Mortality Prediction

The purpose of this research is test if combination of physical tests and blood tests drawn before SCTCT can predict NRM.

In adults and children low or high blood chloride levels are linked to the risk of death. The aim of this observational study is to determine whether there is a relationship between low or high blood chloride levels and the risk of death or long-term lung problems. We will also learn the risk factors and associated conditions of high or low blood chloride levels. We will include infants born before 32 weeks of pregnancy or have a birth weight of less than 1500 grams in the study. The main question it aims to answer is: Is there a relationship between low or high blood chloride levels in the first 4-6 weeks of life and risk of death or long-term lung problems in premature babies? We will examine the medical reports of babies who were followed up in neonatal intensive care unit over the past 5 years.

Emphysema linked to alpha-1-antitrypsin deficiency (DAAT): towards a better prediction of risks Emphysema caused by alpha-1-antitrypsin deficiency (DAAT) is a rare genetic disorder that can lead to serious complications, such as the need for a lung transplant or death, affecting up to 15% of patients. The only specific treatment available is a weekly infusion of alpha-1-antitrypsin (IV-AAT), an expensive and burdensome therapy. Currently, there is no reliable model to predict the course of the disease in these patients. Our study, conducted in several French hospitals, aims to develop a prediction tool combining clinical, biological, functional data and advanced medical image analysis (lung CT). This model will make it possible to identify the most at-risk patients, in order to better adapt their care, anticipate transplant needs and avoid unnecessary treatments for low-risk patients. Ultimately, this approach could also improve access to care for patients who need it most, while optimizing health system resources.

Acute heart failure (AHF) is a leading cause of hospitalization and is associated with high short-term morbidity and mortality, with 20-30% of patients experiencing rehospitalization or death within 30 days. Early adverse events often reflect incomplete recovery, highlighting the need for improved risk stratification after clinical stabilization .Current prognostic approaches mainly focus on hemodynamic congestion. Persistent pulmonary congestion at discharge is a strong predictor of poor outcomes, but these markers primarily assess macrocirculatory abnormalities and do not capture microcirculatory dysfunction, which may persist despite apparent clinical improvement. Lung ultrasound, through the Lung Ultrasound Score (LUS), provides a validated assessment of pulmonary congestion and has demonstrated prognostic value in AHF. However, LUS does not reflect systemic tissue perfusion. In contrast, blood lactate is a robust marker of tissue hypoperfusion, and even mild elevations have been associated with worse outcomes in AHF. A combined score integrating LUS and lactate may therefore better reflect the dual pathophysiology of AHF-persistent congestion and impaired tissue perfusion-and improve prediction of early adverse events. This protocol aims to validate the prognostic value of this combined score for predicting 30-day rehospitalization or death in patients hospitalized for AHF, with the hypothesis that it outperforms LUS alone.

A respiratory intermediate care unit (RICU) is a monitoring and treatment area of respiratory patients who do not required admission to intensive care unit (ICU) but due to complexity, they could not be managed in conventional ward. Aim: To investigate those patients that could better benefit from RICU stay. Hypothesis: a comprehensive and integrative knowledge of all factors that intervene during the RICU admission allow determining probability of survival. Primary outcome: 1. To construct a predictive model of mortality at 30-days after RICU admission for patients admitted to the coordinator RICU based on standard biostatistics: The BELLRICU Model. Secondary outcomes: 2.1. To validate the model in another cohort of patients admitted at the same RICU. 2.2. To validate the model in an external cohort (patients admitted at the rest of Catalan active RICUs at the time of the study). 2.3. To compare the predictive capacity of the BELLRICU model with other previous validated scales but in ICU setting. 2. 4. To explore a new predictive model using artificial intelligence (AI) techniques. 2.4. To design a quick app to implement the BELLRICU model. Methodology: Longitudinal prospective study (3 years), recording variables at baseline, at RICU admission and 30-days follow-up. During the first two years, variables will be collected from the coordinator RICU to construct the BELLRICU model, being "mortality after 30-day of RICU admission" the dependent varialbe and using regression of cox proportional risks analysis. During the third year of the study, the BELLRICU model will be applicated to the rest of the participants RICUs in order to validate the model. Further, the predictive capacity of the BELLRICU model will be compared with the predictive capacity of previous validated scales in ICU setting and with a exploratory model using AI from BELLRICU data base.

