Table 1.

Aspects for inclusion and consideration when conducting a dynamic risk assessment to contain and control CPE in healthcare settings

Component	Considerations
Type of patients	-Transmission is more likely with high dependency patients and patients unable to maintain personal hygiene (e.g. diarrhoea or draining wounds).
Identification of colonised and infected patients, including surveillance	-Early detection of colonised patients on the ward or unit through screening supports the implementation of the IPC activities required to achieve control. -Disease surveillance systems are important for the identification and control of healthcare-associated infections and for understanding the impact of IPC activities [9]. -Consideration should be given to widening screening to assess the degree of spread including through the screening of contacts e.g. patients in the same bay or ward, or those known to have been in contact with the case(s) including those transferred to other wards or units. -The frequency and scope of screening will need to be informed by local expertise, resources and local risk assessments. -Other body screening sites, additional to rectal sampling, may be considered to increase detections, including the axilla and groin [10]. -Patients that initially screened negative, who then require transfer to a high-risk ward (e.g. intensive care), may warrant an additional screen at transfer and isolation pending the result of the screen [11]. -High-risk clinical areas, such as critical care units, should consider weekly or monthly screening to ensure early detection of cases of CPE. -The implementation of a case (colonised or infected) and contacts dynamic registry can help identify (otherwise unrecognisable) epidemiological links leading to the identification, and remediation of an environmental reservoir [12]. -Decolonisation of patients is not recommended and may induce resistance to decolonisation agents [13]. -Screening of patients already discharged from an outbreak ward to their usual home setting is not generally recommended; however, tagging the patient record that the patient has been on an outbreak ward should be considered, so that if readmitted they can be screened and pre-emptively isolated pending the rescreening results. -Screening in community care settings is not generally recommended, however decisions should be risk assessed based on local circumstances, such as if transmission is suspected [[14], [15], [16]].
Laboratories and IPC teams should implement systems for tracking patients with key infections	-Systems should identify and track potential cases (based on phenotypic antibiotic sensitivities) and monitor laboratory confirmed cases. The data require skilled interrogation as these analyses will inform the management of the outbreak. -Systems should take account of the potential for resistance mechanisms to spread to other Gram-negative bacteria. -Line lists of cases are required (including patient demographics, locations, specimen dates and risk factors) for tracking patient risks and movements. -Laboratories should consider adopting assays for the rapid identification of acquired carbapenemases [7]. -Laboratories must ensure that isolates confirmed as carbapenemase producers are reported to public health authority surveillance systems. -Diagnostic laboratories are well placed to support local non-hospital healthcare providers in the rapid identification of clusters or outbreaks in their locations; consideration should be given to how to identify and proactively communicate abnormal findings to these settings.
Staff-patient ratios and IPC expertise	-Transmission events are more likely where staff are looking after more patients than can be safely managed or where ward occupancy is greater than the designed capacity. -OCTs should ensure staff-patient ratios on outbreak wards/units are adequate and address deficiencies. The optimal staff-patient ratios will vary depending on the type of patients and the intensity of care they require [5,[17], [18], [19]]. -It is critical that providers have sufficient IPC expertise and staff experienced in outbreak management [9].
IPC guidelines and standards	-Adherence to IPC guidelines and cleaning standards is vital during CPE outbreaks. Combined antimicrobial stewardship, environmental cleaning and source control through the application of robust IPC practice with standard care (including isolation or cohorting) has been found to be the most effective bundle of interventions to prevent acquisition [20]. -Experienced representatives of the OCT should visit the affected areas to determine that there is robust adherence to IPC guidelines and cleaning standards [[21], [22], [23]]. -Contamination of cleaning equipment (e.g. mops) may occur; consideration should be given to investigating such potential sources [24]. -Healthcare providers should ensure single use patient equipment is used or where equipment must be reused, that appropriate disinfection/decontamination is ensured. -It is vital that patient spaces that have been occupied by patients harbouring CPE are robustly cleaned and decontaminated, particularly after the patient has been discharged; inadequate decontamination can lead to transmission events [25].
Environment	-Environmental microbiological sampling to detect environmental reservoirs guided by microbiological advice on suitable sites and sampling methods may be considered. -Positive environmental samples provide powerful evidence of cleaning deficiencies and can guide improvements, however negative results can provide false reassurance [26,27]. -Staff, patients and visitors must understand that hand hygiene sinks are for the sole purpose of hand-washing, not for disposing of food, drink or waste [28]. -CPE can be carried in the gut and are therefore easily spread through poor hygiene practice, by patients, relatives and staff. Fomites e.g. bedding contaminated with faecal soiling are therefore potential vectors. Providers need to have robust processes in place for diarrhoea management/faecal contamination mitigation. -The frequency of cleaning of toilets should be increased on outbreak wards. -Consideration should be given to staff clothing as a vector and whether additional measures are needed to reduce the transmission potential from the contaminated clothing [[29], [30], [31], [32]]. -Sink and shower waster-traps can harbour high numbers of bacteria which can persist despite cleaning and decontamination efforts due to their protective biofilms. There is some evidence that CPE in waste traps and/or drainage biofilms can transmit to patients [33,34]. Physical removal of biofilm from drains is unlikely to be successful and interventions should aim to reduce transmission from these sites to patients.
Isolation capacity on the ward or unit	-The OCT should be satisfied that that the isolation capacity and approach to cohorting (if adopted) on the ward or unit includes access to en-suite facilities to minimise onwards transmission. -Decisions around isolating or cohorting patients should consider patient safety, ensuring that there is no increased risk of harm to patients; providers should risk assess the decision and monitor for harmful outcomes. -Although the isolation of high risk patients is recommended, this recommendation is not always feasible or followed which may increase the risk of transmission [35]. -Providers should understand what risk mitigation activity has been undertaken since the admission of known cases (e.g. to ascertain if there were delays in identification and isolation of cases) to help determine the potential scale of transmission [8] and to inform further actions. -Cohorting of patients and staff into separate streams may reduce transmission events; the decision to implement cohorting must be led by local risk assessment and can include the following categories in Box B. -Cohorting should NOT be undertaken where patients have differing mechanisms of carbapenem resistance as this risks plasmid transmission between genera and the development of greater antimicrobial resistance. -Cohorting or isolation decisions should be informed by the colonisation pressure on the ward/unit i.e. the likelihood of a patient encountering a colonised patient [36]. Nosocomial transmission is more likely where the number of colonised patients is high. -Decisions around cohorting should be made with the advice of a microbiologist, to avoid inadvertently increasing harm by mixing patients with different resistance mechanisms.
Manage antibiotic pressures in the healthcare facility	-Prescribing formulary changes may be required to minimise patient or environmental exposures to broad spectrum antibiotics, especially carbapenems and third-generation cephalosporins, whilst ensuring access to these where they are truly indicated. -Monitoring of antibiotic usage can help inform changes required.