Assessment Of Illness Severity Research Articles

Decreased cardio-respiratory information transfer is associated with deterioration and a poor prognosis in critically ill patients with sepsis.

Assessing illness severity in the ICU is crucial for early prediction of deterioration and prognosis. Traditional prognostic scores often treat organ systems separately, overlooking the body's interconnected nature. Network physiology offers a new approach to understanding these complex interactions. This study used the concept of transfer entropy (TE) to measure information flow between heart rate (HR), respiratory rate (RR), and capillary oxygen saturation (SpO2) in critically ill sepsis patients, hypothesizing that TE between these signals would correlate with disease outcome. The retrospective cohort study utilized the MIMIC III Clinical Database, including patients who met Sepsis-3 criteria on admission and had 30 minutes of continuous HR, RR, and SpO2 data. TE between the signals was calculated to create physiological network maps. Cox regression assessed the 48 relationship between cardiorespiratory network indices and both deterioration (SOFA score increase of ≥2 points at 48 hours) and 30-day mortality. Among 164 patients, higher information flow from SpO2 to HR [TE(SpO2→HR)] and reciprocal flow between HR and RR [TE(RR→HR) and TE(HR→RR)] were linked to reduced mortality, independent of age, mechanical ventilation, SOFA score, and comorbidity. Reductions in TE(HR → RR), TE(RR→HR), TE(SpO2→RR), and TE(SpO2→HR) were associated with increased risk of 48-hour deterioration. After adjustment for potential confounders, only TE(HR→RR) and TE(RR→HR) remained statistically significant. The study confirmed that physiological network mapping using routine signals in sepsis patients could indicate illness severity and that higher TE values were generally associated with improved outcomes. XXXX XXXX.

Clinical Correlates of Antipsychotic Plasma Levels with Long-Acting Paliperidone: Corrélats cliniques des concentrations plasmiques de palipéridone à libération prolongée.

The majority of patients with schizophrenia experience dramatic improvement in psychotic symptoms when treated with antipsychotic medication. Maintenance treatment can prevent relapses but problems with medication adherence limit effectiveness. Long-acting injectable antipsychotics (LAIs) provide an opportunity to establish adherence but challenges remain in ensuring that the dose selected is therapeutic. Therapeutic drug monitoring has not been established as valuable for LAIs in the maintenance treatment of schizophrenia. This exploratory study was undertaken to describe plasma paliperidone levels in outpatients treated with the LAI paliperidone palmitate and to determine whether paliperidone levels are associated with subjective experience on medication and side effects. Twenty-one outpatients with schizophrenia receiving treatment with LAI paliperidone consented to participation in this study. Blood samples were obtained for measurement of paliperidone and prolactin levels at the first visit. A second paliperidone level was obtained at the time of the next injection for 18 of the participants. Clinical rating scales were administered at the first visit to assess illness severity, attitudes regarding medication, subjective well-being and side effects. Paliperidone levels were highly correlated at the two time points (ρ = .85; P < .001). Mean paliperidone level at the first visit was 34.9 ng/ml and ranged from 5.1 to 73.9 ng/ml. Higher paliperidone levels were correlated with higher prolactin levels (ρ = 0.59, P < .01) and lower sexual desire (ρ = -.58, P < .01). We demonstrated that paliperidone levels can be measured reliably in patients receiving LAI paliperidone. Higher plasma levels were associated with higher prolactin levels and reduced sexual desire but not with measures of subjective experience on medications or other side effects. Measurement of paliperidone levels in patients treated with paliperidone palmitate may have the potential to minimize the dose of medication prescribed and, in turn, the severity of sexual side effects.

Genomic characteristics, disease outcome and heterologous vaccine effectiveness among cases with SARS CoV-2 infection

BackgroundIn the pursuit of global health security, continuous monitoring of vaccine effectiveness across various viral strains emerges as a crucial imperative. The emergence of SARS-CoV-2 major variants of concern (VOCs), including Alpha, Beta, Delta, and Omicron, has added complexity to the COVID-19 vaccination landscape.ObjectivesTo assess illness severity, evaluate vaccine efficacy across varying doses and types, and determine effectiveness against major VOCs within the population.MethodsThis retrospective cohort study, conducted in Abu Dhabi, United Arab Emirates, focuses on a cohort of 44,073 SARS-CoV-2 positive cases from February 2021 to May 2022, dominated by the Delta and Omicron variants. The study employed a nested case-control design, analyzing hospital admissions for confirmed SARS-CoV-2 infection.ResultsVaccine effectiveness was higher among heterologus-boosted individuals at 87% (95% CI:79%-93%) compared to homologus-boosted individuals at 59% (95% CI: 48%-68%) and fully vaccinated, non-boosted adults at 53% (95% CI: 46%-59%). These findings highlight the importance of heterologous boosting, particularly against rapidly evolving viral variants, offering valuable insights for refining pandemic response strategies.ConclusionThe study underscores the critical need for ongoing assessment and adaptation of vaccination strategies to the evolving viral landscape.

