Patricia Garnett
Principal Clinical Microbiologist, Department of Microbiology and Infection Control, Fife Area Laboratory, Victoria Hospital, Kirkcaldy, UK
Healthcare associated infections (HAIs) are both hugely costly and potentially fatal. A conservative estimate of the annual costs to the NHS in the UK is £1 billion.1 In Scotland, annual cost has been estimated at £183m, with an incidence of 9.5% in acute hospitals and 7.3% in community hospitals.2
In 2001, on the recommendation of the Advisory Group on Infection’s Subgroup on Surveillance of Healthcare Associated Infection, a surveillance system was set up to monitor outbreaks in Scotland.2
In January 2003, the Scottish Government HAI Task Force was formed with the remit to; co-ordinate the implementation of the HAI action plan and report on progress to the Health Secretary. The HAI Task Force has completed two programmes, principally to develop new approaches, improve standards and equip NHS boards with the right tools to tackle the root causes of HAI.
The delivery plan for the third programme to be undertaken 2008–2011 includes:
- Continued development of surgical site infection (SSI) surveillance
- Further development of catheter-associated urinary tract infection surveillance, in particular targeting medicine, surgical (and orthopaedics), care of the elderly and in long-term care
- Further exploration as to reasons for variation in Staphylococcus aureus bacteraemia (SAB) rates
- Implementation of intensive care unit surveillance of ventilator-associated pneumonia and central venous catheter infections
- Vascular catheter surveillance
- Surveillance of antimicrobial resistance and use of drugs
- Further development of bacteraemia surveillance (ECOSS interface)
- Develop a prevalence methodology for pressure ulcers
- Explore care home surveillance.3
A user’s experience
The county of Fife on the east coast of Scotland comprises rural areas, industrial towns, post-industrial and fishing communities and tourist landmarks such as St Andrew’s, both the home of golf and of Scotland’s oldest university. NHS Fife – the health board – serves a population of 360,000 via two acute hospitals, a maternity and several community hospitals and three community health partnerships (CHPs).
Its infection control and microbiology departments work together to carry out ongoing surveillance of HAIs as directed by the HAI Task Force and for additional surveillance as appropriate to the population served. The infection control department includes infection control nurses, infection control surveillance and audit nurses and the methicillin-resistant S. aureus (MRSA) screening project team.
Prior to the acquisition of ICNet infection control case management and surveillance software in June 2006, our surveillance of ‘alert’ organisms (MRSA, S. pyogenes, salmonellae species (spp), C. difficile, campylobacter spp, norovirus) was paper-based. This meant it was dependent upon an infection control nurse (ICN) visiting the laboratory for the midday meeting and collecting details of all ‘alert’ organisms reported that day.
ICN colleagues responsible for the hospitals or CHPs with the infected patients were then contacted and further phone calls or visits would ensue; all information being recorded in paper format and filed. Reports based on these files were compiled manually as required for Health Protection Scotland (HPS) and health board statistics.
Infection surveillance nurses collated the results for SSIs and C. difficile, manually producing statistical process control (SPC) charts for distribution to the wards and monthly reports for HPS. The system worked, but report collection was once daily with late results phoned to the ICN in the afternoon. At weekends, there were no meetings, results had to be phoned out, meaning more work and time commitments for both laboratory and ICN staff.
Process greatly simplified
The high profile of HAIs among politicians and the public led to a local process evaluation being undertaken, as a result of which funding was provided for the purchase of the ICNet infection control case management and surveillance software, implementation of which has greatly simplified our surveillance management.
Our first experience of ICNet was in June 2006 with the purchase of version 5 which was interfaced with the laboratory information management system (LIMS) provided by Clinisys, and also with the patient administration system (PAS) OASIS.
Since then, we have regularly upgraded the ICNet system and are currently using version 6.1 which allows us to automate and customise tags for first isolations of MRSA, vancomycin resistant enterococci (VREs), extended spectrum ß-lactamases (ESBLs) and TB. It has the added advantage of the Labstore Module storing all imported microbiology results and enabling us to use additional reports such as antibiograms and MRSA admission screening by selecting the results of interest using the Labsift module.
Our IT department is currently working with both ICNet and OASIS to provide full interoperability and patient visibility down to ward and bay level by bi-directional importing of information. Two imports of laboratory results and one import of patient demographics and admission, discharge and transfer (ADT) updates from the PAS are received daily.
We restricted our initial surveillance to MRSA until users had familiarised themselves with the system, adding all the other common pathogens as everyone became more confident.
The system was set up to show daily organisms imports and also occurrences of these organisms in the ‘alert’ section of the ICNet import sidebar. We selected wards, time periods and numbers of occurrences to trigger the alert and, using the integrated PAS, have been able to include re-admissions of known MRSA, ESBL, VRE and TB cases from whom a new isolation had not occurred during the current admission. This has enabled these patients to be monitored by any ICN anywhere within the NHS Fife computer network at any time of the day, ensuring each case is appropriately managed and the potential for cross-infection reduced.
