Improving surveillance of MRSA bacteraemia
http://www.100md.com
《英国医生杂志》
Should focus more on patients bringing strains to hospital on readmission
Two papers in this week's BMJ consider from different perspectives the limitations of England's mandatory surveillance system for methicillin resistant Staphylococcus aureus (MRSA) bacteraemia.1 2 This surveillance became compulsory in April 2001 in response to the rise in MRSA bacteraemias from less than 2% of all S aureus bacteraemias in 1990 to 42% in 2000.3 The Department of Health publishes the results for each English NHS acute trust every six months.4 In 2004 the secretary of state for health announced that these infections would be halved by 2008 and monitored by the Healthcare Commission.5
The paper by Wyllie and colleagues focuses on the high proportion of MRSA bacteraemias among patients on admission to hospital.1 These bacteraemias were not necessarily acquired in the community, however, since nearly all these patients had previously stayed in hospital, where they could have been exposed to MRSA and some were known to have had MRSA on a previous admission. Indeed, in England most MRSA infections among patients entering hospital are caused by strains that have been exported from hospitals and have created a reservoir in community settings such as residential care homes.6-8 These are distinct from the true community acquired MRSA strains, which have caused serious infections.9
Successful control of the spread of infection depends on having, on average, less than one secondary case arising from each case of MRSA infection. As asymptomatic MRSA carriage may be prolonged, reintroduction from the community is an important factor as secondary cases may span several hospital admissions. We need better understanding of MRSA dynamics inside and outside hospitals to inform future guidance, particularly on whether control of MRSA should span the hospital and its catchment area.6 The high proportion of MRSA bacteraemias that are detectable on entry to hospital has implications for admitting doctors, who need to raise their index of suspicion and treat infected patients accordingly. This situation complicates decisions on empiric treatment: automatically selecting treatment that covers more resistant infections will benefit some patients but will increase the chances of engendering yet more resistant organisms.
Interventions to control MRSA on entry to hospital are also up for debate, such as systematic screening on admission and the use of rapid diagnostic tests as tools in maintaining MRSA-free zones.10 Such tools are unlikely to succeed in the long run if hospitals have insufficient capacity for isolating infected patients and nobody pays enough attention to reservoirs of infection in the community.11 Moreover, competing pressures on the NHS will have to be balanced, so that targets to reduce waiting lists do not adversely affect infection control.
This week's article by Spiegelhalter discusses the use and interpretation of the English surveillance data.2 The author highlights many of the pitfalls of using routine surveillance data to monitor whether a target has been met. Generally, the purpose of public health surveillance is to detect and describe problems, initiate investigation, and suggest hypotheses. The quantity of data, even when obtained through enhanced surveillance, cannot be exhaustive because it is collected by healthcare staff as part of their routine duties. This necessarily introduces limitations and prohibits superficial interpretation of the results.
Spiegelhalter describes elegantly the limitations of such surveillance data, which are often poorly understood and which can allow hospitals to suffer at the hands of the mass media. Although there is undoubtedly some "overdispersion" within the data on MRSA bacteraemia, this infection is unlike many other transmissible infections and much of the overdispersion results from variation due to unmeasured risk factors, such as the specialty under whose care a patient is in hospital (as a proxy for the invasiveness of medical care received). Analysis of data from a voluntary surveillance scheme for bacteraemia 12 shows that there is no overdispersion after adjusting for specialty (Pearson 2 = 6119.5, degrees of freedom = 6011). The inability to remove this additional variation due to risk factors from routine surveillance data also complicates the detection of "problem" hospitals, because the overdispersion factor also includes variation introduced by poor performance.
Despite these limitations, mandatory surveillance of MRSA infection rates has raised the profile of infection control. Infection control is now an essential element of the clinical governance process, with surveillance guiding quality improvement. The paper by Wyllie and colleagues emphasises the importance of certain hospital units as foci of MRSA.1 In any hospital the data should be used to focus investigation to identify what allows MRSA to flourish in affected units. For example, lack of timely access to surgery to create arteriovenous fistulae for dialysis, resulting in prolonged use of intravenous lines, is a risk factor in renal units.13
Some of the limitations discussed here have already been removed through enhancements to the surveillance system in England. The mandatory dataset now includes information on where bacteraemia was acquired and gives dates of admission and infection. Continuing problems with information technology mean, however, that NHS trusts' infection control teams are hampered in collating and manipulating data pertinent to infection control. Improving such systems should be a priority for the national programme for IT (NPfIT), delivered by Connecting for Health.