Effectiveness of the Emergency Severity Index in Early Identification of Sepsis in a Tertiary Care Hospital of a Low-resource Setting

Background: Sepsis is life-threatening organ dysfunction, and a global issue, with incidence and mortality are around 49 M cases in 2017. Annual sepsis incidence from Pakistan is around 1600-2500 per 100, 000 cases with high mortality of approx. 30%. This huge mortality among sepsis and septic shock cases from low-income settings it is necessary to implement a focused and effective triage tool to prioritize and categorize the sick patients from huge bulk of sick patients who attend the ED and to identify those who need to be seen and manage accordingly. Emergency severity index (ESI), a widely acceptable tool that was developed to assess illness severity, hence able to identify the sick patients that need immediate attention at triage. Objectives: Our objective is to determine the effectiveness of ESI triage acuity in early recognition of patient with sepsis and septic shock at triage in a busy ED. Methods: This was a single center, cross sectional and observational study, we include 240 cases who present to the ED of a tertiary care hospital. We follow ESI v. 4.0. All cases presenting to the ED with suspected diagnosis of sepsis or septic shock were included. Data was collected on a predesigned data sheet. Participants were followed by their electronic medical record number through patient care information software where patient’s medical diagnosis and discharge diagnosis are present. Results: Total of 240 patients in the inclusion criteria, area under curve (AUC) for the ESI score I, for septic shock was 0.943 [0.921 - 0.964] with the optimal cutoff value of 2.0, sensitivity of 88.5% and specificity 100%. Similarly, the receiver operating curve (ROC) with area under the curve of ESI Score II for sepsis is found 0.2 [0.129-0.270] with sensitivity 23% and specificity was 12.7%. Conclusion: ESI is a reliable tool which is both sensitive and specific in screening patients with septic shock in a busy emergency department.

Recent onset mental illness severity: pilot study on the role of cognition, sensory modulation, and daily life participation.

Early detection of individuals at risk for onset of severe illness is crucial for prevention and early intervention, aiming to mitigate the long-term impact on both the individual and the community. While well-established models exist for predicting the onset and prolonged severity of illness, there is a gap in understanding illness-onset severity. This pilot study aimed to investigate premorbid objective and subjective dimensions of participation in daily life occupations, as well as sensory and cognitive functions as potential markers of the recent-onset mental illness severity. A total of 50 participants (men: N=26, 52%; women: N=24, 48%), aged 18-40 (M=26.2, SD=5.8) with recent-onset mental illness completed standard, well-established assessments of illness severity, cognitive biases and failures, neurocognitive status, participation in daily life, and sensory responsiveness thorough cross-sectional design. The differences between the groups of the illness severity were explored with descriptive statistics, followed by a Kruskal-Wallis test. Discriminant analysis was used suggesting a multi-varied model for the separation between the groups of illness severity. Three groups of illness severity exhibited differences in premorbid cognitive functions (F(2)=5.8, p<.01) and participation diversity (F(2)=3.8, p<.05). Combining these two indices explained 92% of the variance between the groups (Wilks' Λ = .68, χ2(4) = 17.7, p=.001), accurately classifying mild to marked illness severity (62.5-88.5%). The study contributes to revealing factors involved in the formation of more severe mental illness and suggesting possible avenues for early intervention and prevention. Cognitive biases and sensory modulation dysfunction may contribute to the illness formation. Still, the most effective markers of more severe mental illness onset are functional cognition and limited participation diversity. Since addressing these markers is a unique specialization within occupational therapy, the findings highlight the potential contribution theprofession can make to the early identification of the most vulnerable populations.

Exploring Ventilator-Associated Pneumonia: Microbial Clues and Biomarker Insights from a Retrospective Study.

