Advertisement

Automated room decontamination: report of a Healthcare Infection Society Working Party

Published:January 23, 2022DOI:https://doi.org/10.1016/j.jhin.2022.01.006

      Keywords

      Executive summary

      This report provides advice to hospital managers, hospital-based service providers, infection prevention and control (IPC) teams and end users who intend to employ automated room decontamination devices as part of their IPC regimens. Conventional cleaning and disinfection approaches are long established and can be very effective if thorough, but recently automated systems have become available that offer the effectiveness and safety to supplement manual methods. Some chemicals such as formaldehyde have had a place within the contained laboratory setting for many years but are too toxic for use in patient areas. Biocidal ultraviolet C light has long been used to treat water systems, but whole-room treatment systems have become available following improved electrical safety and componentry.
      Although suppliers of fumigation systems have offered decontamination services for over 20 years, new companies have entered the marketplace providing a greater choice of machine designs, catering for different budgets and usage requirements. As a result of the growth in equipment availability the choice is now much greater. This brings consumer benefits but can also be confusing to the potential end user, who might not be familiar with the wealth of technical specifications for these specialized systems.
      This report is independent and aims to provide useful, generic information that will help healthcare professionals make a well-informed choice if they are intending to buy or rent/lease the automated technology. The aim is to provide guidance on the types of device available, the various active chemicals (where relevant), the biocidal mechanism underpinning the technology, suggested information to be sought from the supplier before purchase, and general precautions recommended for the safe and effective use of the equipment.

      Recommendations

      Consider use of an automated decontamination device as a supplement to manual cleaning in the context of rising or high prevalence of nosocomial infection, such as Clostridioides difficile, meticillin-resistant Staphylococcus aureus, or vancomycin-resistant enterococcus.
      Consider use of hydrogen peroxide vapour or pulsed-xenon ultraviolet light in room surface decontamination during an outbreak of C. difficile infection when other modalities have failed to reduce acquisition.

      Good practice points

      • Manual cleaning should be completed to the same high standard regardless of the subsequent use of automated cleaning devices.
      • On first use of a fumigant or ultraviolet light in a specific room design, efficacy of sealing should be monitored to ensure safety.
      • Prioritize different cleaning systems to the type of infection of the most recent room occupant by use of a red/amber/green rating based on local nosocomial infection rates.
      • Remove foam materials from the room if fumigant is used unless sealed under an impervious cover.
      • Before purchasing or renting a system, run a mock decontamination cycle in a hospital room to determine turnaround times.
      • After purchasing an ultraviolet-light decontamination system, consider the impact on surface finishes such as whitened polyvinyl chloride (PVC) before purchasing other equipment, and ask the equipment supplier to confirm compatibility.
      • Monitor levels of fumigant or ultraviolet light at regular intervals during the contract to ensure efficacy.
      • When adopting a new automated system or disinfecting a new room design, conduct microbiological culture tests (if permitted in the hospital) or take in-use environmental swab tests before and after disinfection to confirm efficacy.

      Lay summary

      Acquiring an infection in hospital is undesirable, especially if the infection is resistant to antibiotic treatment. Manual cleaning and disinfection of patient rooms and areas in which care is delivered can leave surfaces contaminated with micro-organisms (such as bacteria or viruses) that might lead to infection. This report considers the effectiveness of automated (or no-touch) decontamination devices used in addition to ordinary cleaning and disinfection in patient areas. For example, the microbiological benefit versus time taken for automated decontamination of patient rooms between one patient vacating the room and another occupying it. The main types of devices considered are those using ultraviolet light or hydrogen peroxide for the decontamination process. The report describes which devices are recommended in which circumstances, as well as practical advice on their procurement and operation. Although the devices are effective the benefit in terms of preventing patient infections needs further research.
      A glossary explaining key terms used in the report is presented in Appendix A.

      Introduction

      Infection prevention and control (IPC) measures in healthcare settings include manual cleaning (using detergent) and disinfection (using a chemical agent such as bleach). For simplicity, such procedures (which can take many and varied forms) are referred to as ‘manual cleaning/disinfection’ in this report. The procedures can be implemented one or more times per day and when patient rooms and other clinical areas are vacated (the latter being referred to as terminal cleaning/disinfection). The effectiveness of manual cleaning/disinfection depends on the thoroughness of designated procedures and the adherence of the cleaners to those procedures. Microbiological contamination of surfaces in the healthcare environment that persists due to incomplete manual cleaning/disinfection increases the risk of healthcare-associated infection, particularly for people with weakened immune systems. A potential solution is an enhanced approach to environmental surface decontamination, including those offered by automated (no-touch) room decontamination devices and systems.
      This Healthcare Infection Society (HIS) guidance incorporates a systematic evidence review evaluating the effectiveness of automated approaches to room decontamination in healthcare settings compared with manual cleaning/disinfection. The automated decontamination techniques considered include ultraviolet light, either as ultraviolet C (UV-C) or pulsed-xenon ultraviolet (PX-UV) systems, and hydrogen peroxide, either as hydrogen peroxide vapour (HPV) or aerosolized hydrogen peroxide (AHP). HPV and AHP are distinguished by the concentration of hydrogen peroxide used in the decontamination process: HPV is used for systems employing 30–35% hydrogen peroxide whereas AHP refers to systems using 5–6% hydrogen peroxide [
      Health and Safety Executive
      Fumigation: health and safety guidance for employers and technicians carrying out fumigation operations.
      ]. Micro-organisms used to evaluate automated decontamination in healthcare settings include both Gram-positive and Gram-negative bacteria associated with healthcare-associated infection, for example, Acinetobacter spp., Clostridioides difficile, meticillin-resistant Staphylococcus aureus (MRSA), and vancomycin-resistant enterococcus (VRE). The primary focus of the systematic evidence review underpinning the guidance is an evaluation of the effectiveness of automated decontamination in preventing infection or colonization, either in relation to specific micro-organisms or particular types of infection, for example, surgical site infection or device-associated infection (including catheter-associated urinary tract infection (CAUTI), and central line-associated bloodstream infection (CLABSI)). The evidence review highlights published research evaluating the effectiveness of automated decontamination in terms of reducing microbiological environmental contamination in healthcare settings. The guidance overall was intended to address practicalities related to the selection and implementation of an automated decontamination system. These considerations were based on the expertise and experience of the Working Party convened by HIS to develop the guidance.

      Guidance development team

      Acknowledgements

      The work of A.J.B. was supported by the Health and Safety Executive (HSE) project PHF20022. A.P.R.W. was part funded by the National Institute for Health Research University College London Hospitals Biomedical Research Centre. The Working Party gratefully acknowledges the contribution of the late Christina (Tina) Bradley who was influential in proposing the topic, establishing the Working Party and developing the guidance.

      Source of funding

      There was no external funding for development of the guidance.

      Disclosure of potential conflicts of interest

      All members of the Working Party completed conflict-of-interest forms in line with HIS policy. No material conflicts of interest were identified.

      Relationship of authors with sponsor

      HIS commissioned the Working Party to develop the guidance. Several authors are members of HIS (A.Be., A.P.R.W., C.F., C.F.H. and M.G.) or HIS staff (A.Ba., G.L.M. and M.A.M.).

      Responsibility for the guidance

      The views expressed in the report are those of the authors and have been endorsed by HIS following an external consultation process.

      Working Party report

      What is the Working Party Report?

      The report comprises recommendations related to automated room decontamination in healthcare settings. The methodology used to develop the recommendations combines a systematic evidence review and synthesis and expert opinion (see Section Methodology for further details).

      Why do we need a Working Party Report for this topic?

      The need for guidance was prompted by a recognition that many hospitals are either already operating or preparing to purchase automated decontamination systems. Suppliers of such systems act in an intensely competitive market with no independent oversight in terms of responsibility for comparing systems and advising on their application. The guidance was intended to include practical advice for prospective purchasers considering implementation of an automated decontamination system (for example, as part of a tender process).

      What is the purpose of the Working Party Report's recommendations?

      The main purpose of the recommendations is to inform IPC practitioners about the options available for automated room decontamination in healthcare settings. The report includes research recommendations, highlighting gaps in knowledge and evidence.

      What is the scope of the guidance?

