Assessing data sources for sustainable and continuous surveillance: surgical site infections following coronary artery bypass grafts in England

https://doi.org/10.1016/j.jhin.2013.05.004Get rights and content

Summary

Background

Multiple national and local data sources collected by healthcare systems across Europe contain information regarding coronary artery bypass grafts (CABG). Currently surveillance for surgical site infections (SSIs) in this patient group is voluntary in the UK.

Aim

To investigate and compare the currently available data sources in England for SSI surveillance in CABG patients.

Methods

Data were extracted from the Society for Cardiothoracic Surgery in Great Britain and Ireland (SCTS) Adult Cardiac Surgery Registry, the Health Protection Agency SSI surveillance, and the Patient Administration System (PAS) from a single large National Health Service Trust. These data were deterministically linked using patient identifiers, anonymized, and then underwent descriptive analysis.

Findings

From 1 January 2011 to 30 June 2011, 306 patients were recorded in at least one dataset as having undergone CABG. Of these, 76% were recorded by all three data sources being investigated. Of the discordant patients, 5% were recorded by both the local cardiac registry and PAS, and 18% by both the HPA surveillance and PAS. A total of 28 surgical site infections were recorded, of which 21% were identified by all the data sources.

Conclusion

Currently, the databases which collect and store data relating to CABG patients suffer from duplications and discrepancies. Integration of the discordant systems is recommended to create a streamlined, sustainable electronic surveillance system.

Introduction

Coronary artery bypass graft (CABG) remains one of the most frequently performed surgical procedures: 114,300 in England between 2003 and 2008.1 Post-surgical wound infection rates for this procedure range from 0.49% to 18.8%, with the surgical site infection (SSI) rate in the UK being 4.4%.2, 3, 4, 5, 6

Surgical site infection surveillance is routinely carried out across Europe and the USA.7, 8 In the UK, certain orthopaedic procedures have mandatory surveillance with public reporting, but for CABG the surveillance is voluntary.9 In its current form, the surveillance data are collected in a paper format and periodically submitted to the Health Protection Agency (HPA) for cleaning, processing and dissemination. This system entails a large amount of time and effort, usually from the infection prevention and control team, and has a long time delay between data collection and feedback.

The Society for Cardiothoracic Surgery in Great Britain and Ireland (SCTS) Adult Cardiac Surgery Registry in the UK collects copious amounts of information about CABG surgeries. Despite submission of data to the SCTS Registry being close to comprehensive, it has been acknowledged that reporting of postoperative complications can have up to 15% missing data.10 Reoperation, for which one of the indications is a deep sternal wound infection, showed an SSI incidence of 0.5% in 2008, although this variable had 14% missing data.1

The UK is not alone in having a long-running, widely used cardiac registry and national initiatives for SSI surveillance. The National Cardiovascular Data Registry in the USA has been established since 1997 (www.ncdr.com/webncdr/), considerably shorter than the UK register maintained by the Society for Cardiothoracic Surgeons since 1977 (www.ic.nhs.uk). According to Taylor, Sweden and the UK are the only two countries worldwide with all centres reporting to a national cardiology registry.11 The Swedish register is now completely electronic and has been linked to other national data sources, allowing for long-term patient follow-up.11 This registry, like that in the UK, collects data on postoperative complications and specifically reoperation for mediastinitis.12, 13

Hospitals consistently collect and store a wealth of administrative data, including information on admissions, microbiology, diagnoses and procedures.14 Even though the data have not been widely exploited for the support of healthcare-associated infection surveillance, integrative approaches towards the more effective use of administrative data are being pioneered.14, 15, 16 The Patient Administration System (PAS) forms the backbone of administrative data in UK hospitals, with patient's clinical notes converted into coded data.

Utilizing SSI surveillance data, alongside cardiac registry data and administrative hospital data, allows a unique opportunity to investigate these data sources for sustainable SSI surveillance. The aim of the study was to assess the current data sources available within the English National Health Service (NHS) which relate to CABG patients, to evaluate their potential for SSI surveillance.

Section snippets

Methods

This study was carried out by Imperial College NHS Healthcare Trust (ICNHT), a group of hospitals with 1540 beds. Ethical approval for the use of linked anonymized local patient data for research was granted by St Mary's Research Ethics Committee (REC: 09/H0712/85).

Three data sources were used in this study: the SCTS Registry, the voluntary HPA SSI surveillance data, and local administrative hospital data from the PAS.

The HPA SSI surveillance data were collected weekly by infection prevention

Results

In all, 306 patients were recorded as having undergone CABG surgeries between 1 January 2011 and 30 June 2011. Of these, 76% (232) were recorded by the three databases investigated. Figure 1 demonstrates how many patients were identified by the different databases and how they overlap. The PAS documented the most patients, with 303 of the 306 patients having a record. The fewest patients were recorded by the local cardiac registry, which contained 249 patients (82%).

The variables collected by

Discussion

Creating a dataset by linking the standardized HPA surveillance and cardiac registry data has highlighted the overlap in data from these two sources, along with the discrepancies in data recorded. The duplication in data between the cardiac registry and HPA surveillance data relates not only to the basic administrative information, but to surgical features and SSIs. The comparison of these purposeful, traditional data sources with the administrative data demonstrates the potential

Acknowledgements

We thank the IPC nurses, especially I. McCabe and C. Perez, for co-ordinating and collecting the HPA surveillance data and Y. Abdullahi for his role in data management; the cardiac nurses, specifically G. Bleaze, for supporting surveillance activities; and the clinical microbiology department, specifically C. Thomas, for their collaboration and support. We also thank the pathology department and the information technology teams within Imperial College NHS Trust for their collaboration.

References (25)

  • Surveillance of surgical site infections in NHS hospitals in England, 2010/2011

    (2011)
  • R.P. Gaynes et al.

    Surgical site infection (SSI) rates in the United States, 1992–1998: The National Nosocomial Infections Surveillance system basic SSI risk index

    Clin Infect Dis

    (2001)
  • Cited by (3)

    • Mapping national surveillance of surgical site infections in England: needs and priorities

      2018, Journal of Hospital Infection
      Citation Excerpt :

      Automating some aspects of surveillance could help to reduce costs and subjectivity in diagnosis. Several automated or semi-automated surveillance systems have been developed and innovative systems of PDS are also being developed, including an electronic post-discharge questionnaire for the SSISS, which are especially important in surgical categories with a short postoperative length of stay [34–41]. There is growing evidence to support the utility of these systems for SSI surveillance, and growing guidance to support their design and implementation [42,43].

    • The role of antimicrobial sutures in preventing surgical site infection

      2017, Annals of the Royal College of Surgeons of England
    View full text