Skip to main content

Hospitals: Infectious Diseases

Volume 457: debated on Thursday 8 March 2007

To ask the Secretary of State for Health what the evidential basis is for her statement of 23 January 2007, Official Report, column 1305, that local variation in the incidence of Clostridium difficile is greater than local variation in the incidence of meticillin resistant Staphylococcus auerus. (118218)

The statement was made in the context of considering a target for Clostridium difficile and how this differed from the meticillin resistant Staphylococcus aureus (MRSA) target. The MRSA national target was set when mandatory surveillance for this infection was well established. In contrast the position for C difficile is different as mandatory surveillance was only introduced in 2004 and a reliable baseline from which to measure trends is not available yet. In addition, interpretation of C. difficile data is not straightforward as some trusts report cases that occur in the community and other trusts as well as their own cases. After considering these factors and the variation in C. difficile reports in 2006, when rates ranged from over five to less than one per 1,000 bed days we concluded that local targets were the best option as trusts and primary care trusts would be aware of local issues.

To ask the Secretary of State for Health what method the (a) Health Protection Agency and (b) Office of National Statistics use to collect statistics on hospital acquired infections; why the incidence of hospital acquired infections varies between the statistics published by each agency; and if she will make a statement. (124663)

[holding answer 2 March 2007]: Statistics from the Health Protection agency (HPA) and Office for National Statistics (ONS) do not measure the same thing. HPA figures refer to incidence, that is the number of new cases, of infection. Figures from the ONS refer to deaths.

The HPA does not measure the incidence of all healthcare associated infections but collects data on selected healthcare associated infections of significance. All acute NHS trusts in England are obliged to report to the HPA cases in four categories are listed as follows.

Staphylococcus aureus (including meticillin resistant Staphylococcus aureus (MRSA) bacteraemias Clostridium difficile (age 65 and over)

Orthopaedic surgical site infections (SSI)

Glycopeptide resistant enterococci

The MRSA and SSI data are collected via web based systems but most of the other data are provided from routine laboratory reporting. This electronic reporting system also covers infections caused by other micro-organisms and hospital laboratories voluntarily supply these data.

The ONS does not collect information or publish statistics on the incidence (number of new cases) of hospital acquired infection. The ONS produces annual reports on deaths involving MRSA, and on deaths involving Clostridium difficile. The methods ONS uses in preparing these reports is explained fully in the reports themselves. ONS statistics are based on the cause of death information given by doctors and coroners on certificates used to register death and include all deaths that are registered in England and Wales. The cause of death statements from coroners and doctors are coded to the Tenth Revision of the International Classification of Diseases and Related Health Problems (ICD-10), following internationally agreed rules and guidelines. From these data, ONS tabulates the numbers of deaths for which MRSA and/or Clostridium difficile infections were mentioned anywhere on the death certificate, and the numbers for which they were the underlying cause of death. It is not possible to say how many of these infections were acquired in hospital, because this information is not recorded on death certificates. ONS figures therefore represent the actual number of deaths in England and Wales each year with MRSA or Clostridium difficile on the death certificate.

Doctors and coroners should only include on the death certificate conditions which they believe contributed to the death, not all of the conditions present at or immediately before death. Increased concern about health care associated infections, recent outbreaks of a virulent strain of Clostridium difficile and an article in CMO’s update of July 2005 on certifying deaths involving MRSA may have led to increased reporting of these infections on death certificates. It is very difficult to obtain the evidence needed to measure how much of the apparent increase in deaths involving these infections is real and how much may be due to increased reporting


1. The third annual report on MRSA, and the second annual report on C. difficile were published in February 2007.

2. Report: Deaths involving MRSA, England and Wales, 2001-2005. ‘Health Statistics Quarterly’33, 76-81.

3 Report: Deaths involving Clostridium difficile, England and Wales, 2001-2005. ‘Health Statistics Quarterly’33, 71-75.