      The guidance covers automated systems for decontaminating environmental surfaces in healthcare settings. It does not cover decontamination of equipment, devices, or the air in healthcare facilities. The guidance was largely developed with hospitals in mind, but the recommendations might be useful in other healthcare settings where microbiological environmental contamination and associated risk of a healthcare-associated infection is of concern.

      What is the evidence for the guidance?

      The guidance topic was proposed by the former HIS Scientific Development Committee (whose remit was transferred to the HIS Guidelines Committee in 2019) and approved by the HIS Council. The Working Party's considerations regarding the effectiveness of automated room decontamination devices were based on a systematic review and evidence synthesis of peer-reviewed research literature, including quality assessment of the evidence using validated techniques. The members of the Working Party used their experience and expertise to supplement analysis of the published literature regarding the practicalities of selecting and implementing automated decontamination systems.

      Who developed the guidance?

      The Working Party comprised infectious diseases/microbiology clinicians, other IPC specialists such as infection control nurses, microbiologists, engineers and facilities managers specializing in healthcare cleaning. HIS staff with expertise in systematic reviewing prepared the evidence synthesis.

      Who is the guidance for?

      Any healthcare practitioner may use the guidance and adapt it as needed. Users will include clinical staff and IPC teams. The guidance aims to provide recommendations for all health and care settings and to include the available evidence for all settings where microbiological environmental contamination of surfaces is of concern. However, the studies included in the evidence review and synthesis were predominantly conducted in hospital settings. The Working Party believes that whereas many sections of the guidance are especially relevant to hospitals, some evidence and recommendations could be extrapolated to other health and social care settings such as nursing homes.

      How is the guidance structured?

      The rationale for the advice is presented in the context of the supporting evidence identified through systematic literature searches or, in the case of the practicalities of selecting and implementing an automated decontamination system, the expert opinion of the Working Party. Evidence statements summarize the results of the systematic literature searches and evidence synthesis. The phrasing and classification of recommendations reflects the strength of the supporting evidence or reliance on expert opinion. It should be noted that the guidance is of a general nature and that an employer should consider the specific conditions of each individual place of work and comply with all applicable legislation, including the Health and Safety at Work etc. Act 1974 (see https://www.legislation.gov.uk/ukpga/1974/37/contents) and provisions for the regulation, supply, and use of biocides (see https://www.hse.gov.uk/biocides/).

      How frequently will the guidance be reviewed and updated?

      The guidance will be reviewed at least every four years and updated if changes are necessary or if new evidence emerges that requires a change in practice.

      Aim

      The primary aims of the guidance were to evaluate the effectiveness of different approaches to automated room decontamination in healthcare settings and to support decision-making regarding the practicalities of selecting and implementing a particular approach. A secondary aim was to identify areas in need of further research.

      Implementation of the guidance

      How can the guidance be used to improve clinical effectiveness?

      The guidance can be used to inform local IPC advice and in the procurement process for automated room decontamination devices. It provides a framework for audit for quality improvement in maintaining a safe patient environment.

      How much will it cost to implement the guidance?

      Automated room decontamination devices represent significant revenue and capital expenditure which will need to be balanced against potential reduction in hospital-acquired infection and improved quality of life for patients. Similar benefits can be achieved by increasing investment in standard cleaning.

      Summary of audit measures

      The following expressed as percentage compliance:
      • All rooms receive enhanced disinfection (either automated or additional manual cleaning) where a patient with C. difficile infection has been discharged or transferred.
      • All rooms given the appropriate level of cleaning according to the patient pathogens present and not derogated due to patient accommodation pressures.

      Supplementary tools

      Continuing professional development (CPD) questions and model answers for self-assessment are presented in Appendix B.

      Methodology

      Overview

      The processes and methods used to develop the systematic evidence review evaluating the effectiveness of automated approaches to decontamination were based on those described in the National Institute for Health and Care Excellence (NICE) guidelines manual [
      National Institute for Health and Care Excellence
      Developing NICE guidelines: the manual.
      ]. The review question was expressed in the patient–intervention–comparator–outcome (PICO) framework as presented in Table I.
      Table IThe review question formulated using the PICO framework
      Population/settingInterventionComparatorOutcomes
      Patients in any healthcare settingUse of an automated device to decontaminate a patient room or other clinical areaNo cleaning, manual cleaning/disinfection or another automated decontamination devicePatients – infection or colonization with any pathogen
      Additional evidence: micro-organisms, including those experimentally inoculatedMicro-organisms – microbial count (on any surface)
      PICO patient-intervention-comparator-outcome.
      Exclusion criteria: studies describing decontamination of equipment or devices; automated devices used for decontamination of air; non-comparative clinical outbreak studies; studies reporting a total count, but not specific types of micro-organisms.

      Data sources and search strategy

      Three electronic databases (Embase, MEDLINE, and CINAHL) were searched for published articles using medical subject headings (MeSH) and free-text terms. Reference lists from published reviews identified in the literature searches were used to identify additional studies to be considered for inclusion in the guidance review. No date or language restrictions were applied as part of the searches, which were completed in February 2021. Further details of the searches are presented in Appendix C.

      Study eligibility and selection criteria

      Published articles identified through the literature searches were screened for relevance against the PICO framework. One reviewer examined titles, abstracts, and full texts of all records identified through the searches. A second reviewer checked at least 10% of records earmarked for exclusion at each stage of screening. Disagreements were first discussed between the two reviewers and, if consensus was not reached, a third reviewer was consulted. The results are presented in the study selection flowchart in Appendix D. A list of studies excluded after full-text screening is presented in Appendix E.

      Data extraction, preliminary analysis, and quality assessment

      The characteristics of included studies were summarized in the evidence tables presented in Appendix F. For each included study, data were extracted into an evidence table by one reviewer and a second reviewer checked the data extraction for 10% of studies. Priority was given to studies reporting the clinical outcomes of infection or colonization, whereas additional studies reporting only environmental sampling outcomes were highlighted in a separate evidence table.
      The preferred outcome measure for extraction of clinical outcomes was the incidence rate in each treatment arm, since these are used to calculate incidence rate ratios (IRRs) for interventions and comparators in the same study. An IRR of one implies no difference between the incidence rates for two treatments under comparison, whereas an IRR less (greater) than one implies a reduction (increase) in the incidence rate relative to the reference treatment (which, for the purposes of this evidence review, was defined as the most conservative approach to manual cleaning/disinfection evaluated in each study). Further details relevant to the calculation of incidence rates and IRRs are described in Appendix F.
      Included studies reporting the clinical outcomes of infection or colonization were appraised for quality using checklists recommended in the NICE guidelines manual [
      National Institute for Health and Care Excellence
      Developing NICE guidelines: the manual.
      ]. Critical appraisal was conducted by one reviewer, and appraisal outcomes for at least 10% of studies were checked by a second reviewer. The results of study-level quality appraisal are presented in Appendix G, with results stratified (organized) by study design.

