Journal of Hospital Infection 89 (2015) 271e275 Available online at www.sciencedirect.com

Journal of Hospital Infection journal homepage: www.elsevierhealth.com/journals/jhin

Integrating intensive care unit (ICU) surveillance into an ICU clinical care electronic system J.S. Reilly a, *, J. McCoubrey a, S. Cole b, A. Khan c, B. Cook d a

Health Protection Scotland, NHS National Services Scotland, UK NHS Tayside, Scotland, UK c Information and Statistics Division, NHS National Services Scotland, UK d NHS Lothian, Scotland, UK b

A R T I C L E

I N F O

Article history: Received 30 October 2014 Accepted 17 November 2014 Available online 30 December 2014 Keywords: Hospital infection Intensive care unit Surveillance

S U M M A R Y

The intensive care unit (ICU) is the specialty with the highest prevalence of healthcareassociated infection (HCAI) in European hospitals and therefore a priority for surveillance of HCAI. Whereas surveillance is an essential part of an effective infection prevention and control (IPC) programme, all too often it consumes too much clinician and IPC team time, limiting the time available for quality improvement. The case for electronic surveillance is made in the literature from several countries on this basis. These studies indicate that electronic surveillance can improve validity, reduce time spent on surveillance, and provide opportunities for improvement in clinical decision-making and IPC action arising from surveillance. The Scottish ICU HAI surveillance system was established as part of an integrated audit and clinical care system. Investment in this technology infrastructure reduced the burden of data collection and has resulted in a focus on driving improvement in all Scottish ICUs. The experience in Scotland indicates that several critical components are necessary to optimize ICU HCAI surveillance, including: nationally agreed definitions and methods; national investment in information technology infrastructure to make it easier to follow clinical care pathways; leadership of surveillance by intensivists; piloting and validation to ensure confidence in the system; and strategic integration of national and local programmes. These elements have helped improve surveillance data locally, nationally, and at a European level, allowing clinical attention to be focused on the data rather than on the process of data collection. ª 2014 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved.

Introduction The intensive care unit (ICU) is the specialty with the highest prevalence of healthcare-associated infection (HCAI) in European hospitals.1 Multiple comorbidities and complexity of *Corresponding author. Address: Health Protection Scotland, National Services Scotland, 5 Meridian Court, Glasgow, UK. Tel.: þ44 (0)1413 001162. E-mail address: [email protected] (J.S. Reilly).

healthcare delivery make this a population at particular risk, and therefore a priority for surveillance of HCAI. Whereas surveillance is an essential part of an effective infection prevention and control (IPC) programme, very often it consumes too much of clinician and IPC team time, limiting time available for education and quality improvement activities. Electronic, or automated, surveillance has been proposed as a solution to this. A recent systematic review of electronic surveillance over the last decade identified 44 studies, 21 of which assessed performance of electronic versus traditional (observation- and

http://dx.doi.org/10.1016/j.jhin.2014.11.017 0195-6701/ª 2014 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved.

272

J.S. Reilly et al. / Journal of Hospital Infection 89 (2015) 271e275

paper-based) methods, and concluded that sensitivity and specificity varied by HCAI type.2 Scotland has had electronic surveillance of HCAI in place in all ICUs for four years.3 This paper reviews the case for automation of ICU surveillance of HCAI, describes the Scottish experience in establishing such surveillance, and outlines the critical components of successful integrated surveillance in clinical care.

The case for automation in ICU surveillance Several country- and hospital-level studies have been published specifically examining automated ICU surveillance. These studies have examined the validity of, and opportunities for, improvement in IPC action arising from automated surveillance, in comparison with more traditional manual data collection processes.

