COMMENT
Success of the CLEEN study
Dinah Gould, Nicolas Drey
Lancet Infect Dis 2024 Online first August 13, 2024
https://doi.org/10.1016/S1473-3099(24)00454-7
Health-care-associated infections (HAIs) are the most frequently occurring adverse events in health care. These infections pose a major threat to patient safety and were regarded as an inevitable risk of health care for many years; however, contemporary opinion is that 50% of HAIs are avoidable.1 Designing and executing rigorous trials to evaluate interventions to prevent and control HAIs demand methodological ingenuity. Consequently, the evidence base underpinning infection prevention and control (IPC) is fragile, and most IPC guidelines are based on expert opinion.2 The trial by Katrina Browne and colleagues published in The Lancet Infectious Diseases shows that robust IPC research is achievable.3
Potential pathogens survive well on inanimate surfaces. Traditionally, cleaning was done to reduce the number of pathogens present to avoid the risk of transmission, but there was no evidence that cleaning could reduce HAIs.4 Over the past 15 years, however, observational studies have shown that methicillin-resistant Staphylococcus aureus(MRSA), vancomycin-resistant Enterococcus spp, Clostridioides difficile, Acinetobacter spp, and norovirus can be transferred from the environment to patients.5 Risk of infection is compounded because it is unclear which staff are responsible for cleaning medical equipment (eg, blood pressure cuffs and other equipment used to monitor vital signs that is shared between patients) and so cleaning is often overlooked.6 Mitchell and colleagues reviewed this literature and identified an absence of well conducted trials to show that improved decontamination could reduce HAIs.7 The strengths of the stepped-wedge, cluster randomised, controlled trial that Browne and colleagues conducted lie in its rigorous design and the implementation and practicality of the intervention.
Although often large and ambitious studies, stepped-wedge, cluster randomised, controlled trials are manageable because the intervention is rolled out in waves to each new cluster, and it is possible to investigate and address any disadvantages of the intervention in the early clusters as the study progresses.8 A drawback to this type of study is that the clusters receive the intervention sequentially over a period of time8 and might be prone to confounding due to temporal trends, in this case, seasonal variations in infections. Browne and colleagues were aware of these risks when they conducted the CLEEN study3 and stated that the trial, which took place over 9 months and covered the winter season when infections peak, was unlikely to have suffered from confounding. No major policy or other secular changes were documented while the trial was in progress.
The appeal of stepped-wedge, cluster randomised, controlled trials to health providers is that a potentially beneficial intervention will become available to everybody taking part. In the trial by Browne and colleagues,3 the enhanced cleaning and disinfection was additional to usual practice, directed at an in-house dedicated cleaning team, and did not involve the purchase and introduction of costly new technology. Cleaning staff received a 1-h training session, which was conveniently delivered in-house, and completed 3 h of extra work each weekday. Process evaluation (ie, interviews) was used to establish their preferred method of receiving performance feedback.
IPC interventions are most persuasive when their success is shown through decreased HAI rates. All too often, authors report only proxy measures of infection; for example, the effectiveness of hand hygiene campaigns frequently rests on compliance to a hand hygiene intervention, with no attempt to report HAIs.9 Browne and colleagues3 evaluated the effect of their intervention through fortnightly point prevalence surveys of HAIs. Thoroughness of enhanced cleaning was a secondary outcome measure. The economic effect of the trial is an important, substantial element of the work. The authors have sensibly decided to publish a cost-effectiveness evaluation separately.
Browne and colleagues3 conducted their trial in a single hospital. This design is a potential weakness because it is unclear whether the hospital is representative of other hospitals, so external validity is unknown. However, as the authors point out, the issues related to cleaning and disinfecting shared medical equipment have been reported in many different countries and the equipment decontaminated was typical of equipment used in all clinical settings (eg, blood pressure monitors, wheelchairs, and metal trollies).
In summary, this trial was well planned and well executed. The findings will be of major interest to health providers, IPC experts, and policy makers internationally. The effect of reading it is like coming to the end of a good novel: the reader is left wondering what will happen next. Will this successful intervention be sustained in the participating hospital? How soon will the important economic findings become available? How will the IPC community respond to the findings of this important and well conducted trial? Reports of replication in other health provider organisations are awaited with interest.