      Network meta-analysis

      Network meta-analysis (NMA) was considered relevant for quantitative synthesis of the clinical outcomes of infection or colonization because of the multiplicity of automated approaches to decontamination under consideration. Whereas pairwise meta-analysis allows comparison of two treatments (for example, treatment A versus treatment B), NMA allows a unified (and therefore more powerful and informative) comparison of three or more treatments (for example, treatment A versus treatment B, treatment A versus treatment C, and treatment B versus treatment C). Studies involving head-to-head comparisons of treatments provide direct evidence related to those treatment comparisons; however, NMA also allows indirect evidence to be incorporated in the analysis (for example, one study comparing treatments A and B directly and another study comparing treatments A and C directly provide indirect evidence for the comparison of treatments B and C).
      In this evidence review, NMA was planned to allow comparison of a variety of automated decontamination systems with manual cleaning/disinfection and with each other. By contrast, previously published meta-analyses have been restricted to pairwise comparisons of either ultraviolet light or hydrogen peroxide systems with manual cleaning/disinfection (for example, Marra et al. [
      • Marra A.R.
      • Schweizer M.L.
      • Edmond M.B.
      No-touch disinfection methods to decrease multidrug-resistant organism infections: a systematic review and meta-analysis.
      ] and Dong et al. [
      • Dong Z.
      • Zhou N.
      • Liu G.
      • Zhao L.
      Role of pulsed-xenon ultraviolet light in reducing healthcare-associated infections: a systematic review and meta-analysis.
      ]). NMA is becoming more widely used within guideline development, and the statistical methodology used in this evidence review mirrors that used in the NICE guideline development programme [
      National Institute for Health and Care Excellence
      Developing NICE guidelines: the manual, Appendix K: network meta-analysis reporting standards.
      ]. Generic code for Bayesian NMA using the statistical software WinBUGS (see https://www.mrc-bsu.cam.ac.uk/software/bugs/the-bugs-project-winbugs/) was adapted for the analyses conducted for the guidance (see below) [
      • Dias S.
      • Welton N.J.
      • Sutton A.J.
      • Ades A.E.
      NICE DSU Technical Support Document 2: a generalised linear modelling framework for pairwise and network meta-analysis of randomised controlled trials 2011.
      ]. The statistical package R (see https://www.r-project.org/) was used for graphical presentation of NMA data structures.
      The NMAs conducted as part of this evidence review met good practice criteria [
      • Hutton B.
      • Cameron C.
      • Moher D.
      • Salanti G.
      • Chaimani A.
      • Caldwell D.M.
      • et al.
      The PRISMA extension statement for reporting of systematic reviews incorporating network meta-analyses of health care interventions: checklist and explanations.
      ] covering: creation of network diagrams and examination of the geometry of each network and implications for the analysis (for example, in terms of risk of bias); adjustments for correlated outcomes in multi-arm studies (studies evaluating three or more treatments); model fitting (including assessment of convergence in the Bayesian computational framework); model checking (for example, using deviance residuals); interpretation of results both as IRRs for all pairwise treatment contrasts supported by the network and by considering surface under cumulative ranking (SUCRA) scores for individual treatments; and exploration of model assumptions including transitivity (by comparing study designs in relation to the PICO framework) and inconsistency (by comparing direct and indirect treatment effect estimates) where possible. Further details of these aspects of the methodology are presented in Appendix H.
      IRRs for all pairwise treatment contrasts supported by each network were calculated as part of NMA model-fitting. IRRs are easier to interpret than the loge-IRRs and associated standard errors (SEs) that formed the data inputs for the NMAs (see Appendix H). Posterior distributions for the IRRs were summarized in terms of medians and 95% credible intervals (CrIs; analogous to 95% confidence intervals (CIs) used in frequentist approaches to statistical inference). Treatment rankings were also calculated for each iteration of model-fitting, and these were summarized using SUCRA scores expressed as percentages such that a treatment uniformly ranked most (least) effective over all iterations would have a score of 100% (0%).

      Rating of evidence and recommendations

      Evidence synthesized in the guidance review was assessed for quality at outcome level using the approach known as Grading of Recommendations Assessment, Development and Evaluation (GRADE; see https://www.gradeworkinggroup.org/for details). The resulting GRADE tables are presented in Appendix I, with results stratified by the micro-organisms associated with infection or colonization, or type of infection (surgical site infection, device-associated infection, or infection specific to a body organ or system). Using GRADE, the overall quality of the evidence for each clinical outcome of infection or colonization was classified as very low, low, moderate, or high.
      Evidence statements for the clinical outcomes of infection or colonization were constructed by combining the outcome-level classification of evidence quality determined using GRADE and the following terms reflecting the overall confidence in using the evidence to formulate recommendations:
      • strong evidence – further research is unlikely to alter confidence in the estimated effect
      • moderate evidence – further research might alter the estimated effect and its strength
      • weak evidence – further research is very likely to alter the estimated effect and its strength
      • inconsistent evidence – current studies report conflicting evidence and further research is very likely to alter the estimated effect.
      In accordance with the GRADE approach, the Working Party's recommendations related to the clinical outcomes of infection or colonization were phrased to reflect the strength of the evidence and the Working Party's confidence in using it as the basis for developing recommendations.
      Where there was little or no evidence related to the clinical outcomes of infection or colonization that could be used to guide recommendations, the Working Party used informal consensus to formulate good practice points based on their collective experience and expertise. The Working Party also used this approach to formulate advice regarding the practicalities of choosing and implementing automated decontamination systems. In addition, the Working Party formulated recommendations for further research to address identified gaps in the evidence.

      Consultation process

      Feedback on the draft guidance was received from the HIS Guidelines Committee and through consultation with relevant stakeholders. The draft report was placed on the HIS website for 10 working days along with the HIS standard response form, including a conflict-of-interest disclosure form. The availability of the draft guidance was communicated via e-mail and social media. Stakeholders were invited to comment on the format, content, local applicability, patient acceptability, and recommendations. The Working Party reviewed stakeholder comments and collectively agreed revisions in response to the comments (see Appendix J). Comments received from individuals who disclosed conflicts of interest, or who did not submit a conflict-of-interest disclosure form, were not considered by the Working Party.

      Rationale for recommendations

      Which automated room decontamination devices are effective for reducing microbial burden and preventing infection or colonization in healthcare settings?