Background and Objectives: Ventilator-associated pneumonia (VAP) is a common complication in critically ill patients receiving mechanical ventilation. The incidence rates of VAP vary, and it poses significant challenges due to microbial resistance and the potential for adverse outcomes. This study aims to explore the microbial profile of VAP and evaluate the utility of biomarkers and illness severity scores in predicting survival. Materials and Methods: A retrospective cohort study was conducted involving 130 patients diagnosed with VAP. Microbial analysis of bronchoalveolar lavage (BAL) fluid, as well as measurements of C-reactive protein (CRP) and procalcitonin (PCT) levels, were performed. Sequential Organ Failure Assessment (SOFA) and Acute Physiology and Chronic Health Evaluation II (APACHE II) scores were calculated to assess illness severity. Statistical analyses were conducted to determine correlations and associations. Results: The study revealed that Klebsiella pneumoniae (K. pneumoniae) (50.7%) and Pseudomonas aeruginosa (P. aeruginosa) (27.69%) were the most identified microorganisms in VAP cases. SOFA (p-value < 0.0001) and APACHE II (p-value < 0.0001) scores were effective in assessing the severity of illness and predicting mortality in VAP patients. Additionally, our investigation highlighted the prognostic potential of CRP levels (odds ratio [OR]: 0.980, 95% confidence interval [CI] 0.968 to 0.992, p = 0.001). Elevated levels of CRP were associated with reduced survival probabilities in VAP patients. Conclusion: This study highlights the microbial profile of VAP and the importance of biomarkers and illness severity scores in predicting survival. Conclusions: The findings emphasize the need for appropriate management strategies to combat microbial resistance and improve outcomes in VAP patients.

2203 Avoiding therapeutic nihilism in critically ill older adults: a single Centre approach to enhanced care in patients with frailty

Abstract Introduction Enhanced care units (ECU) are a novel concept, targeting the gap between ward-level and critical care. They care for patients requiring intensive medical or nursing care, who may not require, desire, or be suitable for, escalation above ward care (Society of Acute Medicine and Intensive Care Society, 2022). The ECU at Barnet Hospital opened in March 2022, and, because of the local population demographic, admits a high number of older patients living with frailty. We aimed to assess the performance of the ECU for this subset of patients. Methods A retrospective audit of electronic records of 75 randomly selected patients admitted to ECU between March and August 2023. Data were gathered on Clinical Frailty Score (CFS) at baseline, comorbidity, escalation status, APACHE II illness severity score, and outcome measures. Results The majority of patients in the sample, 52 of 75 (69.3%), were over 65 years of age with an average of 69.1 years. Baseline frailty score was high, with a modal CFS of 6. Of these patients, 32 (61.5%) had a DNACPR, and 17 (32.7%) had treatment ceiling at ECU level. Illness severity was similar across CFS groups, with a mean APACHE II score of 15.2 (representing a 25% mortality risk). Overall mortality in the over 65 s was 23.1% (12/52), without significant change when stratified by CFS. Mortality in the under 65 s was 8.7% (2/23). Conclusions Acutely unwell patients with frailty may benefit from ECU level care. In our centre, we found no significant increase in mortality linked to a higher frailty score. We suggest that this may represent good case selection by clinicians experienced in working with frailty: admitting patients with more reversibility and targeting therapies towards reversible causes. Limitations remain, especially in assessing illness severity, as the assessment tools are not targeted to this cohort.

Sepsis Huddles in the Neonatal Intensive Care Unit: A Retrospective Cohort Study of Late-onset Infection Recognition and Severity Assessment

To analyze relationships between provider-documented signs prompting sepsis evaluations, assessments of illness severity, and late-onset infection (LOI). Retrospective cohort study of all infants receiving a sepsis huddle in conjunction with a LOI evaluation. Participants were ≥3 days old and admitted to a level IV neonatal intensive care unit (NICU) from September 2018 through May 2021. Data were extracted from standardized sepsis huddle notes in the electronic health record, including clinical signs prompting LOI evaluations, illness severity assessments (from least to most severe: green, yellow, and red), and management plans. To analyze relationships of sepsis huddle characteristics with the detection of culture-confirmed LOI (bacteremia, urinary tract infection, or meningitis), we utilized diagnostic test statistics, area under the receiver-operator characteristic (AUROC) analyses, and multivariable logistic regression. We identified 1209 eligible sepsis huddles among 604 infants. There were 111 culture-confirmed LOI episodes (9% of all huddles). Twelve clinical signs of infection poorly distinguished infants with and without LOI, with sensitivity for each ranging from 2 to 36% and AUROC ranging 0.49-0.53. Multivariable logistic regression identified increasing odds of infection with higher perceived illness severity at the time of sepsis huddle, adjusted for gestational age and receipt of intensive care supports. Clinical signs prompting sepsis huddles were nonspecific and not predictive of concurrent LOI. Higher perceived illness severity was associated with presence of infection, despite some misclassification based on objective criteria. In level IV NICUs, antimicrobial stewardship through development of criteria for antibiotic non-initiation may be challenging, as presenting signs of LOI are similar among infants with and without confirmed infection.

