Predicting Clostridium difficile infection: Preoperative risk factors


Principal Investigator: GRETA L KRAPOHL
Affiliation: University of Michigan
Country: USA
Abstract: DESCRIPTION (provided by applicant): Clostridium difficile is the most common organism causing hospital-acquired infections. The steady rise in the incidence of Clostridium difficile infection (CDI) is compounded with a parallel increase in the severity and morbidity of the disease and a growing elderly population that is disproportionately affected by CDI. The broad, long-term objectives of this proposal are to develop, validate and implement a clinical prediction rule (risk score) for the quantification of infection risk for clinical bedside implementation. We propose that a clinical prediction rule to identify the patients most vulnerable for CDI early in their hospitalization is a strategy to improve patient outcomes so that preventive interventions and treatments can be targeted to at-risk patients before, not in response to, infectious disease. This proposal directly addresses the National Institute of Nursing Research's (NINR) research strategies for building nursing science by the use of interdisciplinary collaboration to improve patient safety. Specifically, this work will enable the translation of research-based evidence into operational clinical bedside practice. Our proposal is the first phase, in a series of studies, to develop a clinical prediction rule to prevent CDI in high-risk patients. Using a retrospective, cohort design, we will determine the preoperative risk factors of CDI from over two years of colectomy surgical patients data (1800 patients) from the 23 hospitals enrolled in the Michigan Surgical Quality Collaborative (MSQC). The prediction rule will be constructed through three specific aims: (1) Determination of the preoperative risk factors of colectomy patients diagnosed with CDI as compared with colectomy patients without CDI, (2) Integration of the most robust preoperative variables associated with CDI into a clinical prediction rule model and (3) Evaluation of the predictive accuracy of the CDI prediction rule with statistical confirmation. This study is the foundation for a valid, accurate and relevant prediction rule to detect vulnerable, at-risk patients. Relevance of Research to Public Health: The escalation of CDI in hospitals is emerging as a serious medical and public health problem. This research proposes a strategy to prevent CDI and improve patient outcomes by heightening the ability to identify the most vulnerable patients and target preventative intervention.
Funding Period: ----------------2009 - ---------------2011-
more information: NIH RePORT