      Search results and study selection

      The literature searches, which were performed in accordance with the search terms in Tables C.1 and C.2, identified 1041 articles; a further 13 articles were identified by handsearching reference lists, etc. (see Figure D.1). One thousand and one articles were eventually excluded, with those considered at the full-text stage being listed in Table E.1 together with reasons for exclusion. A total of 53 articles were selected for inclusion, representing 29 distinct studies reporting clinical outcomes (see Table F.1) [
      • Anderson D.J.
      • Chen L.F.
      • Weber D.J.
      • Moehring R.W.
      • Lewis S.S.
      • Triplett P.F.
      • et al.
      Enhanced terminal room disinfection and acquisition and infection caused by multidrug-resistant organisms and Clostridium difficile (the Benefits of Enhanced Terminal Room Disinfection study): a cluster-randomised, multicentre, crossover study.
      ,
      • Anderson D.J.
      • Moehring R.W.
      • Weber D.J.
      • Lewis S.S.
      • Chen L.F.
      • Schwab J.C.
      • et al.
      Effectiveness of targeted enhanced terminal room disinfection on hospital-wide acquisition and infection with multidrug-resistant organisms and Clostridium difficile: a secondary analysis of a multicentre cluster randomised controlled trial with crossover design (BETR Disinfection).
      ,
      • Attia F.
      • Whitener C.
      • Mincemoyer S.
      • Houck J.
      • Julian K.
      The effect of pulsed xenon ultraviolet light disinfection on healthcare-associated Clostridioides difficile rates in a tertiary care hospital.
      ,
      • Boyce J.M.
      • Havill N.L.
      • Otter J.A.
      • McDonald L.C.
      • Adams N.M.T.
      • Cooper T.
      • et al.
      Impact of hydrogen peroxide vapor room decontamination on Clostridium difficile environmental contamination and transmission in a healthcare setting.
      ,
      • Boyce J.M.
      • Guercia K.A.
      • Sullivan L.
      • Havill N.L.
      • Fekieta R.
      • Kozakiewicz J.
      • et al.
      Prospective cluster controlled crossover trial to compare the impact of an improved hydrogen peroxide disinfectant and a quaternary ammonium-based disinfectant on surface contamination and health care outcomes.
      ,
      • Brite J.
      • McMillen T.
      • Robilotti E.
      • Sun J.
      • Chow H.Y.
      • Stell F.
      • et al.
      Effectiveness of ultraviolet disinfection in reducing hospital-acquired Clostridium difficile and vancomycin-resistant Enterococcus on a bone marrow transplant unit.
      ,
      • Catalanotti A.
      • Abbe D.
      • Simmons S.
      • Stibich M.
      Influence of pulsed-xenon ultraviolet light-based environmental disinfection on surgical site infections.
      ,
      • Doll M.E.
      • Zhao J.
      • Kang L.
      • Rittmann B.
      • Alvarez M.
      • Fleming M.
      • et al.
      Chasing the rate: An interrupted time series analysis of interventions targeting reported hospital onset Clostridioides difficile, 2013–2018.
      ,
      • Green C.
      • Pamplin J.C.
      • Chafin K.N.
      • Murray C.K.
      • Yun H.C.
      Pulsed-xenon ultraviolet light disinfection in a burn unit: Impact on environmental bioburden, multidrug-resistant organism acquisition and healthcare associated infections.
      ,
      • Haas J.P.
      • Menz J.
      • Dusza S.
      • Montecalvo M.A.
      Implementation and impact of ultraviolet environmental disinfection in an acute care setting.
      ,
      • Hardy K.J.
      • Gossain S.
      • Henderson N.
      • Drugan C.
      • Oppenheim B.A.
      • Gao F.
      • et al.
      Rapid recontamination with MRSA of the environment of an intensive care unit after decontamination with hydrogen peroxide vapour.
      ,
      • Horn K.
      • Otter J.A.
      Hydrogen peroxide vapor room disinfection and hand hygiene improvements reduce Clostridium difficile infection, methicillin-resistant Staphylococcus aureus, vancomycin-resistant enterococci, and extended-spectrum beta-lactamase.
      ,
      • Kitagawa H.
      • Mori M.
      • Kawano R.
      • Hara T.
      • Kashiyama S.
      • Hayashi Y.
      • et al.
      Combining pulsed xenon ultraviolet disinfection with terminal manual cleaning helps reduce the acquisition rate of methicillin-resistant Staphylococcus aureus.
      ,
      • Kovach C.R.
      • Taneli Y.
      • Neiman T.
      • Dyer E.M.
      • Arzaga A.J.A.
      • Kelber S.T.
      Evaluation of an ultraviolet room disinfection protocol to decrease nursing home microbial burden, infection and hospitalization rates.
      ,
      • Levin J.
      • Riley L.S.
      • Parrish C.
      • English D.
      • Ahn S.
      The effect of portable pulsed xenon ultraviolet light after terminal cleaning on hospital-associated Clostridium difficile infection in a community hospital.
      ,
      • Manian F.A.
      • Griesnauer S.
      • Bryant A.
      Implementation of hospital-wide enhanced terminal cleaning of targeted patient rooms and its impact on endemic Clostridium difficile infection rates.
      ,
      • McCord J.
      • Prewitt M.
      • Dyakova E.
      • Mookerjee S.
      • Otter J.A.
      Reduction in Clostridium difficile infection associated with the introduction of hydrogen peroxide vapour automated room disinfection.
      ,
      • McMullen K.
      • Guth R.M.
      • Wood H.
      • Mueller C.
      • Dunn G.
      • Wade R.
      • et al.
      Impact of no-touch ultraviolet light room disinfection systems on Clostridioides difficile infections.
      ,
      • Miller R.
      • Simmons S.
      • Dale C.
      • Stachowiak J.
      • Stibich M.
      Utilization and impact of a pulsed-xenon ultraviolet room disinfection system and multidisciplinary care team on Clostridium difficile in a long-term acute care facility.
      ,
      • Mitchell B.G.
      • Digney W.
      • Locket P.
      • Dancer S.J.
      Controlling methicillin-resistant Staphylococcus aureus (MRSA) in a hospital and the role of hydrogen peroxide decontamination: an interrupted time series analysis.
      ,
      • Morikane K.
      • Suzuki S.
      • Yoshioka J.
      • Yakuwa J.
      • Nakane M.
      • Nemoto K.
      Clinical and microbiological effect of pulsed xenon ultraviolet disinfection to reduce multidrug-resistant organisms in the intensive care unit in a Japanese hospital: a before-after study.
      ,
      • Murphy P.
      • Kang L.
      • Fleming M.
      • Atkinson C.
      • Pryor R.
      • Cooper K.
      • et al.
      Effect of ultraviolet-C light disinfection at terminal patient discharge on hospital-acquired infections in bone marrow transplant and oncology units.
      ,
      • Murrell L.J.
      • Hamilton E.K.
      • Johnson H.B.
      • Spencer M.
      Influence of a visible-light continuous environmental disinfection system on microbial contamination and surgical site infections in an orthopedic operating room.
      ,
      • Nagaraja A.
      • Visintainer P.
      • Haas J.P.
      • Menz J.
      • Wormser G.P.
      • Montecalvo M.A.
      Clostridium difficile infections before and during use of ultraviolet disinfection.
      ,
      • Napolitano N.A.
      • Mahapatra T.
      • Tang W.
      The effectiveness of UV-C radiation for facility-wide environmental disinfection to reduce health care-acquired infections.
      ,
      • Passaretti C.L.
      • Otter J.A.
      • Reich N.G.
      • Myers J.
      • Shepard J.
      • Ross T.
      • et al.
      An evaluation of environmental decontamination with hydrogen peroxide vapor for reducing the risk of patient acquisition of multidrug-resistant organisms.
      ,
      • Pegues D.A.
      • Han J.
      • Gilmar C.
      • McDonnell B.
      • Gaynes S.
      Impact of ultraviolet germicidal irradiation for no-touch terminal room disinfection on Clostridium difficile infection incidence among hematology–oncology patients.
      ,
      • Raggi R.
      • Archulet K.
      • Haag C.W.
      • Tang W.
