Intended for healthcare professionals

CCBYNC Open access

Rapid response to:


Evaluation of the national Cleanyourhands campaign to reduce Staphylococcus aureus bacteraemia and Clostridium difficile infection in hospitals in England and Wales by improved hand hygiene: four year, prospective, ecological, interrupted time series study

BMJ 2012; 344 doi: (Published 03 May 2012) Cite this as: BMJ 2012;344:e3005

Rapid Response:

We note that the results of Stone et al. are consistent with mathematical models of MRSA transmission within hospitals. This can be demonstrated using the framework suggested by Cooper et al (Proc Natl Acad Sci USA 2004; 101: 10223-8). In the model, the transmission within hospital is described by an effective contact rate between susceptible and infectious individuals. Hence improved hygiene conditions correspond to a reduction in the probability that contact between individuals results in transmission and hence a reduction in the contact parameter.

In the figure below, we have plotted the expected prevalence (number of bacteraemia cases per 10,000 bed days) as a function of changes in the contact parameter, for the long-term equilibrium of the system. The model predicts a notable sensitivity response in MRSA rates to small reductions in contact rate, suggesting that relatively small reductions in transmission, via programmes such as hand hygiene programmes, can impact considerably on overall bacteraemia rates.

The delay in the efficacy of the programme described by Stone et al is not readily explained by the long term behaviour of the model alone, as the relationship between transmission reduction and equilibrium prevalence is approximately linear (for those parameters of relevance here), and as noted in the paper, there may have been significantly increased compliance with hygiene measures in the last year of the study only. However, the delay could also possibly be due to transition time from one endemic equilibrium to another. This time depends on the contact parameter, and the state of the system at the onset of changes in hygiene measures.

Simulations of the model show that this is dependent on the initial conditions (i.e. how close to an equilibrium the system was in prior to the intervention) as well as the magnitude of changes in the effective contact rate, but time periods in the order of many months or several years can result under a range of reasonable scenarios.

Competing interests: No competing interests

07 June 2012
Yasmin Friedmann
Research assistant
Institute of Life Sciences, Swansea University
Swansea SA2 8PP