Assessment of Illness Severity in Adults Hospitalized With Acute Respiratory Tract Infection due to Influenza, Respiratory Syncytial Virus, or Human Metapneumovirus.

Influenza, respiratory syncytial virus (RSV), and human metapneumovirus (hMPV) are common respiratory viruses causing similar symptoms. Optimal tools to assess illness severity for these viruses have not been defined. Using the Hospitalized Acute Respiratory Tract Infection (HARTI) study data, we report symptom severity by clinician-rated clinical severity scores (CSS) in adults with influenza, RSV, or hMPV and correlations between CSS and patient-reported outcomes (PROs). HARTI was a global epidemiologic study in adults hospitalized with acute respiratory tract infections. Patients were assessed at enrollment within 24 h of admission with CSS and twice during hospitalization with CSS, Respiratory Infection Intensity and Impact Questionnaire™ (RiiQ™), and EQ-5D-5L. Data were summarized descriptively, stratified by pathogen and baseline and hospitalization characteristics. Domain (general, upper respiratory, and lower respiratory) and sign/symptom subscores are presented for CSS; sign/symptom subscores are presented for RiiQ™ results. Data from 635 patients with influenza, 248 with RSV, and 107 with hMPV were included. At enrollment, total CSS and general and lower respiratory signs/symptoms (LRS) scores were higher for RSV and hMPV than influenza. Between-pathogen differences were greatest for LRS scores. Dyspnea, rales/rhonchi, wheezing, and shortness of breath scores trended higher for RSV and hMPV than influenza. RiiQ™ scores for cough, fatigue, and short of breath were strongly correlated with corresponding clinician-rated symptoms. These findings support the use of PROs (e.g., the RiiQ™) correlating with clinician assessments to gauge patient well-being and aid patient management by accurately assessing respiratory illness severity due to RSV, hMPV, or influenza.

Integrated Analysis Identified TGFBI as a Biomarker of Disease Severity and Prognosis Correlated with Immune Infiltrates in Patients with Sepsis.

Sepsis is a major contributor to morbidity and mortality among hospitalized patients. This study aims to identify markers associated with the severity and prognosis of sepsis, providing new approaches for its management and treatment. Data were mined from the Gene Expression Omnibus (GEO) databases and were analyzed by multiple statistical methods like the Spearman correlation coefficient, Kaplan-Meier analysis, Cox regression analysis, and functional enrichment analysis. Candidate indicator' associations with immune infiltration and roles in sepsis development were evaluated. Additionally, we employed techniques such as flow cytometry and neutral red staining to evaluate its impact on macrophage functions like polarization and phagocytosis. Twenty-eight genes were identified as being closely linked to the severity of sepsis, among which transforming growth factor beta induced (TGFBI) emerged as a distinct marker for predicting clinical outcomes. Notably, reductions in TGFBI expression during sepsis correlate with poor prognosis and rapid disease progression. Elevated expression of TGFBI has been observed to mitigate abnormalities in sepsis-related immune cell infiltration that are critical to the pathogenesis and prognosis of the disease, including but not limited to type 17 T helper cells and activated CD8 T cells. Moreover, the protein-protein interaction network revealed the top ten genes that interact with TGFBI, showing significant involvement in the regulation of the actin cytoskeleton, extracellular matrix-receptor interactions, and phagosomes. These are pivotal elements in the formation of phagocytic cups by macrophages, squaring the findings of the Human Protein Atlas. Additionally, we discovered that TGFBI expression was significantly higher in M2-like macrophages, and its upregulation was found to inhibit lipopolysaccharide-induced polarization and phagocytosis in M1-like macrophages, thereby playing a role in preventing the onset of inflammation. TGFBI warrants additional exploration as a promising biomarker for assessing illness severity and prognosis in patients with sepsis, considering its significant association with immunological and inflammatory responses in this condition.

A Comprehensive Review on Serum Ferritin as a Prognostic Marker in Intensive Care Units: Insights Into Ischemic Heart Disease.

