The study was supported by the National Institute of Health Research Oxford Biomedical Research Centre -- a collaboration between Oxford University Hospitals NHS Trust and Oxford University.

The team, led by Tim Peto and Rosalind Harding at the University of Oxford, sequenced the genomes of C. difficile isolated from 486 patients treated at four hospitals in Oxfordshire between 2006 and 2010. By counting the number of genetic differences between different isolates and estimating the mutation rate of the bacteria, the researchers were able to determine the likely time at which any two isolates became genetically separate and thus, whether the two patients in question could have plausibly caught the infection from each other in the hospital. In other words, genetic divergence implies a time-scale that can be used for judging the likelihood of direct transmission.

The results of the study indicated that, although transmission between patients is likely to occur, it actually happens at relatively low frequency. In particular, concerns that healthcare teams were spreading infection between different hospitals seem to be misplaced. One exception to this general finding is that there were a large number of cases of infection from one particular strain that does appear to have been due to patient-to-patient transmission, emphasising the epidemic nature of this lineage. Notably, this strain has declined in UK hospitals in the last five years.

Dr Xavier Didelot, the study's lead author, said: " This research opens up very exciting opportunities for better understanding how bacterial infections are spread, and what we can do to stop them. The reduced cost of sequencing whole bacterial genomes means we now have the technology for identifying very recent transmissions of infection. Moreover, we can apply this technology even in cases when infection control teams have no suspicion that routes of contact between patients might exist."