      Clinical, operational, and financial impact of an ultraviolet-C terminal disinfection intervention at a community hospital.
      ,
      • Rutala W.A.
      • Kanamori H.
      • Gergen M.F.
      • Knelson L.P.
      • Sickbert-Bennett E.E.
      • Chen L.F.
      • et al.
      Enhanced disinfection leads to reduction of microbial contamination and a decrease in patient colonization and infection.
      ,
      • Sampathkumar P.
      • Folkert C.
      • Barth J.E.
      • Nation L.
      • Benz M.
      • Hesse A.
      • et al.
      A trial of pulsed xenon ultraviolet disinfection to reduce Clostridioides difficile infection.
      ,
      • Schaffzin J.K.
      • Wilhite A.W.
      • Li Z.
      • Finney D.
      • Ankrum A.L.
      • Moore R.
      Maximizing efficiency in a high occupancy setting to utilize ultraviolet disinfection for isolation rooms.
      ,
      • Vianna P.G.
      • Dale C.R.
      • Simmons S.
      • Stibich M.
      • Licitra C.M.
      Impact of pulsed xenon ultraviolet light on hospital-acquired infection rates in a community hospital.
      ] and 21 further studies reporting only environmental sampling outcomes involving either detection of clinically occurring environmental contamination [
      • Barbut F.
      • Menuet D.
      • Verachten M.
      • Girou E.
      Comparison of the efficacy of a hydrogen peroxide dry-mist disinfection system and sodium hypochlorite solution for eradication of Clostridium difficile spores.
      ,
      • Blazejewski C.
      • Wallet F.
      • Rouze A.
      • Le Guern R.
      • Ponthieux S.
      • Salleron J.
      • et al.
      Efficiency of hydrogen peroxide in improving disinfection of ICU rooms.
      ,
      • French G.L.
      • Otter J.A.
      • Shannon K.P.
      • Adams N.M.T.
      • Watling D.
      • Parks M.J.
      Tackling contamination of the hospital environment by methicillin-resistant Staphylococcus aureus (MRSA): a comparison between conventional terminal cleaning and hydrogen peroxide vapour decontamination.
      ,
      • Ghantoji S.S.
      • Stibich M.
      • Stachowiak J.
      • Cantu S.
      • Adachi J.A.
      • Raad II,
      • et al.
      Non-inferiority of pulsed xenon UV light versus bleach for reducing environmental Clostridium difficile contamination on high-touch surfaces in Clostridium difficile infection isolation rooms.
      ,
      • Jinadatha C.
      • Quezada R.
      • Huber T.W.
      • Williams J.B.
      • Zeber J.E.
      • Copeland L.A.
      Evaluation of a pulsed-xenon ultraviolet room disinfection device for impact on contamination levels of methicillin-resistant Staphylococcus aureus.
      ,
      • Lerner A.O.
      • Abu-Hanna J.
      • Carmeli Y.
      • Schechner V.
      Environmental contamination by carbapenem-resistant Acinetobacter baumannii: the effects of room type and cleaning methods.
      ,
      • Maclean M.
      • MacGregor S.J.
      • Anderson J.G.
      • Woolsey G.A.
      • Coia J.E.
      • Hamilton K.
      • et al.
      Environmental decontamination of a hospital isolation room using high-intensity narrow-spectrum light.
      ,
      • Mosci D.
      • Marmo G.W.
      • Sciolino L.
      • Zaccaro C.
      • Antonellini R.
      • Accogli L.
      • et al.
      Automatic environmental disinfection with hydrogen peroxide and silver ions versus manual environmental disinfection with sodium hypochlorite: a multicentre randomized before-and-after trial.
      ,
      • Sitzlar B.
      • Deshpande A.
      • Fertelli D.
      • Kundrapu S.
      • Sethi A.K.
      • Donskey C.J.
      An environmental disinfection odyssey: evaluation of sequential interventions to improve disinfection of Clostrid-ium difficile isolation rooms.
      ,
      • Warren B.G.
      • Turner N.
      • Smith B.
      • Addison R.
      • Marden S.
      • Weber D.J.
      • et al.
      Measuring the impact of continuous disinfection strategies on environmental burden in outpatient settings: a prospective randomized controlled trial.
      ,
      • Wong T.
      • Woznow T.
      • Petrie M.
      • Murzello E.
      • Muniak A.
      • Kadora A.
      • et al.
      Postdischarge decontamination of MRSA, VRE, and Clostridium difficile isolation rooms using 2 commercially available automated ultraviolet-C-emitting devices.
      ,
      • Yui S.
      • Ali S.
      • Muzslay M.
      • Jeanes A.
      • Wilson A.P.R.
      Identification of Clostridium difficile reservoirs in the patient environment and efficacy of aerial hydrogen peroxide decontamination.
      ,
      • Zeber J.E.
      • Pfeiffer C.
      • Baddley J.W.
      • Cadena-Zuluaga J.
      • Stock E.M.
      • Copeland L.A.
      • et al.
      Effect of pulsed xenon ultraviolet room disinfection devices on microbial counts for methicillin-resistant Staphylococcus aureus and aerobic bacterial colonies.
      ,
      • Zeber J.E.
      • Coppin J.D.
      • Villamaria F.C.
      • Williams M.D.
      • Copeland L.A.
      • Chatterjee P.
      • et al.
      Use of ultraviolet irradiation in addition to commonly used hospital disinfectants or cleaners further reduces the bioburden on high-touch surfaces.
      ] or experimental inoculation of surfaces [
      • Ali S.
      • Muzslay M.
      • Bruce M.
      • Jeanes A.
      • Moore G.
      • Wilson A.P.R.
      Efficacy of two hydrogen peroxide vapour aerial decontamination systems for enhanced disinfection of meticillin-resistant Staphylococcus aureus, Klebsiella pneumoniae and Clostridium difficile in single isolation rooms.
      ,
      • Ali S.
      • Yui S.
      • Muzslay M.
      • Wilson A.P.R.
      Comparison of two whole-room ultraviolet irradiation systems for enhanced disinfection of contaminated hospital patient rooms.
      ,
      • Doan L.
      • Forrest H.
      • Fakis A.
      • Craig J.
      • Claxton L.
      • Khare M.
      Clinical and cost effectiveness of eight disinfection methods for terminal disinfection of hospital isolation rooms contaminated with Clostridium difficile 027.
      ,
      • Havill N.L.
      • Moore B.A.
      • Boyce J.M.
      Comparison of the microbiological efficacy of hydrogen peroxide vapor and ultraviolet light processes for room decontamination.
      ,
      • Jelden K.C.
      • Gibbs S.G.
      • Smith P.W.
      • Hewlett A.L.
      • Iwen P.C.
      • Schmid K.K.
      • et al.
      Ultraviolet (UV)-reflective paint with ultraviolet germicidal irradiation (UVGI) improves decontamination of nosocomial bacteria on hospital room surfaces.
      ,
      • Rutala W.A.
      • Gergen M.F.
      • Tande B.M.
      • Weber D.J.
      Rapid hospital room decontamination using ultraviolet (UV) light with a nanostructured UV-reflective wall coating.
      ,
      • Rutala W.A.
      • Gergen M.F.
      • Tande B.M.
      • Weber D.J.
      Room decontamination using an ultraviolet-C device with short ultraviolet exposure time.
      ] (see Table F.2).
      Among the 29 studies reporting clinical outcomes, some focused specifically on infection and one focused specifically on colonization; the remainder focused on acquisition (infection or colonization without distinguishing between the two). Subsequent sections of this report are, therefore, structured and phrased according to the clinical outcomes of infection or acquisition. The evidence identified for inclusion covered infection or acquisition due to specific micro-organisms or groups of micro-organisms, surgical site infection, device-associated infection, and infection specific to body organ or system (see below for further details).
      The most frequently evaluated automated room decontamination systems in terms of clinical outcomes were UV-C (eight studies [
      • Anderson D.J.
      • Chen L.F.
      • Weber D.J.
      • Moehring R.W.
      • Lewis S.S.
      • Triplett P.F.
      • et al.
      Enhanced terminal room disinfection and acquisition and infection caused by multidrug-resistant organisms and Clostridium difficile (the Benefits of Enhanced Terminal Room Disinfection study): a cluster-randomised, multicentre, crossover study.