Serum ferritin has garnered considerable attention as a prognostic marker in intensive care units (ICUs), offering valuable insights into patient outcomes and clinical management strategies. This comprehensive review examines the role of serum ferritin in predicting outcomes among critically ill patients, with a particular focus on its implications for ischemic heart disease (IHD). Elevated serum ferritin levels have consistently been associated with adverse outcomes in ICU settings, including increased mortality, prolonged hospital stays, and higher morbidity rates. Furthermore, the relationship between serum ferritin levels and IHD underscores its potential as a biomarker for cardiovascular risk assessment in critically ill populations. The review synthesizes existing literature to highlight the predictive value of serum ferritin in assessing illness severity and guiding clinical decision-making in the ICUs. It also explores potential mechanisms linking serum ferritin to adverse outcomes and discusses implications for clinical practice. Integrating serum ferritin measurements into routine assessments could enhance prognostication and risk stratification in ICU patients, while further research is needed to elucidate optimal management strategies and therapeutic targets. Collaborative efforts between clinicians and researchers are essential to advance our understanding of serum ferritin's prognostic value in the ICUs and translate this knowledge into improved patient care and outcomes.

Interfacility Referral Communication for PICU Transfer.

For patients requiring transfer to a higher level of care, excellent interfacility communication is essential. Our objective was to characterize verbal handoffs for urgent interfacility transfers of children to the PICU and compare these characteristics with known elements of high-quality intrahospital shift-to-shift handoffs. Mixed methods retrospective study of audio-recorded referral calls between referring clinicians and receiving PICU physicians for urgent interfacility PICU transfers. Academic tertiary referral PICU. Children 0-18 years old admitted to a single PICU following interfacility transfer over a 4-month period (October 2019 to January 2020). None. We reviewed interfacility referral phone calls for 49 patients. Referral calls between clinicians lasted a median of 9.7 minutes (interquartile range, 6.8-14.5 min). Most referring clinicians provided information on history (96%), physical examination (94%), test results (94%), and interventions (98%). Fewer clinicians provided assessments of illness severity (87%) or code status (19%). Seventy-seven percent of referring clinicians and 6% of receiving PICU physicians stated the working diagnosis. Only 9% of PICU physicians summarized information received. Interfacility handoffs usually involved: 1) indirect references to illness severity and diagnosis rather than explicit discussions, 2) justifications for PICU admission, 3) statements communicating and addressing uncertainty, and 4) statements indicating the referring hospital's reliance on PICU resources. Interfacility referral communication was similar to intrahospital shift-to-shift handoffs with some key differences: 1) use of contextual information for appropriate PICU triage, 2) difference in expertise between communicating clinicians, and 3) reliance of referring clinicians and PICU physicians on each other for accurate information and medical/transport guidance. Interfacility PICU referral communication shared characteristics with intrahospital shift-to-shift handoffs; however, communication did not adhere to known elements of high-quality handovers. Structured tools specific to PICU interfacility referral communication must be developed and investigated for effectiveness in improving communication and patient outcomes.

Use of bitemporal NACA score documentation in prehospital emergency medical services– a retrospective study

BackgroundThe assessment of illness severity in the prehospital setting is essential for guiding appropriate medical interventions. The National Advisory Committee for Aeronautics (NACA) score is a validated tool commonly used for this purpose. However, the potential benefits of using bitemporal documentation of NACA scores to capture the dynamic changes in emergency situations remain uncertain. The objective of this study was to evaluate the potential benefit of bitemporal NACA score documentation in the prehospital setting, specifically in assessing the dynamic changes of emergencies and facilitating quality improvement through enhanced documentation practices.MethodsIn this retrospective study, data from prehospital emergency patients were analyzed who received care from the physician response unit between January 1, 2018, and May 31, 2021. Patient demographics, NACA scores, indications for emergency care, and changes in NACA scores were extracted from medical records. Statistical analyses were performed to examine the associations between NACA scores, emergency categories, indications, and changes in NACA scores.ResultsThe study included 4005 patients, predominantly categorized as NACA III (33.7% at initial assessment, 41.8% at subsequent assessment) and NACA IV (31.6% at initial assessment, 22.4% at subsequent assessment). There was a significant improvement in NACA scores during the provision of prehospital care (p < 0.01). Notably, prehospital emergencies attributed to internal medical, neurological, traumatic, and paediatric causes demonstrated significant improvements in NACA scores (p < 0.01). Gender-specific differences were also observed.ConclusionOur study suggests that the bitemporal documentation of NACA scores can be advantageous in the prehospital setting and may have implications for research, practice, and policy.

