This prospective study was performed to determine the incidence, risk factors, severity and outcomes of community-associated Clostridium difficile infection (CA-CDI) in the SE of Scotland.

All patients (335) diagnosed with laboratory confirmed CDI in the city of Edinburgh, East Lothian and Midlothian regions of Scotland between August 2010 and July 2011 were followed up for one year after diagnosis. Clinical details and laboratory markers were recorded. Stool samples were tested for C. difficile, other bacterial pathogens and norovirus. Molecular epidemiology of C. difficile isolates was studied by PCR-ribotyping.

Of the total 335 confirmed CDI cases, PCR-ribotype 001 was the commonest (14.1%), followed by PCR-ribotypes 078 (12.9%) and 015 (11.7%), respectively. CA-CDI represented 12.5% of the cases. In these, PCR-ribotype 078 was the commonest (19.0%), followed by PCR-ribotypes 014/020 (16.7%), PCR-ribotype 015 (14.3% and PCR-ribotype 001 (11.9%). A lower Charlson co-morbidity index and a lower age was observed in the CA-CDI group as was total number of different antibiotic classes whereas age >75 was more common in the HA-CDI group. On multivariable analysis presence of PCR-ribotype 078 was significantly associated with community acquisition (p = 0.006) whereas a greater proportion of immunosuppressed patients and those on antibiotics 8 weeks preceding diagnosis (p = 0.035 and p = 0.005 respectively) were found among HA-CDI cases. Charlson co-morbidity index, number of different antibiotics given in the eight weeks preceding onset, severity of infection and rural residence were not significantly different between the two groups.

This study demonstrates that patients with CA-CDI may also present with severe infection, are less likely to receive antibiotics prior to CDI, more likely to be younger in age and have a greater proportion of PCR-ribotype 078 compared with CDI acquired in a hospital setting. Hence a high level of vigilance must be maintained to detect CDI cases which present in the community without the traditional predisposing factors.