Recently, recording of results from the government-funded MRSA screening project has been transferred from an independent database to ICNet so that, at the close of the project in February 2011, the infection control team can continue surveillance of pre-admission screening results using a menu of associated questions designed to replace the project paper questionnaires.
We are now looking to include surveillance for carbapenamase-resistant enterobacteriaceae, following recent reports of the isolation of these organisms within the UK. Fortunately, ICNet has proven to be adaptable to new situations and organisms so that we are able to use it to extend our surveillance as and when required.
Patient records
Surveillance has been greatly enhanced by the ability to ‘tag’ patient records in ICNet once an isolate of an organism has occurred. The bidirectional exchange between ICNet and the PAS enables these records to be tagged also.
The ICNet tags are used to ‘recognise’ re-admitted patients with no current microbiology result and alert the ICNs to those patients’ presence in a particular hospital and ward by means of the Alert organism sidebar. Patient tracking is, therefore, more precise and ensures that appropriate infection control procedures can be put in place.
We have embraced the associated feature facility whole-heartedly, as it has enabled us to create a menu of questions associated with each organism that previously required paper records. Once populated from either a drop-down menu or a free text box, these features can be selected to include in reports and are used to analyse case management.
All case notes added by ICNs are immediately available to the rest of the team, so that there is continuity of management without the need to personally hand over paper notes or leave phone messages. The notes are specific to the particular case records and dated and timed on addition so that the flow of information is linear.
As patients are moved rapidly from area to area, it is important to know of cases or contacts of, for example, MRSA, who may have been moved to another ward by the time the result comes through. The PAS interface records the ADT history for each ICNet patient, a great help in following the patient and identifying previous inpatient wards and contacts of the infected case. However, as the PAS relies on manual input of data and information about patient placement to be reported in a timely fashion by the bed bureau, the ADT information is not real time, but certainly better than pre-ICNet days.
We find the software particularly useful for root cause analysis, especially for establishing the source of a case of C. difficile and ascertaining from ICN case notes and imported laboratory patient notes when and what information was passed to the ward, to whom and by whom and that advice on isolation procedures is logged. So we can know whether what should be done has been done, we have slipped up or there is another cause of the problem. ICNet has given us full and rapid traceability – in the past, that would have involved going through a huge amount of paperwork and taken considerable time.
Compiling reports
Surveillance requires the results to be collected, analysed and disseminated, a remit fulfilled both easily and rapidly by ICNet’s reports. They are very quick to build and save and can be exported to spreadsheet software for analysis or produced as SPC charts. Every month, each ward receives a SPC chart on its C.difficile infections as one of the initiatives to enable the wards to monitor and control the potential problem areas for C. difficile acquisition and cross-infection.
We use the bacteraemia report for collection of data on SABs for the HAI Task Force and also on a weekly basis to identify positive cultures in the renal unit that is linked in with haemodialysis line surveillance. In the event of a positive culture, checks can be made to determine whether the line has been removed or if a particular organism has caused the infection.
We build most reports on organism isolations, refine them to particular wards, hospitals, specimen types, and add in associated features from the patient record, such as whether or not the isolate is considered to be an HAI. Saved reports are recovered at the click of a mouse button and give a rolling update, but are simple to alter to a time period of interest by resetting the date range.
Our ICD sets up routine reports for cases of C.difficile and new cases of MRSA to check that we are keeping control on the numbers of these specific infections. We should be, but are we?
The ICNs build contact tracing reports to identify all other patients with whom an infected patient has come into contact, and a benefit of the PAS interface is that all contact patients, whether imported as a case into ICNet or not, will be included in this report.
Outbreak Management
Currently, we are trialling the use of ICNet for outbreak management with the intention of reducing paper records, containing all the outbreak information in one easily accessible database, and producing standard reports.
A panel of questions about the patient’s history relevant to the outbreak condition are included in the patient record and can be searched when building an outbreak report and included in report for analysis on spreadsheet software. The status of an outbreak can be included in each report and updated daily.
We are hoping that the trial will show the benefits of moving from a paper-based to an electronic record, but there are hiccups at weekends and on call when access to ICNet on a home computer is not available.
Training
The help menu on ICNet has been very good – it’s sensibly intuitive and we have now set up our own Fife user group to get feedback, to see how we are using it and to brainstorm ideas for what we would like to have added on. We know this is worthwhile, as ideas and change requests have been taken on board by ICNet and dealt with by continuous updates and improvements.
NHS Fife ICT is delighted at the results that have been achieved using ICNet and with the level of control and surveillance of infection that is now possible due to the accessibility of the data. We couldn’t work without it now.
References
- C&AG’s Report: Four Country Healthcare Associated Infection Prevalence Survey 2006. Journal of Hospital Infection 2008, 1-19; and results from Scottish National HAI Prevalence Survey. Journal of Hospital Infection 2008;69:62-68.
- House of Commons’ Public Accounts Committee: Reducing Healthcare Associated Infection in Hospitals in England, 52nd Report of Session 2008-09, p13, published 10 November 2009.
- www.who.int/surveill/index.html
