Trends in mortality following Clostridium difficile infection in Scotland, 2010–2016: a retrospective cohort and case–control study. Issue 2 (October 2018)

• Record Type: Journal Article
• Authors: Banks, A.; Moore, E.K.; Bishop, J.; Coia, J.E.; Brown, D.; Mather, H.; Wiuff, C.
• Is Part Of: Journal of hospital infection. Volume 100:Issue 2(2018)
• Page Start: 133
• Page End: 141
• Publication Date: 2018-10
• Subjects: Clostridium difficile infection -- PCR ribotype -- Mortality -- Risk factors -- Surveillance -- Length of stay; Cross infection -- Periodicals; Cross infection -- Prevention -- Periodicals; Nosocomial infections -- Periodicals; Nosocomial infections -- Prevention -- Periodicals; Cross Infection -- Periodicals; Cross Infection -- prevention & control -- Periodicals; Infection Control -- Periodicals; Electronic journals; 614.44
• DOI: 10.1016/j.jhin.2018.07.023
• Languages: English
• ISSNs: 0195-6701

Abstract

Summary: Background: National surveillance of Clostridium difficile infection (CDI) in Scotland enables the monitoring of trends in incidence rates but not mortality. Aim: To assess factors associated with mortality for all CDI cases aged ≥15 years in Scotland between 2010 and 2016. Methods: All CDI cases aged ≥15 years in Scotland between 2010 and 2016 were linked to hospital admission and mortality datasets. Logistic regression was used to assess factors associated with mortality (30-day all-cause). A case–control study of a hospitalized subset of cases and matched hospitalized controls assessed the impact of CDI on mortality and length of stay. Findings: Thirty-day all-cause mortality decreased over the seven-year period (from 20.5% to 15.6%; P < 0.001), mainly among healthcare-associated CDI (HA-CDI). Increased age, higher Charlson score, HA-CDI, as well as liver, heart and malignancy comorbidities were associated with higher mortality. No association was observed between polymerase chain reaction ribotype and higher mortality, though 015 and 078 were associated with lower mortality. Adjusted odds ratio (OR) for 30-day mortality in hospitalized CDI cases compared to controls was 2.67 (95% confidence interval (CI): 2.42–2.94; P < 0.001). Whereas mortality declined over time in cases and controls, the trend in ORs remained relatively stable. Having CDI increased additional mean length of stay beyond infection by 22.3% (95% CI: 18.0–26.8%; P < 0.001). Conclusion: CDI is associated with an almost three-fold increase in 30-day mortality and places an increased burden on hospital resources by increasing mean LOS beyond the infection date by 22.3%. The decreasing CDI mortality trends may be due to overall improvements in mortality among the general and hospital population of Scotland. Therefore, despite large declines in incidence rates, CDI remains a serious healthcare problem.