      ,
      • Doll M.E.
      • Zhao J.
      • Kang L.
      • Rittmann B.
      • Alvarez M.
      • Fleming M.
      • et al.
      Chasing the rate: An interrupted time series analysis of interventions targeting reported hospital onset Clostridioides difficile, 2013–2018.
      ,
      • McMullen K.
      • Guth R.M.
      • Wood H.
      • Mueller C.
      • Dunn G.
      • Wade R.
      • et al.
      Impact of no-touch ultraviolet light room disinfection systems on Clostridioides difficile infections.
      ,
      • Murphy P.
      • Kang L.
      • Fleming M.
      • Atkinson C.
      • Pryor R.
      • Cooper K.
      • et al.
      Effect of ultraviolet-C light disinfection at terminal patient discharge on hospital-acquired infections in bone marrow transplant and oncology units.
      ,
      • Napolitano N.A.
      • Mahapatra T.
      • Tang W.
      The effectiveness of UV-C radiation for facility-wide environmental disinfection to reduce health care-acquired infections.
      ,
      • Pegues D.A.
      • Han J.
      • Gilmar C.
      • McDonnell B.
      • Gaynes S.
      Impact of ultraviolet germicidal irradiation for no-touch terminal room disinfection on Clostridium difficile infection incidence among hematology–oncology patients.
      ,
      • Raggi R.
      • Archulet K.
      • Haag C.W.
      • Tang W.
      Clinical, operational, and financial impact of an ultraviolet-C terminal disinfection intervention at a community hospital.
      ,
      • Schaffzin J.K.
      • Wilhite A.W.
      • Li Z.
      • Finney D.
      • Ankrum A.L.
      • Moore R.
      Maximizing efficiency in a high occupancy setting to utilize ultraviolet disinfection for isolation rooms.
      ]), PX-UV (13 studies [
      • Attia F.
      • Whitener C.
      • Mincemoyer S.
      • Houck J.
      • Julian K.
      The effect of pulsed xenon ultraviolet light disinfection on healthcare-associated Clostridioides difficile rates in a tertiary care hospital.
      ,
      • Brite J.
      • McMillen T.
      • Robilotti E.
      • Sun J.
      • Chow H.Y.
      • Stell F.
      • et al.
      Effectiveness of ultraviolet disinfection in reducing hospital-acquired Clostridium difficile and vancomycin-resistant Enterococcus on a bone marrow transplant unit.
      ,
      • Catalanotti A.
      • Abbe D.
      • Simmons S.
      • Stibich M.
      Influence of pulsed-xenon ultraviolet light-based environmental disinfection on surgical site infections.
      ,
      • Green C.
      • Pamplin J.C.
      • Chafin K.N.
      • Murray C.K.
      • Yun H.C.
      Pulsed-xenon ultraviolet light disinfection in a burn unit: Impact on environmental bioburden, multidrug-resistant organism acquisition and healthcare associated infections.
      ,
      • Haas J.P.
      • Menz J.
      • Dusza S.
      • Montecalvo M.A.
      Implementation and impact of ultraviolet environmental disinfection in an acute care setting.
      ,
      • Kitagawa H.
      • Mori M.
      • Kawano R.
      • Hara T.
      • Kashiyama S.
      • Hayashi Y.
      • et al.
      Combining pulsed xenon ultraviolet disinfection with terminal manual cleaning helps reduce the acquisition rate of methicillin-resistant Staphylococcus aureus.
      ,
      • Kovach C.R.
      • Taneli Y.
      • Neiman T.
      • Dyer E.M.
      • Arzaga A.J.A.
      • Kelber S.T.
      Evaluation of an ultraviolet room disinfection protocol to decrease nursing home microbial burden, infection and hospitalization rates.
      ,
      • Levin J.
      • Riley L.S.
      • Parrish C.
      • English D.
      • Ahn S.
      The effect of portable pulsed xenon ultraviolet light after terminal cleaning on hospital-associated Clostridium difficile infection in a community hospital.
      ,
      • McMullen K.
      • Guth R.M.
      • Wood H.
      • Mueller C.
      • Dunn G.
      • Wade R.
      • et al.
      Impact of no-touch ultraviolet light room disinfection systems on Clostridioides difficile infections.
      ,
      • Miller R.
      • Simmons S.
      • Dale C.
      • Stachowiak J.
      • Stibich M.
      Utilization and impact of a pulsed-xenon ultraviolet room disinfection system and multidisciplinary care team on Clostridium difficile in a long-term acute care facility.
      ,
      • Morikane K.
      • Suzuki S.
      • Yoshioka J.
      • Yakuwa J.
      • Nakane M.
      • Nemoto K.
      Clinical and microbiological effect of pulsed xenon ultraviolet disinfection to reduce multidrug-resistant organisms in the intensive care unit in a Japanese hospital: a before-after study.
      ,
      • Sampathkumar P.
      • Folkert C.
      • Barth J.E.
      • Nation L.
      • Benz M.
      • Hesse A.
      • et al.
      A trial of pulsed xenon ultraviolet disinfection to reduce Clostridioides difficile infection.
      ,
      • Vianna P.G.
      • Dale C.R.
      • Simmons S.
      • Stibich M.
      • Licitra C.M.
      Impact of pulsed xenon ultraviolet light on hospital-acquired infection rates in a community hospital.
      ]), and HPV (six studies [
      • Boyce J.M.
      • Havill N.L.
      • Otter J.A.
      • McDonald L.C.
      • Adams N.M.T.
      • Cooper T.
      • et al.
      Impact of hydrogen peroxide vapor room decontamination on Clostridium difficile environmental contamination and transmission in a healthcare setting.
      ,
      • Hardy K.J.
      • Gossain S.
      • Henderson N.
      • Drugan C.
      • Oppenheim B.A.
      • Gao F.
      • et al.
      Rapid recontamination with MRSA of the environment of an intensive care unit after decontamination with hydrogen peroxide vapour.
      ,
      • Horn K.
      • Otter J.A.
      Hydrogen peroxide vapor room disinfection and hand hygiene improvements reduce Clostridium difficile infection, methicillin-resistant Staphylococcus aureus, vancomycin-resistant enterococci, and extended-spectrum beta-lactamase.
      ,
      • Manian F.A.
      • Griesnauer S.
      • Bryant A.
      Implementation of hospital-wide enhanced terminal cleaning of targeted patient rooms and its impact on endemic Clostridium difficile infection rates.
      ,
      • McCord J.
      • Prewitt M.
      • Dyakova E.
      • Mookerjee S.
      • Otter J.A.
      Reduction in Clostridium difficile infection associated with the introduction of hydrogen peroxide vapour automated room disinfection.
      ,
      • Passaretti C.L.
      • Otter J.A.
      • Reich N.G.
      • Myers J.
      • Shepard J.
      • Ross T.
      • et al.
      An evaluation of environmental decontamination with hydrogen peroxide vapor for reducing the risk of patient acquisition of multidrug-resistant organisms.
      ]). AHP was compared with manual cleaning/disinfection in one study [
      • Mitchell B.G.
      • Digney W.
      • Locket P.
      • Dancer S.J.
      Controlling methicillin-resistant Staphylococcus aureus (MRSA) in a hospital and the role of hydrogen peroxide decontamination: an interrupted time series analysis.
      ], as was a visible (indigo and white) light continuous disinfection system [
      • Murrell L.J.
      • Hamilton E.K.
      • Johnson H.B.
      • Spencer M.
      Influence of a visible-light continuous environmental disinfection system on microbial contamination and surgical site infections in an orthopedic operating room.
      ]. Most studies focused on the use of automated decontamination devices after manual cleaning/disinfection (most frequently in the context of terminal cleaning/disinfection of patient rooms). However, one study [
      • Murphy P.
      • Kang L.
      • Fleming M.
      • Atkinson C.
      • Pryor R.
      • Cooper K.
      • et al.
      Effect of ultraviolet-C light disinfection at terminal patient discharge on hospital-acquired infections in bone marrow transplant and oncology units.
      ] compared UV-C at every terminal discharge to UV-C only at terminal discharge of patients with C. difficile infection.