In the meantime teams need interim solutions, possibly based on the linkage technology described by Wyllie and colleagues.1 We also need robust multicentre studies to assess the efficacy of interventions. Lastly, it is time to turn the spotlight on bacteraemias caused by methicillin sensitive Staphylococcus aureus, whose prevalence is also rising.
Georgia Duckworth, director, department of healthcare-associated infection and antimicrobial resistance
(georgia.duckworth@hpa.org.uk)
Health Protection Agency Centre for Infections, London NW9 5EQ.
Andr Charlett, interim director, statistics, modelling, and bioinformatics department
Health Protection Agency Centre for Infections, London NW9 5EQ.
Papers p 992 and Education and debate p 1013
Competing interests: GD has been involved in implementing the mandatory surveillance as required by the Department of Health. GD and AC have been involved in analyses of the data and in recent enhancements of the surveillance system.
References
Wyllie DH, Peto TEA, Crook D. MRSA bacteraemia in patients on arrival in hospital: a cohort study in Oxfordshire 1997-2003. BMJ 2005;331: 992-5.
Spiegelhalter DJ. Problems in assessing rates of infection with methicillin resistant Staphylococcus aureus. BMJ 2005;331: 1013-5.
Department of Health. All hospitals to monitor hospital acquired infection . www.dh.gov.uk/PublicationsAndStatistics/PressReleases/PressReleasesNotices/fs/en?CONTENT_ID=4007286&chk=rHpzGU (accessed 24 Oct 2005).
Department of Health. MRSA surveillance system: results . www.dh.gov.uk/PublicationsAndStatistics/Publications/PublicationsStatistics/Publications StatisticsArticle/fs/en?CONTENT_ID=4085951&chk=HBt2QD (accessed 20 Oct 2005).
Department of Health. Bloodborne MRSA infection rates to be halved by 2008—Reid . www.dh.gov.uk/PublicationsAndStatistics/PressReleases/PressReleasesNotices/fs/en?CONTENT_ID=4093533&chk=MY%2BkD/ (accessed 24 Oct 2005).
Cox RA, Mallaghan C, Conquest C, King J. Epidemic methicillin-resistant Staphylococcus aureus: controlling the spread outside hospital. J Hosp Infect 1995;29: 107-19.
Grundmann H, Tami A, Hori S, Halwani M, Slack R. Nottingham Staphylococcus aureus population study: prevalence of MRSA among elderly people in the community. BMJ 2002;324: 1365-6.
Maudsley J, Stone SP, Kibbler CC, Iliffe SR, Conaty SJ, Cookson BD, et al. The community prevalence of methicillin-resistant Staphylococcus aureus (MRSA) in older people living in their own homes: implications for treatment, screening and surveillance in the UK. J Hosp Infect 2004;57: 258-62.
Community MRSA in England and Wales: definition through strain characterisation. Commun Dis Rep CDR Wkly 2005;15(11). www.hpa.org.uk/cdr/archives/archive05/News/news1105.htm#CMRSA (accessed 20 Oct 2005).
Biant LC, Teare EL, Williams WW, Tuite JD. Eradication of methicillin resistant Staphylococcus aureus by "ring fencing" of elective orthopaedic beds. BMJ 2004;329: 149-51.
Cooper BS, Medley GF, Stone SP, Kibbler CC, Cookson BD, Roberts JA, et al. Methicillin-resistant Staphylococcus aureus in hospitals and the community: stealth dynamics and control catastrophes. Proc Natl Acad Sci USA 2004;101: 10223-8.
Coello R, Charlett A, Ward V, Wilson J, Pearson A, Sedgwick J, et al. Device-related sources of bacteraemia in English hospitals—opportunities for the prevention of hospital-acquired bacteraemia. J Hosp Infect 2003;53: 46-57.
Department of Health. MRSA—Learning from the Best conference. www.dh.gov.uk/NewsHome/ConferenceAndEventReports/ConferenceReportsConference ReportsArticle/fs/en?CONTENT_ID=4102049&chk=GgHAEH (accessed 20 Oct 2005).