Validity and time De Bruin et al. examined the validity and economics of automated surveillance in an ICU in a hospital in Austria.4 This study assessed the validity of a fully automated surveillance system for the detection of HCAI. The results were compared with manual surveillance using paper-based data collection. For HCAI case finding, manual versus electronic surveillance had sensitivities of 40% versus 87%, specificities of 94% versus 99%, positive predictive values of 71% versus 96%, and negative predictive values of 80% versus 95%, respectively. Time spent on manual surveillance was 82.5 versus 12.5 h for electronic. Thus, automated surveillance was found not only to have greater validity but also to require less clinical time. In Belgium, De Bus et al. developed an electronic decision support programme in an ICU that enabled ICU physicians to add clinical information during patient rounds, resulting in prospective compilation of an HCAI database.5 They assessed the validity of this computer-assisted surveillance (CAS) of ICU infection by retrospective review of the content by a specialist panel. The validation study findings did not report sensitivity and specificity of case ascertainment, but conventional paper-based surveillance identified 89 ICU-acquired infections (13 BSI, 18 UTI, 58 RTI) and CAS identified 90 ICU-acquired infections (14 BSI, 17 UTI, 59 RTI) in 876 ICU admissions. There was agreement on a total of 69 infections (77.5%) (kappa score of 0.74), of which 13 were BSI (100%), 14 UTI (77.8 %) and 42 RTI (72.4%). Disagreement occurred for 16 episodes that were predominantly RTI and UTI. Electronic surveillance took 30% less time than manual data collection, so the authors concluded that it was a feasible surveillance method with good validity and time effectiveness. A few studies have also identified opportunity for improvement in IPC practice arising from automated surveillance.

Opportunities for IPC improvement Existing electronic surveillance systems in The Netherlands, based on classification algorithms using microbiology results, antibiotic use data and discharge codes were found to have increased the efficiency and completeness of surveillance by preselecting high-risk patients for manual review by clinicians.6 The benefits of this are twofold: first, making it easy and time effective for data collection for numerator purposes in surveillance; second, prompting the review of the at-risk

patients, providing an opportunity for intervention to reduce risk of HCAI. The possibility of moving beyond automation for HCAI data collection and activities associated with surveillance towards data-driven decision-making for ICU patient care has been explored in Austria.7 The focus of this work was about making it easy for practitioners to do the right thing as a by-product of electronic surveillance. Their HCAI data import interfaces with other healthcare information technology (IT) systems within the hospital, and they used a stepwise pipeline of aggregation and interpretation, for presence or absence of signs and symptoms for HCAI, using fuzzy logic methodologies. This enabled real-time alerts to the intensive care team and clinicians. This might be described as surveillance utopia, wherein the automation of surveillance drives good decision-making at the point of care, rather than retrospective feedback of data for action in more traditional surveillance systems. Electronic surveillance using multivariable prediction models, based on available clinical patient data, will enable better detection of infection in the future. With ongoing developments in healthcare IT, implementation of electronic surveillance systems will become increasingly feasible.2 Some challenges remain, such as IT infrastructure in some hospitals and countries, validity of using records and systems designed for another purpose, and adequate adjustment for underlying risk factors from routine records. Further evidence has been published on electronic capture of post discharge surveillance, which will become increasingly important as healthcare delivery models change over time, and lengths of stay in hospital decrease. The importance of IT infrastructure in supporting IPC is emphasized by the studies detailed in this review thus far, but there are other important elements to consider in ensuring that any surveillance system has the maximum chances of being successful in preventing infection.

Critical components of successful surveillance Whereas there are clearly determined requirements for electronic surveillance, automation alone will not result in successful surveillance.8,9 The ‘mother ship’ for surveillance, as we know it in modern day IPC, is the Study on the Efficacy of Nosocomial Infection Control (SENIC) study.10 In the hospitals studied, four elements were identified which maximized the impact of the surveillance on outcome: intensive surveillance, intensive control, an infection control nurse to collect data (as well as a programme of regular feedback to clinicians), and a physician actively involved. Although this seminal work has not been replicated in the 30 years since it was published, other authors have sought to describe the elements required for successful surveillance, such as Gaynes and Solomon, who described these as: defining what outcomes to measure, reliably collecting the data in a standardized manner, analysing data for inter-hospital comparison, and using the data locally in a timely manner to improve quality of care.11 In developing national surveillance systems in recent years, much of the content of these frameworks has been adopted by many countries. Within Scotland, national surveillance was established in the same way. In recognition of the population at risk of HCAI in ICU, surveillance was established in 2010 and focused on selected HCAIs that frequently occur in the ICU: ventilator-associated pneumonia (VAP), central venous catheter-related bloodstream infection (CRBSI), and

J.S. Reilly et al. / Journal of Hospital Infection 89 (2015) 271e275 bloodstream infection (BSI).1 The remainder of this review describes the establishment of the ICU surveillance in Scotland, focusing on the critical components for success, with reference to these surveillance frameworks.

