The estimated US burden of CAP is substantial, with >1.5 million unique adults being hospitalized annually, 100000 deaths occurring during hospitalization, and approximately 1 of 3 patients hospitalized with CAP dying within 1 year.
Machine learning approaches to modeling of epidemiologic data are becoming increasingly more prevalent in the literature. These methods have the potential to improve our understanding of health and opportunities for intervention, far beyond our past capabilities. This article provides a walkthrough for creating supervised machine learning models with current examples from the literature. From identifying an appropriate sample and selecting features through training, testing, and assessing performance, the end-to-end approach to machine learning can be a daunting task. We take the reader through each step in the process and discuss novel concepts in the area of machine learning, including identifying treatment effects and explaining the output from machine learning models.
Few patients with community-acquired pneumonia (CAP) require admission to the intensive care unit (ICU-CAP). However, they represent the most severe form of the disease. An understanding of the etiologic agents of ICU-CAP may lead to better treatment decisions and patient outcomes. The objective of this study was to determine the incidence of respiratory viruses in patients with ICU-CAP. This was an observational study conducted in six Kentucky hospitals from December 2008 through October 2011. A case of ICU-CAP was defined as a patient admitted to an ICU with the diagnosis of CAP. The Luminex xTAG multiplex polymerase chain reaction (PCR) assay was used for viral identification. A total of 468 adult and pediatric patients with ICU-CAP were enrolled in the study. A total of 92 adult patients (23 %) and 14 pediatric patients (19 %) had a respiratory virus identified. Influenza was the most common virus identified in adults and the second most common in pediatric patients. This study suggests that respiratory viruses may be common etiologic agents of pneumonia in patients with ICU-CAP. The Centers for Disease Control and Prevention (CDC) recommend empiric anti-influenza therapy during the winter for hospitalized patients with CAP. This study supports this recommendation in patients with ICU-CAP.
The objective of this study was to compare systemic and local cytokine profiles and neutrophil responses in patients with severe versus non-severe community-acquired pneumonia (CAP). Hospitalized patients with CAP were grouped according to the pneumonia severity index (PSI), as non-severe (PSI < 91 points) or severe (PSI ≥ 91 points). Blood and sputum samples were collected upon admission. Compared to non-severe CAP patients, the severe CAP group showed higher plasma levels of pro- and anti-inflammatory cytokines but in contrast, lower sputum concentrations of pro-inflammatory cytokines. Blood neutrophil functional responses were elevated in CAP patients compared to healthy controls. However, neutrophils from severe CAP patients showed reduced respiratory burst activity compared to the non-severe group. Results indicate that patients with severe CAP fail to mount a robust local pro-inflammatory response but exhibit instead a more substantial systemic inflammatory response, suggesting that a key driver of CAP severity may be the ability of the patient to generate an optimal local inflammatory response.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.