ICU mortality indicates the severity of disease, healthcare quality, and the efficacy of interventions. The severity scores are tools to predict the risk of mortality in the ICU, and the APACHE II score is frequently used for this purpose. However, studies validating the score in Colombia are limited. There is uncertainty about the precision and discrimination capacity of the APACHE II score in a population that varies from the original, with varying diseases, and in a different timeline. The investigators determined to evaluate: 1. Evaluate the rate of mortality in the ICU by type of disease and type of admission. 2. The factors associated with mortality. 3. Validate the performance of the APACHE II score as a predictor of mortality.

There are several scores for predicting mortality within 28 days in the intensive care unit, including APACHE 2, MPM 2 and MODS. These models for predicting mortality can help intensive care physicians to identify patients at high risk of mortality, thus helping them in the decision regarding their admission to the ICU and the management of their care during hospitalization. However, it is not entirely clear whether these models can predict mortality better than an experienced ICU physician or nurse. The purpose of the study is to compare the ability of a senior ICU physician and nurse to predict mortality within 28 days in intensive care patients and between the predictive ability of three models for mortality prediction, as well as to investigate what were the parameters that particularly influenced the mortality prediction of the medical and nursing staff. The study will be performed as a prospective observational study and will include approximately 2000 patients. For each patient admitted to intensive care, the risk of mortality will be calculated according to APACHE 2, MPM 2 and MODS. In addition, an independent evaluation by 3 specialist ICU physicians and an experienced ICU nurse will be carried out. The medical and nursing staff's reasons for the risk provided by them will be presented.

Mortality Risk Assessment by Skilled Caregivers Compared to Existing Validated Tools in Skilled Nursing Departments at Shmuel Harofeh Geriatric Hospital Background The elderly population in Israel and globally is growing, increasing demand for medical services, particularly palliative care. Recommendations from 2016 emphasized the need for geriatric and skilled nursing departments to focus on end-of-life care, but implementation has been limited. High mortality and frequent readmissions are reported in long-term care, yet accurate mortality prediction tools for elderly patients remain limited. Improved mortality prediction can help identify patients who would benefit from palliative care and reduce unnecessary interventions. Research Objectives 1. Assess life expectancy of patients in skilled nursing departments. 2. Compare the effectiveness of various tools in predicting six-month mortality. Hypothesis Caregiver assessments will more accurately predict mortality than current validated tools. Study Design Type: Prospective cohort study. Location: Shmuel Harofeh Hospital Study Population Approximately 250 patients admitted to skilled nursing departments at Shmuel Harofeh Hospital. Recruitment Period: Two years. Follow-up Period: Up to one year. Methods Epidemiological and clinical data (age, comorbidities, functional and cognitive status, lab results) will be collected. Mortality risk will be assessed using: 1. Validated Tools: Including the MITCHELL scale (for patients with advanced dementia) and the POROCK scale (for institutionalized patients). 2. Caregiver Assessment: Subjective life expectancy estimates by attending geriatricians and nursing staff within three days of admission and again 7-10 days later. An external geriatrician will also provide an assessment based on brief, non-invasive observation. Data Processing Data will be coded, entered into an electronic dataset, and undergo statistical analysis after collection. No interventions beyond routine care are included. Ethical Considerations As an observational study without intervention, a waiver for informed consent was granted. Importance of Research Skilled nursing facilities increasingly need to provide palliative care for elderly patients. This study aims to improve mortality prediction methods, helping to identify patients for end-of-life care, ultimately enhancing care quality, and reducing costs by avoiding unnecessary hospitalizations and treatments.