Assessment of the variation of the pulmonary gas exchange model in the critical care setting

Background and Aims: In intensive care units (ICUs), clinicians use various mathematical models to assess and monitor respiratory illness. However, evidence suggests that some pulmonary internal and external physiolog¬ical factors are not integrated in the calculation of these models. This research aimed to demonstrate the variation in these models within a short period in critical care settings. Methods: A retrospective cohort study was conducted from 2017 to 2019 with adults admitted to an ICU and mechanically ventilated for more than 24 hours. Two different arterial blood gas samples were selected within a mean of 120 minutes. The coefficient of variation percent [(CV%)=SD/Mean] was calculated. Results: In total, 211 intubated ICU patients were included. The mean age was 57 (SD +/- 19.62) years, and 33% were female. The degree of variation of the average CV% for all the mathematical models was 14%-30%. The mean CV% for the arterial oxygen tension to inspired oxygen fraction (PaO2/FiO2) and arterial/alveolar oxygen tension ratio (PaO2/ PAO2) was 14%, which is a low variation compared to the respiratory index (RI) for which the mean CV% was 30%. Conclusions: The findings of this study indicate that most of the existing oxygen mathematical models used in clinical settings lack physiological realism, necessitating a new mathematical model that integrate both assess illness severity and enhance other models’ accuracy.

IntelliCare: Integrating IoT and Machine Learning for Remote Patient Monitoring in Healthcare: A Comprehensive Framework

The development of smart health monitoring systems has emerged as a consequence of the integration of Internet of Things (IoT) and Machine Learning (ML) technologies within the healthcare sector. This transformation has significantly reshaped patient care methodologies, shifting from traditional approaches to electronic healthcare systems. Leveraging IoT technology fosters a contemporary medical device ecosystem, fostering seamless communication among healthcare professionals, patients, and medical devices. Through the deployment of IoT devices, encompassing sensors and transmitters, coupled with Machine Learning algorithms, various applications have arisen, spanning from remote patient monitoring to real-time health assessment during ambulance transit to medical facilities. This proposed framework aims to monitor essential physiological parameters including body temperature, blood pressure, heart rate, sweat analysis, glucose levels, ECG, EEG, and pulse oximetry, transmitting pertinent data for tailored processing and analysis. Implantable IoT devices serve as conduits for wireless communication, data storage, centralized computation, and portable processing, facilitating connectivity among sensors, GPS-enabled ambulances, medical practitioners, and patients. To mitigate potential health risks, sensors are equipped with Machine Learning capabilities to promptly assess illness severity and recommend appropriate interventions, potentially triggering automated alerts to healthcare providers. This seamless exchange of information via IoT and wireless networks enables rapid communication between doctors and patients, facilitating personalized medical recommendations, prescription management, and hospital selection based on individual health profiles.

Illness severity assessment of older adults in critical illness using machine learning (ELDER-ICU): an international multicentre study with subgroup bias evaluation

Comorbidity, frailty, and decreased cognitive function lead to a higher risk of death in elderly patients (more than 65 years of age) during acute medical events. Early and accurate illness severity assessment can support appropriate decision making for clinicians caring for these patients. We aimed to develop ELDER-ICU, a machine learning model to assess the illness severity of older adults admitted to the intensive care unit (ICU) with cohort-specific calibration and evaluation for potential model bias. In this retrospective, international multicentre study, the ELDER-ICU model was developed using data from 14 US hospitals, and validated in 171 hospitals from the USA and Netherlands. Data were extracted from the Medical Information Mart for Intensive Care database, electronic ICU Collaborative Research Database, and Amsterdam University Medical Centers Database. We used six categories of data as predictors, including demographics and comorbidities, physical frailty, laboratory tests, vital signs, treatments, and urine output. Patient data from the first day of ICU stay were used to predict in-hospital mortality. We used the eXtreme Gradient Boosting algorithm (XGBoost) to develop models and the SHapley Additive exPlanations method to explain model prediction. The trained model was calibrated before internal, external, and temporal validation. The final XGBoost model was compared against three other machine learning algorithms and five clinical scores. We performed subgroup analysis based on age, sex, and race. We assessed the discrimination and calibration of models using the area under receiver operating characteristic (AUROC) and standardised mortality ratio (SMR) with 95% CIs. Using the development dataset (n=50 366) and predictive model building process, the XGBoost algorithm performed the best in all types of validations compared with other machine learning algorithms and clinical scores (internal validation with 5037 patients from 14 US hospitals, AUROC=0·866 [95% CI 0·851-0·880]; external validation in the US population with 20 541 patients from 169 hospitals, AUROC=0·838 [0·829-0·847]; external validation in European population with 2411 patients from one hospital, AUROC=0·833 [0·812-0·853]; temporal validation with 4311 patients from one hospital, AUROC=0·884 [0·869-0·897]). In the external validation set (US population), the median AUROCs of bias evaluations covering eight subgroups were above 0·81, and the overall SMR was 0·99 (0·96-1·03). The top ten risk predictors were the minimum Glasgow Coma Scale score, total urine output, average respiratory rate, mechanical ventilation use, best state of activity, Charlson Comorbidity Index score, geriatric nutritional risk index, code status, age, and maximum blood urea nitrogen. A simplified model containing only the top 20 features (ELDER-ICU-20) had similar predictive performance to the full model. The ELDER-ICU model reliably predicts the risk of in-hospital mortality using routinely collected clinical features. The predictions could inform clinicians about patients who are at elevated risk of deterioration. Prospective validation of this model in clinical practice and a process for continuous performance monitoring and model recalibration are needed. National Institutes of Health, National Natural Science Foundation of China, National Special Health Science Program, Health Science and Technology Plan of Zhejiang Province, Fundamental Research Funds for the Central Universities, Drug Clinical Evaluate Research of Chinese Pharmaceutical Association, and National Key R&D Program of China.

