychometric testing, and more aggressive control of constipation and other factors that may precipitate encephalopathy.

Biliary stenting in patients with malignant biliary obstruction: comparison of double layer, plastic and metal stents.

BACKGROUND AND AIM: The double layer stent (DLS) has a unique design and has been used for palliation of malignant biliary obstruction, but literature on this stent is limited. Our aim was to compare plastic (PS), DLS and metal stents (MS) in terms of complication rates, time to occlusion, and patency rate in patients with malignant biliary obstruction (MBO). METHODS: A retrospective review of stents placed for MBO at our institution in the period between January 2009 and April 2011 was conducted. A total of 114 stents were identified, of which 44 were MS (39 %), 37 DLS (32 %), and 33 PS (29 %). A stent was considered occluded when an unplanned stent removal or intervention occurred due to clinical suspicion of biliary obstruction. RESULTS: Stents remained patent for 95 days (range 7-359 days) in the DLS group and 59 days (range 7-228 days) in the PS group (P = 0.014) and 128.7 days (range 4-602 days) in the metal stent group. Twenty-seven percent (n = 9) of PS occluded after a mean of 60 days while 16 % (n = 7) of MS occluded after a mean of 87 days and 5 % (n = 2) of DLS occluded after a mean of 85 days (DLS vs. PS P = 0.012, DLS vs. MS P = 0.13, MS vs. PS P = 0.22). CONCLUSIONS: DLS are superior to PS in patients with MBO and appear to be comparable to MS. MS had a longer patency rate but were comparable to DLS in early and late complications. We speculate that the less expensive DLS may be a cost effective alternative in the palliation of MBO.