Mortality and Clostridium Difficile Infection in an Australian Setting

Abstract

<p>Aim: To quantify the risk of death associated with <strong>Clostridium</strong><strong> </strong><strong>difficile</strong><strong> </strong><strong>infection</strong><strong>, </strong><strong>in</strong><strong> </strong><strong>an</strong><strong> </strong><strong>Australian</strong> tertiary hospital. Background Two reviews examining <strong>Clostridium</strong><strong> </strong><strong>difficile</strong><strong> </strong><strong>infection</strong> and <strong>mortality</strong> indicate that <strong>Clostridium</strong><strong> </strong><strong>difficile</strong><strong> </strong><strong>infection</strong> is associated with increased <strong>mortality</strong><strong> </strong><strong>in</strong> hospitalized patients. Studies investigating the <strong>mortality</strong><strong> </strong>of<strong> </strong><strong>Clostridium</strong><strong> </strong><strong>difficile</strong><strong> </strong><strong>infection</strong><strong> </strong><strong>in</strong><strong> </strong><strong>settings</strong> outside of Europe and North America are required, so that the epidemiology of <strong>Clostridium</strong><strong> </strong><strong>difficile</strong><strong> </strong><strong>infection</strong><strong> </strong><strong>in</strong> these regions can be understood and appropriate prevention strategies made.</p> <p>Design: <strong>An</strong> observational non-concurrent cohort study design was used.</p> <p>Methods: Data from all persons who had (exposed) and a matched sample of persons who did not have <strong>Clostridium</strong><strong> </strong><strong>difficile</strong><strong> </strong><strong>infection</strong>, for the calendar years 2007-2010, were analysed. The risk of dying within 30, 60, 90 and 180 days was compared using the two groups. Kaplan-Meier survival analysis and conditional logistic regression models were applied to the data to examine time to death and <strong>mortality</strong> risk adjusted for comorbidities using the Charlson Comorbidity Index.</p> <p>Results: One hundred and fifty-eight cases of <strong>infection</strong> were identified. A statistically significant difference <strong>in</strong> all-cause <strong>mortality</strong> was identified between exposed and non-exposed groups at 60 and 180 days. <strong>In</strong> a conditional regression model, <strong>mortality</strong> <strong>in</strong> the exposed group was significantly higher at 180 days.</p> <p>Conclusion: <strong>In</strong> this <strong>Australian</strong> study, <strong>Clostridium</strong> <strong>difficile</strong><strong> </strong><strong>infection</strong> was associated with increased <strong>mortality</strong>. <strong>In</strong> doing so, it highlights the need for nurses to immediately instigate contact precautions for persons suspected of having <strong>Clostridium</strong><strong> </strong><strong>difficile</strong><strong> </strong><strong>infection</strong> and to facilitate a timely faecal collection for testing. Our findings support ongoing surveillance of <strong>Clostridium</strong><strong> </strong><strong>difficile</strong><strong> </strong><strong>infection</strong> and associated prevention and control activities.</p>