Using surveillance to change practice

The current issue of the Journal of Infection Prevention (JIP) has a number of articles that illustrate the relevance of surveillance to improving the practice of infection prevention and control. Of course, this is not a new idea. We learnt from the SENIC project published in 1985 that surveillance was a critical component of successful infection prevention and control programmes. This seminal work clearly demonstrated that prevention and control activities alone were not sufficient to reduce healthcare-associated infections (HCAI). There is a good reason for this, as the authors concluded: ‘infection control problems and the need for prevention efforts were not apparent to physicians, nurses or administrators until they were given quantitative measures of the problem derived from surveillance data’ (Haley et al., 1985).

The accepted definition of surveillance is ‘the ongoing collection, analysis and interpretation of health data, closely integrated with the timely dissemination to those who need to know, and the application to preventing and controlling disease’ (Thacker and Berkelman, 1988). In the UK today, the surveillance efforts of most infection control teams are primarily directed at the capture of data to meet national imperatives. While such surveillance may meet the criteria of being ‘ongoing’, it often fails in relation to the other criteria because analysis, dissemination and application occur at a national rather than a local level. In addition, such national surveillance initiatives, which need to consume minimal resources, tend make us of readily available data and are therefore not always focused on those HCAI responsible for the main burden of disease. The findings of SENIC were drawn from surveillance programmes that covered all the main HCAI rather than single pathogens. This suggests that opportunities to prevent many more HCAI are being missed through the lack of coherent and comprehensive surveillance programmes, targeted at the broad range of HCAI that affect at least 5% of patients admitted to hospital.

The paper by Hallam et al. in this issue of JIP provides an excellent example of using surveillance in a practical and effective way in the prevention of catheter-related bloodstream infections (CR-BSI). The approach they used is locally framed but has the potential to be applied in similar way in other hospitals. This study aligns nicely with the concepts described by Haley et al. (1985). They had a clear aim for surveillance, and by integrating it with an improvement programme, used the data in decision-making; designing changes to care and monitoring their impact on rates of CR-BSI. By adapting standard definitions so that they applied locally and involving the microbiologist in identifying infections, they were able to establish a consistent approach to measurement. They applied epidemiological methods to data analysis by using a denominator of number of catheter days to report rates.

The surveillance demonstrated a steady decline in rate of CRBSI associated with the improvement programme, with a fivefold decrease in rate of CRBSI over a four-year period. Over the course of the programme, the rate of BSI reduced from 5 BSI/1000 catheter days to 0.23. This change reflects a real impact on patient morbidity and a knock-on effect on antimicrobial prescribing. While the activity to improve practice was critical to the prevention of infections, the surveillance was a primary driver of improvement. As the authors acknowledge, although the surveillance was difficult to set up, it was critical to engaging and motivating staff, who, as the project progressed, came to recognise the value of a consistent approach to measurement.

Another strength of this study is the use of improvement science methodology to make changes to care. The use of Plan Do Study Act Cycles to make incremental change and test ideas on a small scale before implementation is a concept that merits wider use in infection prevention and control (IPC) practice. Too often, IPC practitioners impose rather than test solutions with front-line staff, despite the latter approach being far more likely to achieve sustained change.

Publishing practical and effective solutions to further the art and science of infection prevention is a key aim of this journal and Hallam et al have elegantly demonstrated the impact of infection control and surveillance activity on rates of HCAI and patient safety. This would not be possible without the strong focus on developing a systematic and ongoing approach to measurement of HCAI.

A second article in this issue by Al-Mousa et al. also illustrates the key role that surveillance plays in driving improvement in infection prevention and control. It is focused on the prevention of ventilator-associated pneumonia (VAP), an important HCAI associated with considerable morbidity, healthcare resource and antimicrobial prescribing. The intervention comprised a 13-component care bundle for VAP prevention. The outcome of the intervention was measured by surveillance using standard definitions for VAP, data collected by infection prevention and control practitioners, and analysed using epidemiological methods. The study was conducted in three intensive care units in Kuwait City over an 18-month period and demonstrated a reduction in rate of VAP adjusted for ventilator days and unit of 61%, which reflected a reduction in rate from seven to three VAP/1000 ventilator days. Feedback was given to ICU staff as monthly reports of rates of VAP for their ICU compared to standard rates (based on the same case definition) for other countries participating in the surveillance system, and specific countries within the region. Interestingly, this change occurred despite process audit not being conducted due to lack of resources. Although, this means that the link between compliance with the bundle and reduction in VAP was not monitored, it does provide some evidence that the initial education programme followed by the surveillance and feedback of data on rates of VAP, was on its own sufficient to drive improvements and reduce the risk of infection.

In a previous editorial I discussed the lack of evidence supporting the effectiveness of audit in actually improving care (Wilson, 2017). I suggested that IPC audit programmes should be evaluated to ensure they are better aligned with the Model for Improvement (Nolan and Berwick, 2006) and hence constructed around the three key questions:

• What are we trying to accomplish?

• How will we know change is an improvement?

• What changes can we make that will result in improvement?

This issue of JIP clearly illustrates the value of surveillance in both achieving and demonstrating improvement and suggests that infection prevention and control teams need to devote more of their precious resource to this critical activity.