      Assessment of methodological quality

      Two of the studies reporting clinical outcomes were conducted as controlled trials [
      • Anderson D.J.
      • Chen L.F.
      • Weber D.J.
      • Moehring R.W.
      • Lewis S.S.
      • Triplett P.F.
      • et al.
      Enhanced terminal room disinfection and acquisition and infection caused by multidrug-resistant organisms and Clostridium difficile (the Benefits of Enhanced Terminal Room Disinfection study): a cluster-randomised, multicentre, crossover study.
      ,
      • Boyce J.M.
      • Guercia K.A.
      • Sullivan L.
      • Havill N.L.
      • Fekieta R.
      • Kozakiewicz J.
      • et al.
      Prospective cluster controlled crossover trial to compare the impact of an improved hydrogen peroxide disinfectant and a quaternary ammonium-based disinfectant on surface contamination and health care outcomes.
      ], five were conducted as controlled before–after studies [
      • Kitagawa H.
      • Mori M.
      • Kawano R.
      • Hara T.
      • Kashiyama S.
      • Hayashi Y.
      • et al.
      Combining pulsed xenon ultraviolet disinfection with terminal manual cleaning helps reduce the acquisition rate of methicillin-resistant Staphylococcus aureus.
      ,
      • Morikane K.
      • Suzuki S.
      • Yoshioka J.
      • Yakuwa J.
      • Nakane M.
      • Nemoto K.
      Clinical and microbiological effect of pulsed xenon ultraviolet disinfection to reduce multidrug-resistant organisms in the intensive care unit in a Japanese hospital: a before-after study.
      ,
      • Murrell L.J.
      • Hamilton E.K.
      • Johnson H.B.
      • Spencer M.
      Influence of a visible-light continuous environmental disinfection system on microbial contamination and surgical site infections in an orthopedic operating room.
      ,
      • Passaretti C.L.
      • Otter J.A.
      • Reich N.G.
      • Myers J.
      • Shepard J.
      • Ross T.
      • et al.
      An evaluation of environmental decontamination with hydrogen peroxide vapor for reducing the risk of patient acquisition of multidrug-resistant organisms.
      ,
      • Sampathkumar P.
      • Folkert C.
      • Barth J.E.
      • Nation L.
      • Benz M.
      • Hesse A.
      • et al.
      A trial of pulsed xenon ultraviolet disinfection to reduce Clostridioides difficile infection.
      ], seven were conducted as interrupted time series [
      • Brite J.
      • McMillen T.
      • Robilotti E.
      • Sun J.
      • Chow H.Y.
      • Stell F.
      • et al.
      Effectiveness of ultraviolet disinfection in reducing hospital-acquired Clostridium difficile and vancomycin-resistant Enterococcus on a bone marrow transplant unit.
      ,
      • Doll M.E.
      • Zhao J.
      • Kang L.
      • Rittmann B.
      • Alvarez M.
      • Fleming M.
      • et al.
      Chasing the rate: An interrupted time series analysis of interventions targeting reported hospital onset Clostridioides difficile, 2013–2018.
      ,
      • McCord J.
      • Prewitt M.
      • Dyakova E.
      • Mookerjee S.
      • Otter J.A.
      Reduction in Clostridium difficile infection associated with the introduction of hydrogen peroxide vapour automated room disinfection.
      ,
      • Mitchell B.G.
      • Digney W.
      • Locket P.
      • Dancer S.J.
      Controlling methicillin-resistant Staphylococcus aureus (MRSA) in a hospital and the role of hydrogen peroxide decontamination: an interrupted time series analysis.
      ,
      • Murphy P.
      • Kang L.
      • Fleming M.
      • Atkinson C.
      • Pryor R.
      • Cooper K.
      • et al.
      Effect of ultraviolet-C light disinfection at terminal patient discharge on hospital-acquired infections in bone marrow transplant and oncology units.
      ,
      • Pegues D.A.
      • Han J.
      • Gilmar C.
      • McDonnell B.
      • Gaynes S.
      Impact of ultraviolet germicidal irradiation for no-touch terminal room disinfection on Clostridium difficile infection incidence among hematology–oncology patients.
      ,
      • Schaffzin J.K.
      • Wilhite A.W.
      • Li Z.
      • Finney D.
      • Ankrum A.L.
      • Moore R.
      Maximizing efficiency in a high occupancy setting to utilize ultraviolet disinfection for isolation rooms.
      ], and the remainder were conducted as quasi-experimental (uncontrolled before–after) studies [
      • Attia F.
      • Whitener C.
      • Mincemoyer S.
      • Houck J.
      • Julian K.
      The effect of pulsed xenon ultraviolet light disinfection on healthcare-associated Clostridioides difficile rates in a tertiary care hospital.
      ,
      • Boyce J.M.
      • Havill N.L.
      • Otter J.A.
      • McDonald L.C.
      • Adams N.M.T.
      • Cooper T.
      • et al.
      Impact of hydrogen peroxide vapor room decontamination on Clostridium difficile environmental contamination and transmission in a healthcare setting.
      ,
      • Catalanotti A.
      • Abbe D.
      • Simmons S.
      • Stibich M.
      Influence of pulsed-xenon ultraviolet light-based environmental disinfection on surgical site infections.
      ,
      • Green C.
      • Pamplin J.C.
      • Chafin K.N.
      • Murray C.K.
      • Yun H.C.
      Pulsed-xenon ultraviolet light disinfection in a burn unit: Impact on environmental bioburden, multidrug-resistant organism acquisition and healthcare associated infections.
      ,
      • Haas J.P.
      • Menz J.
      • Dusza S.
      • Montecalvo M.A.
      Implementation and impact of ultraviolet environmental disinfection in an acute care setting.
      ,
      • Hardy K.J.
      • Gossain S.
      • Henderson N.
      • Drugan C.
      • Oppenheim B.A.
      • Gao F.
      • et al.
      Rapid recontamination with MRSA of the environment of an intensive care unit after decontamination with hydrogen peroxide vapour.
      ,
      • Horn K.
      • Otter J.A.
      Hydrogen peroxide vapor room disinfection and hand hygiene improvements reduce Clostridium difficile infection, methicillin-resistant Staphylococcus aureus, vancomycin-resistant enterococci, and extended-spectrum beta-lactamase.
      ,
      • Kovach C.R.
      • Taneli Y.
      • Neiman T.
      • Dyer E.M.
      • Arzaga A.J.A.
      • Kelber S.T.
      Evaluation of an ultraviolet room disinfection protocol to decrease nursing home microbial burden, infection and hospitalization rates.
      ,
      • Levin J.
      • Riley L.S.
      • Parrish C.
      • English D.
      • Ahn S.
      The effect of portable pulsed xenon ultraviolet light after terminal cleaning on hospital-associated Clostridium difficile infection in a community hospital.
      ,
      • Manian F.A.
      • Griesnauer S.
      • Bryant A.
      Implementation of hospital-wide enhanced terminal cleaning of targeted patient rooms and its impact on endemic Clostridium difficile infection rates.
      ,
      • McMullen K.
      • Guth R.M.
      • Wood H.
      • Mueller C.
      • Dunn G.
      • Wade R.
      • et al.
      Impact of no-touch ultraviolet light room disinfection systems on Clostridioides difficile infections.
      ,
      • Miller R.
      • Simmons S.
      • Dale C.
      • Stachowiak J.
      • Stibich M.
      Utilization and impact of a pulsed-xenon ultraviolet room disinfection system and multidisciplinary care team on Clostridium difficile in a long-term acute care facility.
      ,
      • Napolitano N.A.
      • Mahapatra T.
      • Tang W.
      The effectiveness of UV-C radiation for facility-wide environmental disinfection to reduce health care-acquired infections.
      ,
      • Raggi R.
      • Archulet K.
      • Haag C.W.
      • Tang W.
      Clinical, operational, and financial impact of an ultraviolet-C terminal disinfection intervention at a community hospital.
      ,
      • Vianna P.G.
      • Dale C.R.
      • Simmons S.
      • Stibich M.
      • Licitra C.M.
      Impact of pulsed xenon ultraviolet light on hospital-acquired infection rates in a community hospital.
      ]. Where controlled before–after studies reported adjusted IRRs these were used to calculate data inputs for the relevant NMAs (see Appendix H). Methodological quality assessments for the included studies are presented according to study design in Tables G.1, G.2, G.3, and G.4, respectively.