The Scottish experience of establishing electronic ICU surveillance Clinical engagement and national infrastructure In 2002, a working group was established in NHS Scotland that included intensive care society members, IPC professionals and public health professionals from the national health protection organization. It was recognized at the outset that clinical engagement was key and this was initiated with the Scottish Intensive Care Society Audit Group (SICSAG) chair. Initial discussions involved making the case for surveillance and debating definitions and methods for surveillance. This scoping work continued for the next two years, during which time decisions were made to adopt the Hospitals in Europe Link for Infection Control through Surveillance (HELICS) definitions and methods.12 These definitions were being promoted within Europe at that time and have since been adopted by the European Centre for Disease Prevention and Control (ECDC).13 Thereafter a business case was submitted to the Scottish Government for investment in the IT infrastructure for an existing ICU clinical care administrative and audit electronic system (Ward Watcher, Critical Care Audit Ltd, Otley, UK), in order to facilitate the electronic capture of data for the surveillance programme. Ward Watcher had many of the demographic variables required for the minimum dataset required for HELICS, so one page was added for HCAI containing the variables not collected elsewhere, such as criteria for case definitions of HCAI. The system was designed with internal validation to ensure that cases meeting the criteria for HCAI would be reliably collected. Whereas a period of two years seems lengthy to get to the point of collecting data, building relationships and trust is important for successful surveillance and takes time, as does accessing the funding for the project. A number of principles required to be agreed with the Scottish Intensive Care Society about central data analysis and feedback to units by SICSAG, and reassurance was needed that data were being used for local quality improvement and not to generate league tables. These principles had to be in place to initiate the work. The time required for establishing surveillance has been recognized by others involved in national and European surveillance to date; ‘The process is slow because it often requires important political decisions and an investment at national and hospital level.’13

Feasibility studies A study was carried out for a period of three months in 2005 to assess the feasibility of using the existing Scottish ICU audit database (Ward Watcher) to collect HCAI surveillance data electronically using HELICS definitions for ICU-associated infection in Scotland.12,14 Five of the 25 adult, general ICUs in Scotland participated in a prospective pilot study of HCAI in ICUs. Data were collected at HELICS patient level (level 2) surveillance of BSI, CRBSI, and VAP.12 The findings indicated that the HELICS definitions for ICU-associated infection were

273

applicable in Scotland; definitions for VAP, CRBSI, and BSI could be applied in all hospitals.12 The study provided crude infection rates, although numbers were very small, and a questionnaire sent to all participating ICUs provided information on the time taken to collect the data, the staff involved in the data collection, and any problems with case definition. The findings indicated that data collection took 10% of patients admitted to ICUs in Europe experienced a severe HCAI, such as VAP or BSI. In addition, a model-based simulation of individual patient profiles over time in the ICU was carried out, and it was estimated that 52% of VAP and 69% of BSI was preventable.

The Scottish Intensive Care Society (SICS) and Scottish Intensive Care Society Audit Group (SICSAG) are gratefully acknowledged, as are all the clinical staff in the ICUs in Scotland for their contribution to the national ICU surveillance programme.

Clinical ownership In order to maximize the prevention potential from surveillance, clinical ownership is critical. Since 2012 the Scottish Intensive Care Society Audit Group (SICSAG) included participation in HCAI surveillance as a quality improvement indicator within their professional standards.22 This demonstrates their commitment to electronic surveillance of HCAI in ICU and to use it to drive improvements in clinical care and patient outcomes.

Conclusion The case for automation in ICU surveillance has been made in several countries.49 These studies have demonstrated that

Conflict of interest statement None declared. Funding sources The Scottish Government funded the development of Ward Watcher for national surveillance (HAITF PID 3.4 and 7.13).

