English hospitals under-report surgical site infections, survey shows

BMJ 2013; 346 doi: (Published 16 January 2013) Cite this as: BMJ 2013;346:f345

Re: English hospitals under-report surgical site infections, survey shows

We thank Lamagini and colleagues for their interest in our paper. These authors from the HPA claim that we are misinformed and lacking in understanding. Yet, our criticisms are the same as those expressed by the Public Accounts Committee and the DH Advisory Committee on HCAIs. Even the European Centers for Disease Control says the English SSI surveillance system ‘lags’ behind the rest of Europe.

The SSI surveillance data published by the HPA does not include post discharge surveillance (save for readmission data in the mandatory scheme) which account up to 80% of SSIs. This results in the ‘true’ scale of SSIs being hugely under reported. As length of stay after surgery continues to fall this becomes ever more important. An SSI surveillance system which does not include post discharge surveillance is akin to describing the size of iceberg by measuring only the part seen above the water.

Regarding inconsistencies in data collection methods, in our study which included 106/158 NHS acute trusts, 10% of trusts did not include superficial infections, 8% of trusts only submitted in-patient data (as opposed to in-patient and readmission as per HPA protocol) and 15% of trusts did not use the HPA definition of an SSI. Thirty trusts did not submit all the data they collected to the HPA, choosing not to submit post discharge data, non-mandatory data or submitting only a minimum of 3 months data when continuous data had been collected. Inconsistencies with the HPA programme were also highlighted in 2009 by the DH Advisory Committee on HCAIs which recommended the development of a nationally agreed methodology for case ascertainment after discharge; however, little progress on this has been made. While the HPA is confident that the majority of trusts follow the HPA protocol it would be interesting to be shown the basis of this confidence.

The confidence of the HPA in their current system will be no comfort to trusts, or individual surgeons and their teams, which carry out high quality post-discharge surveillance in good faith and are branded as outliers or suffer financially through having ‘higher’ rates than other trusts.

Given that SSI rates are publicly reported they should reflect as accurate a picture of the true extent of SSIs as possible, to ensure public confidence. This is the same for commissioners who take published rates of infection into account when deciding where to place contracts.

There is no doubt that there is an appetite for surveillance amongst trusts responding to the survey (the like of which has never been carried out by the HPA). We feel that the HPA should undertake a consultation with users of the system in order that inconsistencies can be minimised and so that the published data can be viewed and interpreted with confidence. A new initiative to undertake surveillance of infections in critical care areas is being led by a steering group comprising professional societies with an interest in this field. We feel that adopting a similar approach with surgical colleagues, and other specialist societies with an interest in this area, would greatly enhance the confidence of the users of the current surveillance systems and foster engagement with stakeholders.

Competing interests: No competing interests

21 January 2013
Judith Tanner
Professor of Clinical Research
Martin Kiernan, David Leaper, Peter Norrie, Wendy Padley
De Montfort University
The Gateway, Leicester, LE1 9BH