0319 Evaluating Illness Severity from Acute to Chronic Insomnia: Is the First the Worst?

Abstract Introduction The Spielman 3P and Perlis 4P models represent illness severity over the course of insomnia disorder. The 3P model suggests that illness severity is worst when subjects experience acute illness (as compared to the subchronic and chronic phases of the disorder). The 4P model suggests that illness severity crescendos with chronicity. The present analysis uses an archival data set to assess illness severity with new onset illness (i.e.,from Good Sleep [GS] to Acute Insomnia [AI] and from AI to Chronic Insomnia [CI]). Illness severity is quantified in terms of Total Wake Time (TWT). Number of affected days/week is a tabulation of nights/week with clinical insomnia symptoms. Methods Subjects were recruited as GSs(N=911), followed up to 1-year with digital sleep diaries, and were classified as GS, AI, or CI based on standardized criteria. Data for CI subjects were anchored to the 1st of 14 days with insomnia so that day-to-day TWT, SL, WASO, and EMA were represented for ~100 days prior-to-and-following AI onset. A similar graphic (+/-Acute onset) was constructed for number of days per week with insomnia. GS data were matched to CI subjects dates (i.e., the data were temporally tethered). Segmented linear mixed regression models were applied to examine the change in slopes in the AI-to-CI period compared to GS and AI period. Results 883 subjects remained as GSs (Mage=53.1±10.8years;35.1%female;83.0%white), while 28 indi-viduals transitioned to AI and then CI (Mage=53.6±12.6years;39.3%female;89.3%white). Average TWT rose over the course of the first 2 weeks from the AI onset(b=1.6,SE=0.52minutes,p=0.002) and then was stable for 3 months(b=-0.004,SE=0.03,p=0.91). This worsening of sleep continuity was almost entirely related to increased WASO. Average number of affected days was found to be stable from AI to CI (b=0.002,SE=0.002,p=0.20). That is, while there was week-to-week variability in the number of days affected, no linear trend was evident. Conclusion These data suggest that, in our sample of CIs with primarily middle insomnia, the average severity and number of affected days was worst with the onset of AI (worst is first) and stable thereafter. Whether this temporal trend applies equally to new onset Initial and Late insomnia remains to be determined. Analyses ongoing. Support (if any) R01AG041783

A CROSS-SECTIONAL STUDY ON IMPACT OF MEDICATIONS ON COGNITION IN PATIENTS WITH SCHIZOPHRENIA

Background: Cognitive decits in schizophrenia might start before onset of symptoms. Medications reduce the illness severity, but their effect on cognition is not clearly established. Poor cognition may worsen the prognosis and hence impact the overall functioning. To assess the effect Aim: of ongoing medications on cognition in patients with schizophrenia. A cross-sectiona Methods: l, observational study was conducted from JulyAugust, 2022 in a tertiary psychiatric hospital using convenience sampling in patients with schizophrenia on atleast 6-8weeks of same treatment medications. Patients with neurodevelopmental, neurological, neurodegenerative, chronic substance use or known personality disorders and drug naive patients were excluded. Brief Psychiatric Rating Scale was used to assess illness severity and Cognitive Assessment Interview was conducted to assess cognitive decit. 107 subjects(n=60;56.1% men and n=47;43.9% women) with Results: diagnosis of schizophrenia were considered; mean age being 35.8610.14 years, mean duration of illness being 8.015.62 years. Majority had mild cognitive impairment(n=50;46.7%). Pearson's correlation showed signicant positive correlation between cognitive functioning and olanzapine and uoxetine use(r= 0.195 and r=0.296 respectively; p &lt;0.05). Signicant negative correlation was observed between cognitive functioning and trihexyphenidyl use(-0.195; p&lt;0.05). Medications used in schizophrenia can have both benecial Conclusion: and deleterious effects on cognition. Further longitudinal studies on patients maintaining on the same medications will help to establish the association and hence their relative impact on cognition.

