Epidemiology of Non- Invasive Mechanical Ventilation in Argentina: a Multicenter Observational Study

Introduction Non-invasive mechanical ventilation (NIMV) is a fundamental strategy in the management of acute respiratory failure, preventing intubation and reducing complications. However, the epidemiology of its use in Argentina is not well understood, motivating this study. Justification Unequal access to ICU resources in Argentina affects the quality of care. This study aims to describe the implementation, effectiveness, and outcomes of NIMV across different regions of the country to improve clinical decision-making and healthcare policies. Objectives * Primary: Analyze the survival of ICU patients receiving NIMV. * Secondary: Evaluate success/failure factors, regional disparities in resource availability, the impact of healthcare staff training, comparison between public and private sectors, and establish a national database. Methods Study Design Observational, multicenter, prospective, and analytical study. Both public and private hospitals in Argentina will be included. Data will be collected over seven months, with patient follow-up until discharge or death. Study Population Adult patients (\>18 years) admitted to the ICU requiring NIMV for acute respiratory failure or as part of weaning from invasive mechanical ventilation. Inclusion Criteria * Age \>18 years. * Use of NIMV in the ICU for respiratory failure or during weaning. * Signed informed consent. Exclusion Criteria * Patients using home NIMV without exacerbation. * Patients without acute respiratory impairment. Data Collection and Management * Data will be recorded using an electronic form in RedCap®, with restricted access. * Confidentiality will be ensured through alphanumeric coding. * A monitoring committee will supervise data quality. Statistical Analysis * Continuous variables: Mean and standard deviation or median and interquartile range. * Comparisons: Student's t-test, Mann-Whitney U test, Chi² test, or Fisher's exact test. * Multivariate models: Logistic regression and Kaplan-Me