References 1. European Centre for Disease Prevention and Control. Point prevalence survey of healthcare associated infections and antimicrobial use in European acute care hospitals. Stockholm: ECDC; 2013. ´lvarez L, Charlett A, Holmes A. 2. Freeman R, Moore LSP, Garcı´a A Advances in electronic surveillance for healthcare-associated infections in the 21st century: a systematic review. J Hosp Infect 2013;84:106e111. 3. Health Protection Scotland. Surveillance of healthcare associated infections in Scottish intensive care units. Annual report of data from January to December 2013. Glasgow: HPS; 2014. 4. de Bruin JS, Adlassnig KP, Blacky A, Mandl H, Fehre K, Koller W. Effectiveness of an automated surveillance system for intensive care unit-acquired infections. J Am Med Inform Assoc 2013;20: 369e372. 5. De Bus L, Diet G, Gadeyne B, et al. Validity analysis of a unique infection surveillance system in the intensive care unit by analysis of a data warehouse built through a workflow-integrated software application. J Hosp Infect 2014;87:159e164. 6. van Mourik MS, Troelstra A, van Solinge WW, Moons KG, Bonten MJ. Automated surveillance for healthcare-associated infections: opportunities for improvement. Clin Infect Dis 2013;57:85e93. 7. Koller W, Blacky A, Mandl H, Rappelsberger A, Adlassnig KP. Can we bridge the definition diversity in healthcare-associated infection surveillance? From IT-supported surveillance to IT-supported infection prevention and control. Stud Health Technol Inform 2013;192:1112.

J.S. Reilly et al. / Journal of Hospital Infection 89 (2015) 271e275 8. Woeltje KF, Lin MY, Klompas M, Wright MO, Zuccotti G, Trick WE. SHEA White paper: data requirements for electronic surveillance of healthcare-associated infections. Infect Control Hosp Epidemiol 2014;35:1083e1091. 9. Graham D, Greene LR, Cain TA, et al. APIC position infections. Am J Infect Control 2009;37:510e513. 10. Haley RW, Culver DH, White JW, et al. The efficacy of infection surveillance and control programs in preventing nosocomial infections in US hospitals. Am J Epidemiol 1985;121:182e205. 11. Gaynes RP, Solomon S. Hospital Infections Program, Centers for Disease Control and Prevention, Atlanta. The Joint Commission Journal on Quality Improvement 1996;22:457e467. 12. Suetens C, Morales I, Savey A, et al. The ICU-HELICS programme: towards European surveillance of hospital-acquired infections in intensive care units. Eurosurveillance 2002;7:127e128. 13. Suetens C, Morales I, Savey A, et al. European surveillance of ICUacquired infections (HELICS-ICU): methods and main results. J Hosp Infect 2007;65(Suppl. 2):171e173. 14. Booth M, McCoubrey J, Reilly J, MacKirdy F. Feasibility of surveillance of ICU-associated infections in Scotland. Crit Care 2006;10(Suppl.):P114. 15. Haradan C, Leitch J. Scotland’s successful national approach to improving patient safety in acute care. Health Affairs 2011;30: 755e763.

275

16. McCoubrey J, MacKirdy F, Reilly J, et al. A pilot study of surveillance of intensive care unit associated infections in Scotland. J Infect Prev 2010;11:24. 17. Belgian National Surveillance of Infections in the Hospital Data validation study of the National surveillance of nosocomial infections in intensive care units (SIZ-IPH). Study protocol 2002. http://www.nsih.be/download/val_protocol_en.pdf 18. Health Protection Scotland. Intensive care unit associated infection national surveillance programme report. Glasgow: HPS; 2011. 19. Morris AC, Hay AW, Swann DG, et al. Reducing ventilatorassociated pneumonia in intensive care: impact of implementing a care bundle. Crit Care Med 2011;39:2218e2224. 20. European Centre for Disease Prevention and Control. Annual Epidemiological Report 2013. Reporting on 2011 surveillance data and 2012 epidemic intelligence data. Stockholm: ECDC; 2013. 21. Lambert ML, Silversmit G, Savey A. Preventable proportion of severe infections acquired in intensive care units: case-mix adjusted estimations from patient-based surveillance data. Infect Control Hosp Epidemiol 2014;35:494e501. 22. SICSQIP. Quality indicators for quality care in Scotland. Edinburgh: NHS National Services Scotland; 2012.

Integrating intensive care unit (ICU) surveillance into an ICU clinical care electronic system.

The intensive care unit (ICU) is the specialty with the highest prevalence of healthcare-associated infection (HCAI) in European hospitals and therefo...
202KB Sizes 0 Downloads 6 Views