Two papers in this week's BMJ consider from different perspectives the limitations of England's mandatory surveillance system for methicillin resistant Staphylococcus aureus (MRSA) bacteraemia.1 2 This surveillance became compulsory in April 2001 in response to the rise in MRSA bacteraemias from less than 2% of all S aureus bacteraemias in 1990 to 42% in 2000.3 The Department of Health publishes the results for each English NHS acute trust every six months.4 In 2004 the secretary of state for health announced that these infections would be halved by 2008 and monitored by the Healthcare Commission.5
The paper by Wyllie and colleagues focuses on the high proportion of MRSA bacteraemias among patients on admission to hospital.1 These bacteraemias were not necessarily acquired in the community, however, since nearly all these patients had previously stayed in hospital, where they could have been exposed to MRSA and some were known to have had MRSA on a previous admission. Indeed, in England most MRSA infections among patients entering hospital are caused by strains that have been exported from hospitals and have created a reservoir in community settings such as residential care homes.6-8 These are distinct from the true community acquired MRSA strains, which have caused serious infections.9
Successful control of the spread of infection depends on having, on average, less than one secondary case arising from each case of MRSA infection. As asymptomatic MRSA carriage may be prolonged, reintroduction from the community is an important factor as secondary cases may span several hospital admissions. We need better understanding of MRSA dynamics inside and outside hospitals to inform future guidance, particularly on whether control of MRSA should span the hospital and its catchment area.6 The high proportion of MRSA bacteraemias that are detectable on entry to hospital has implications for admitting doctors, who need to raise their index of suspicion and treat infected patients accordingly. This situation complicates decisions on empiric treatment: automatically selecting treatment that covers more resistant infections will benefit some patients but will increase the chances of engendering yet more resistant organisms.
Interventions to control MRSA on entry to hospital are also up for debate, such as systematic screening on admission and the use of rapid diagnostic tests as tools in maintaining MRSA-free zones.10 Such tools are unlikely to succeed in the long run if hospitals have insufficient capacity for isolating infected patients and nobody pays enough attention to reservoirs of infection in the community.11 Moreover, competing pressures on the NHS will have to be balanced, so that targets to reduce waiting lists do not adversely affect infection control.
This week's article by Spiegelhalter discusses the use and interpretation of the English surveillance data.2 The author highlights many of the pitfalls of using routine surveillance data to monitor whether a target has been met. Generally, the purpose of public health surveillance is to detect and describe problems, initiate investigation, and suggest hypotheses. The quantity of data, even when obtained through enhanced surveillance, cannot be exhaustive because it is collected by healthcare staff as part of their routine duties. This necessarily introduces limitations and prohibits superficial interpretation of the results.
Spiegelhalter describes elegantly the limitations of such surveillance data, which are often poorly understood and which can allow hospitals to suffer at the hands of the mass media. Although there is undoubtedly some "overdispersion" within the data on MRSA bacteraemia, this infection is unlike many other transmissible infections and much of the overdispersion results from variation due to unmeasured risk factors, such as the specialty under whose care a patient is in hospital (as a proxy for the invasiveness of medical care received). Analysis of data from a voluntary surveillance scheme for bacteraemia 12 shows that there is no overdispersion after adjusting for specialty (Pearson 2 = 6119.5, degrees of freedom = 6011). The inability to remove this additional variation due to risk factors from routine surveillance data also complicates the detection of "problem" hospitals, because the overdispersion factor also includes variation introduced by poor performance.
Despite these limitations, mandatory surveillance of MRSA infection rates has raised the profile of infection control. Infection control is now an essential element of the clinical governance process, with surveillance guiding quality improvement. The paper by Wyllie and colleagues emphasises the importance of certain hospital units as foci of MRSA.1 In any hospital the data should be used to focus investigation to identify what allows MRSA to flourish in affected units. For example, lack of timely access to surgery to create arteriovenous fistulae for dialysis, resulting in prolonged use of intravenous lines, is a risk factor in renal units.13
Some of the limitations discussed here have already been removed through enhancements to the surveillance system in England. The mandatory dataset now includes information on where bacteraemia was acquired and gives dates of admission and infection. Continuing problems with information technology mean, however, that NHS trusts' infection control teams are hampered in collating and manipulating data pertinent to infection control. Improving such systems should be a priority for the national programme for IT (NPfIT), delivered by Connecting for Health.