The illness severity of ward remaining patients reviewed by the medical emergency team: A retrospective cohort study.

Although progress has been made in identifying and responding to acutely deteriorating ward patients, judgements about the level of care required for patients after medical emergency team review are complex, rarely including a formal assessment of illness severity. This challenges staff and resource management practices and patient safety. This study sought to quantify the illness severity of ward patients after medical emergency team review. This retrospective cohort study examined the clinical records of 1500 randomly sampled adult ward patients following medical emergency team review at a metropolitan tertiary hospital. Outcome measures were the derivation of patient acuity and dependency scores using sequential organ failure assessment and nursing activities score instruments. Findings are reported using the STROBE guideline for cohort studies. No direct patient contact was made during the data collection and analysis phases of the study. Patients were male (52.6%), unplanned (73.9%) medical admissions (57.5%), median age of 67 years. The median sequential organ failure assessment score was 4% and 20% of patients demonstrated multiple organ system failure requiring non typical monitoring and coordination arrangements for at least 24 h. The median nursing activities score was 86% suggestive of a near 1:1 nurse-to-patient ratio. More than half of all patients required enhanced levels of assistance with mobilization (58.8%) and hygiene (53.9%) activities. Patients who remain on the ward following medical emergency team review had complex combinations of organ dysfunction, with levels of dependency similar to those found in intensive care units. This has implications for ward and patient safety and continuity of care arrangements. Profiling illness severity at the conclusion of the medical emergency team review may help determine the need for special resource and staffing arrangements or placement within the ward environment.

A novel cardiac arrest severity score for the early prediction of hypoxic-ischemic brain injury and in-hospital death

Abstract Funding Acknowledgements Type of funding sources: None. Background Out-of-hospital cardiac arrest (OHCA) outcomes are unsatisfactory despite postcardiac arrest care. Early prediction of prognoses might help stratify patients and provide tailored therapy. Purpose In this study, we derived and validated a novel scoring system to predict hypoxic-ischaemic brain injury (HIBI) and in-hospital death (IHD). Methods We retrospectively analysed Korean Hypothermia Network prospective registry data collected from in Korea between 2015 and 2018. Patients without neuroprognostication data were excluded, and the remaining patients were randomly divided into derivation and validation cohorts. HIBI was defined when at least one prognostication predicted a poor outcome. IHD meant all deaths regardless of cause. In the derivation cohort, stepwise multivariate logistic regression was conducted for HIBI and IHD scores, and model performance was assessed. We then classified patients into four categories and analysed associations between the categories and cerebral performance categories (CPCs) at hospital discharge. Finally, we validated our models in the internal validation cohort. Results Among 1373 patients, 240 were excluded, and 1133 were randomised into derivation (n=754) and validation cohorts (n=379). In the derivation cohort, 7 and 8 predictors were selected for HIBI (0–8) and IHD scores (0–11), respectively, and the area under the curve (AUC) was 0.85 (95% CI 0.82–0.87) and 0.80 (95% CI 0.77–0.82), respectively. Applying optimum cutoff values of ≥6 points for HIBI and ≥7 points for IHD, patients were classified as follows: HIBI (-)/IHD (-), Category 1 (n=424); HIBI (-)/IHD (+), Category 2 (n=100); HIBI (+)/IHD (-), Category 3 (n=21); and HIBI (+)/IHD (+), Category 4 (n=209). CPCs at discharge were significantly different in each category (p&amp;lt;0.001). In the validation cohort, the model showed moderate discrimination (AUC 0.83, 95% CI 0.79–0.87 for HIBI and AUC 0.77, 95% CI 0.72–0.81 for IHD) with good calibration. Each category of the validation cohort showed a significant difference in discharge outcomes (p&amp;lt;0.001) and a similar trend to the derivation cohort. Conclusions We presented a novel approach for assessing illness severity after OHCA. Although external prospective studies are warranted, risk stratification for HIBI and IHD could help provide OHCA patients with appropriate treatment.