Research projects

Research projects

Research projects

Epidemiology

Over the years, we have conducted research on the incidence, mortality, and consequences of sepsis. Additionally, we study the relationship between modifiable lifestyle factors and the risk of severe infections. Among other things, we have published articles on the relationship between obesity, physical activity, mental health, and sleep deprivation and the future risk of sepsis. The group has also focused on the relationship between other medical conditions such as iron deficiency anemia, thyroid disease, and chronic kidney failure and the development of bloodstream infections and sepsis. For these studies, we have used classical epidemiological methods such as Cox regression analyses, as well as more modern methods such as mediation analyses and Mendelian randomization.

Human genetics

We have studied the genetic risk for various infectious diseases such as skin and soft tissue infections, urinary tract infections, and sepsis. This has often been done through genome-wide association studies (GWAS) in large cohorts such as HUNT and the UK Biobank. This has provided insights into entirely new disease mechanisms that could be new treatment targets for these diseases. Equally important, GWAS has given us a new tool for studies in genetic epidemiology by serving as the basis for Mendelian randomization.

Clinical studies and infection prevention 

We have conducted several clinical studies on various issues within bloodstream infections and sepsis. In particular, we have focused on the identification of infections associated with intravascular catheters, the effect of antibiotic treatment in accordance with national guidelines for sepsis and the benefit of introducing a standardised sepsis course with a monitoring system at the ward. "Stop Sepsis Nurse" received the Improvement Award for 2016 and the Norwegian Quality Award for the Health Service in 2017 to show that targeted and structured work among nurses reduced mortality from sepsis. This was a collaboration between Levanger Hospital, Nord University and the Centre for Sepsis Research in Central Norway. We have conducted both retrospective studies with a review of hospital records and prospective studies such as studies of the quality of care of peripheral venous catheters.

We have several ongoing studies on characterizing the course of the disease in detail for a large number of patients with sepsis and comparing this with a deep dive into the properties of the bacteria, in the hope of finding warning signs of a severe course for the individual patient and to map possible new treatment targets for sepsis. In a pilot study, we investigate the immunomodulatory effect of a newly developed and patented peptide from CEMIR, NTNU.

Microbiology

Interaction between Bacteria and Host: The Importance of Bacterial and Host Genotype and Phenotype in Sepsis (Research Project)
At Levanger Hospital, clinical data and bacterial isolates have been collected from more than 5000 patients aged 18 and older who have been hospitalized in Nord-Trøndelag with bloodstream infection (BSI) since 1994. Clinical data from these patients are now available in a local quality registry (Sepsis Registry). We are currently working on two projects where we analyze clinical data from the Sepsis registry and whole genome sequence data for E. coli and S. aureus isolates from blood cultures. The main goal of the project is to understand how the characteristics of the bacteria causing BSI and sepsis, and the characteristics of the patients who get such infections, influence susceptibility, severity, and outcome of BSI and sepsis.

Rapid Culture-Independent Diagnosis of Bloodstream Infection and Sepsis (Innovation Project)
Rapid and effective antibiotic treatment is crucial to ensure the best possible outcome of serious infections such as bloodstream infection (BSI) and sepsis. Early identification of the microbe causing the infection is important to be able to target the treatment as early as possible. The purpose of the project is to investigate the usefulness of rapid culture-independent methods for detecting bacteria and fungi in blood within a few hours compared to blood culture, which takes one to several days. We include patients admitted to the main intensive care unit and the emergency department at St. Olavs University Hospital with suspected BSI or sepsis.

Streptococcus agalactiae - Systemic Infection in Adults
In this project, we describe the incidence and mortality of systemic Streptococcus agalactiae (group B streptococci, GBS) infection in adults in Norway during the period 1996-2019, focusing on factors that may influence the incidence and outcome of the disease. The project is done in collaboration with the National reference laboratory for GBS at Department of medical microbiology, St. Olavs hospital.

Artificial Intelligence 

Computational SEpsis Mining and modelling (CoSeM) 
The CoSeM project aims to provide prevention and intervention for patients at risk of bloodstream infections and sepsis. The focus is on capturing individual patient histories to characterize signs of sepsis susceptibility which can be utilized for patient risk mitigation and monitoring of infection and sepsis. Funding was provided by the NTNU Health and Life Sciences Strategic Area. For more information, see the CoSeM website.

De-identifying Medical Records Project
De-identifying Medical Records Project focuses on developing software to de-identify medical records and integrating it into a pipeline at the Central Norway Regional Health Authority's IT department (Hemit). This project has de-identified over 100,000 medical records and/or adverse event reports for various projects related to sepsis and mental health. Funding was partially provided by Strategic Funding from the NTNUs Department of Computer Science (IDI).
 

  
Machine Learning for Adverse Event Analysis (MLAA)
The MLAA project focuses on developing new efficient methods for processing adverse events to better monitor adverse events and identify 
potential areas for improvement as learning opportunities. Funding was provided by Innovation Funding from the Central Norway Regional Health Authority (https://www.helse-midt.no/).
 


Effect of Written Guidance for Best Practices in Staphylococcus aureus Bacteremia – A Retrospective Observational Study Using a Large Language Model for Data Extraction (Innovation Project)
This retrospective observational study aims to evaluate the effect of standardized written guidance on the management of Staphylococcus aureus bacteremia (SAB). We compare patient outcomes and adherence to best practices before and after the implementation in 2019 at Helse Møre og Romsdal (HMR HF). Additionally, the study will investigate the use of a large language model (LLM) as a tool for extracting research data from electronic patient records (EPR).
This project serves as a pilot for an innovation initiative aimed at developing an LLM-based research tool and a pipeline for research on patient record data and for quality improvement in Helse Midt-Norge.