In the meantime teams need interim solutions, possibly based on the linkage technology described by Wyllie and colleagues.1 We also need robust multicentre studies to assess the efficacy of interventions. Lastly, it is time to turn the spotlight on bacteraemias caused by methicillin sensitive Staphylococcus aureus, whose prevalence is also rising.
Georgia Duckworth, director, department of healthcare-associated infection and antimicrobial resistance
(georgia.duckworth@hpa.org.uk)
Health Protection Agency Centre for Infections, London NW9 5EQ.
Andr Charlett, interim director, statistics, modelling, and bioinformatics department
Health Protection Agency Centre for Infections, London NW9 5EQ.
Papers p 992 and Education and debate p 1013
Competing interests: GD has been involved in implementing the mandatory surveillance as required by the Department of Health. GD and AC have been involved in analyses of the data and in recent enhancements of the surveillance system.
References
Wyllie DH, Peto TEA, Crook D. MRSA bacteraemia in patients on arrival in hospital: a cohort study in Oxfordshire 1997-2003. BMJ 2005;331: 992-5.
Spiegelhalter DJ. Problems in assessing rates of infection with methicillin resistant Staphylococcus aureus. BMJ 2005;331: 1013-5.
Department of Health. All hospitals to monitor hospital acquired infection . www.dh.gov.uk/PublicationsAndStatistics/PressReleases/PressReleasesNotices/fs/en?CONTENT_ID=4007286&chk=rHpzGU (accessed 24 Oct 2005).
Department of Health. MRSA surveillance system: results . www.dh.gov.uk/PublicationsAndStatistics/Publications/PublicationsStatistics/Publications StatisticsArticle/fs/en?CONTENT_ID=4085951&chk=HBt2QD (accessed 20 Oct 2005).
Department of Health. Bloodborne MRSA infection rates to be halved by 2008—Reid . www.dh.gov.uk/PublicationsAndStatistics/PressReleases/PressReleasesNotices/fs/en?CONTENT_ID=4093533&chk=MY%2BkD/ (accessed 24 Oct 2005).
Cox RA, Mallaghan C, Conquest C, King J. Epidemic methicillin-resistant Staphylococcus aureus: controlling the spread outside hospital. J Hosp Infect 1995;29: 107-19.
Grundmann H, Tami A, Hori S, Halwani M, Slack R. Nottingham Staphylococcus aureus population study: prevalence of MRSA among elderly people in the community. BMJ 2002;324: 1365-6.
Maudsley J, Stone SP, Kibbler CC, Iliffe SR, Conaty SJ, Cookson BD, et al. The community prevalence of methicillin-resistant Staphylococcus aureus (MRSA) in older people living in their own homes: implications for treatment, screening and surveillance in the UK. J Hosp Infect 2004;57: 258-62.
Community MRSA in England and Wales: definition through strain characterisation. Commun Dis Rep CDR Wkly 2005;15(11). www.hpa.org.uk/cdr/archives/archive05/News/news1105.htm#CMRSA (accessed 20 Oct 2005).
Biant LC, Teare EL, Williams WW, Tuite JD. Eradication of methicillin resistant Staphylococcus aureus by "ring fencing" of elective orthopaedic beds. BMJ 2004;329: 149-51.
Cooper BS, Medley GF, Stone SP, Kibbler CC, Cookson BD, Roberts JA, et al. Methicillin-resistant Staphylococcus aureus in hospitals and the community: stealth dynamics and control catastrophes. Proc Natl Acad Sci USA 2004;101: 10223-8.
Coello R, Charlett A, Ward V, Wilson J, Pearson A, Sedgwick J, et al. Device-related sources of bacteraemia in English hospitals—opportunities for the prevention of hospital-acquired bacteraemia. J Hosp Infect 2003;53: 46-57.
Department of Health. MRSA—Learning from the Best conference. www.dh.gov.uk/NewsHome/ConferenceAndEventReports/ConferenceReportsConference ReportsArticle/fs/en?CONTENT_ID=4102049&chk=GgHAEH (accessed 20 Oct 2005).