      Network meta-analysis

      NMA was performed for the clinical outcomes of infection or acquisition due to Acinetobacter spp. (four studies [
      • Anderson D.J.
      • Chen L.F.
      • Weber D.J.
      • Moehring R.W.
      • Lewis S.S.
      • Triplett P.F.
      • et al.
      Enhanced terminal room disinfection and acquisition and infection caused by multidrug-resistant organisms and Clostridium difficile (the Benefits of Enhanced Terminal Room Disinfection study): a cluster-randomised, multicentre, crossover study.
      ,
      • Morikane K.
      • Suzuki S.
      • Yoshioka J.
      • Yakuwa J.
      • Nakane M.
      • Nemoto K.
      Clinical and microbiological effect of pulsed xenon ultraviolet disinfection to reduce multidrug-resistant organisms in the intensive care unit in a Japanese hospital: a before-after study.
      ,
      • Napolitano N.A.
      • Mahapatra T.
      • Tang W.
      The effectiveness of UV-C radiation for facility-wide environmental disinfection to reduce health care-acquired infections.
      ,
      • Raggi R.
      • Archulet K.
      • Haag C.W.
      • Tang W.
      Clinical, operational, and financial impact of an ultraviolet-C terminal disinfection intervention at a community hospital.
      ]), C. difficile (18 studies [
      • Anderson D.J.
      • Chen L.F.
      • Weber D.J.
      • Moehring R.W.
      • Lewis S.S.
      • Triplett P.F.
      • et al.
      Enhanced terminal room disinfection and acquisition and infection caused by multidrug-resistant organisms and Clostridium difficile (the Benefits of Enhanced Terminal Room Disinfection study): a cluster-randomised, multicentre, crossover study.
      ,
      • Attia F.
      • Whitener C.
      • Mincemoyer S.
      • Houck J.
      • Julian K.
      The effect of pulsed xenon ultraviolet light disinfection on healthcare-associated Clostridioides difficile rates in a tertiary care hospital.
      ,
      • Boyce J.M.
      • Havill N.L.
      • Otter J.A.
      • McDonald L.C.
      • Adams N.M.T.
      • Cooper T.
      • et al.
      Impact of hydrogen peroxide vapor room decontamination on Clostridium difficile environmental contamination and transmission in a healthcare setting.
      ,
      • Boyce J.M.
      • Guercia K.A.
      • Sullivan L.
      • Havill N.L.
      • Fekieta R.
      • Kozakiewicz J.
      • et al.
      Prospective cluster controlled crossover trial to compare the impact of an improved hydrogen peroxide disinfectant and a quaternary ammonium-based disinfectant on surface contamination and health care outcomes.
      ,
      • Brite J.
      • McMillen T.
      • Robilotti E.
      • Sun J.
      • Chow H.Y.
      • Stell F.
      • et al.
      Effectiveness of ultraviolet disinfection in reducing hospital-acquired Clostridium difficile and vancomycin-resistant Enterococcus on a bone marrow transplant unit.
      ,
      • Green C.
      • Pamplin J.C.
      • Chafin K.N.
      • Murray C.K.
      • Yun H.C.
      Pulsed-xenon ultraviolet light disinfection in a burn unit: Impact on environmental bioburden, multidrug-resistant organism acquisition and healthcare associated infections.
      ,
      • Haas J.P.
      • Menz J.
      • Dusza S.
      • Montecalvo M.A.
      Implementation and impact of ultraviolet environmental disinfection in an acute care setting.
      ,
      • Horn K.
      • Otter J.A.
      Hydrogen peroxide vapor room disinfection and hand hygiene improvements reduce Clostridium difficile infection, methicillin-resistant Staphylococcus aureus, vancomycin-resistant enterococci, and extended-spectrum beta-lactamase.
      ,
      • Levin J.
      • Riley L.S.
      • Parrish C.
      • English D.
      • Ahn S.
      The effect of portable pulsed xenon ultraviolet light after terminal cleaning on hospital-associated Clostridium difficile infection in a community hospital.
      ,
      • Manian F.A.
      • Griesnauer S.
      • Bryant A.
      Implementation of hospital-wide enhanced terminal cleaning of targeted patient rooms and its impact on endemic Clostridium difficile infection rates.
      ,
      • McCord J.
      • Prewitt M.
      • Dyakova E.
      • Mookerjee S.
      • Otter J.A.
      Reduction in Clostridium difficile infection associated with the introduction of hydrogen peroxide vapour automated room disinfection.
      ,
      • McMullen K.
      • Guth R.M.
      • Wood H.
      • Mueller C.
      • Dunn G.
      • Wade R.
      • et al.
      Impact of no-touch ultraviolet light room disinfection systems on Clostridioides difficile infections.
      ,
      • Miller R.
      • Simmons S.
      • Dale C.
      • Stachowiak J.
      • Stibich M.
      Utilization and impact of a pulsed-xenon ultraviolet room disinfection system and multidisciplinary care team on Clostridium difficile in a long-term acute care facility.
      ,
      • Napolitano N.A.
      • Mahapatra T.
      • Tang W.
      The effectiveness of UV-C radiation for facility-wide environmental disinfection to reduce health care-acquired infections.
      ,
      • Passaretti C.L.
      • Otter J.A.
      • Reich N.G.
      • Myers J.
      • Shepard J.
      • Ross T.
      • et al.
      An evaluation of environmental decontamination with hydrogen peroxide vapor for reducing the risk of patient acquisition of multidrug-resistant organisms.
      ,
      • Pegues D.A.
      • Han J.
      • Gilmar C.
      • McDonnell B.
      • Gaynes S.
      Impact of ultraviolet germicidal irradiation for no-touch terminal room disinfection on Clostridium difficile infection incidence among hematology–oncology patients.
      ,
      • Sampathkumar P.
      • Folkert C.
      • Barth J.E.
      • Nation L.
      • Benz M.
      • Hesse A.
      • et al.
      A trial of pulsed xenon ultraviolet disinfection to reduce Clostridioides difficile infection.
      ,
      • Vianna P.G.
      • Dale C.R.
      • Simmons S.
      • Stibich M.
      • Licitra C.M.
      Impact of pulsed xenon ultraviolet light on hospital-acquired infection rates in a community hospital.
      ]), MRSA (12 studies [
      • Anderson D.J.
      • Chen L.F.
      • Weber D.J.
      • Moehring R.W.
      • Lewis S.S.
      • Triplett P.F.
      • et al.
      Enhanced terminal room disinfection and acquisition and infection caused by multidrug-resistant organisms and Clostridium difficile (the Benefits of Enhanced Terminal Room Disinfection study): a cluster-randomised, multicentre, crossover study.
      ,
      • Boyce J.M.
      • Guercia K.A.
      • Sullivan L.
      • Havill N.L.
      • Fekieta R.
      • Kozakiewicz J.
      • et al.
      Prospective cluster controlled crossover trial to compare the impact of an improved hydrogen peroxide disinfectant and a quaternary ammonium-based disinfectant on surface contamination and health care outcomes.
      ,
      • Green C.
      • Pamplin J.C.
      • Chafin K.N.
      • Murray C.K.
      • Yun H.C.
      Pulsed-xenon ultraviolet light disinfection in a burn unit: Impact on environmental bioburden, multidrug-resistant organism acquisition and healthcare associated infections.
      ,
      • Haas J.P.
      • Menz J.
      • Dusza S.
      • Montecalvo M.A.
      Implementation and impact of ultraviolet environmental disinfection in an acute care setting.
      ,
      • Horn K.
      • Otter J.A.
      Hydrogen peroxide vapor room disinfection and hand hygiene improvements reduce Clostridium difficile infection, methicillin-resistant Staphylococcus aureus, vancomycin-resistant enterococci, and extended-spectrum beta-lactamase.
      ,
      • Kitagawa H.
      • Mori M.
      • Kawano R.
      • Hara T.
      • Kashiyama S.
      • Hayashi Y.
      • et al.
      Combining pulsed xenon ultraviolet disinfection with terminal manual cleaning helps reduce the acquisition rate of methicillin-resistant Staphylococcus aureus.
      ,
      • Mitchell B.G.
      • Digney W.
      • Locket P.
      • Dancer S.J.
      Controlling methicillin-resistant Staphylococcus aureus (MRSA) in a hospital and the role of hydrogen peroxide decontamination: an interrupted time series analysis.
      ,
      • Morikane K.
      • Suzuki S.
      • Yoshioka J.
      • Yakuwa J.
      • Nakane M.
      • Nemoto K.
      Clinical and microbiological effect of pulsed xenon ultraviolet disinfection to reduce multidrug-resistant organisms in the intensive care unit in a Japanese hospital: a before-after study.
      ,
      • Napolitano N.A.
      • Mahapatra T.
      • Tang W.
      The effectiveness of UV-C radiation for facility-wide environmental disinfection to reduce health care-acquired infections.
      ,
      • Passaretti C.L.
      • Otter J.A.
      • Reich N.G.
      • Myers J.
      • Shepard J.
      • Ross T.
      • et al.
      An evaluation of environmental decontamination with hydrogen peroxide vapor for reducing the risk of patient acquisition of multidrug-resistant organisms.
      ,
      • Raggi R.
      • Archulet K.
      • Haag C.W.
      • Tang W.
      Clinical, operational, and financial impact of an ultraviolet-C terminal disinfection intervention at a community hospital.
      ,
      • Vianna P.G.
      • Dale C.R.
      • Simmons S.
      • Stibich M.
      • Licitra C.M.
      Impact of pulsed xenon ultraviolet light on hospital-acquired infection rates in a community hospital.
      ]), and VRE (10 studies [
      • Anderson D.J.
      • Chen L.F.
      • Weber D.J.
      • Moehring R.W.
      • Lewis S.S.
      • Triplett P.F.
      • et al.
      Enhanced terminal room disinfection and acquisition and infection caused by multidrug-resistant organisms and Clostridium difficile (the Benefits of Enhanced Terminal Room Disinfection study): a cluster-randomised, multicentre, crossover study.
      ,
      • Boyce J.M.
      • Guercia K.A.
      • Sullivan L.
      • Havill N.L.
      • Fekieta R.
      • Kozakiewicz J.
      • et al.
      Prospective cluster controlled crossover trial to compare the impact of an improved hydrogen peroxide disinfectant and a quaternary ammonium-based disinfectant on surface contamination and health care outcomes.
      ,
      • Brite J.
      • McMillen T.
      • Robilotti E.
      • Sun J.
      • Chow H.Y.
      • Stell F.
      • et al.
      Effectiveness of ultraviolet disinfection in reducing hospital-acquired Clostridium difficile and vancomycin-resistant Enterococcus on a bone marrow transplant unit.
      ,
      • Haas J.P.
      • Menz J.
      • Dusza S.
      • Montecalvo M.A.
      Implementation and impact of ultraviolet environmental disinfection in an acute care setting.
      ,
      • Horn K.
      • Otter J.A.
      Hydrogen peroxide vapor room disinfection and hand hygiene improvements reduce Clostridium difficile infection, methicillin-resistant Staphylococcus aureus, vancomycin-resistant enterococci, and extended-spectrum beta-lactamase.
      ,
      • Napolitano N.A.
      • Mahapatra T.
      • Tang W.
      The effectiveness of UV-C radiation for facility-wide environmental disinfection to reduce health care-acquired infections.
      ,
      • Passaretti C.L.
      • Otter J.A.
      • Reich N.G.
      • Myers J.
      • Shepard J.
      • Ross T.
      • et al.
      An evaluation of environmental decontamination with hydrogen peroxide vapor for reducing the risk of patient acquisition of multidrug-resistant organisms.
      ,
      • Raggi R.
      • Archulet K.
      • Haag C.W.
      • Tang W.
      Clinical, operational, and financial impact of an ultraviolet-C terminal disinfection intervention at a community hospital.
      ,
      • Sampathkumar P.
      • Folkert C.
      • Barth J.E.
      • Nation L.
      • Benz M.
      • Hesse A.
      • et al.
      A trial of pulsed xenon ultraviolet disinfection to reduce Clostridioides difficile infection.
      ,
      • Vianna P.G.
      • Dale C.R.
      • Simmons S.
      • Stibich M.
      • Licitra C.M.
      Impact of pulsed xenon ultraviolet light on hospital-acquired infection rates in a community hospital.
      ]). The data inputs for the NMAs (including loge-IRRs and associated SEs) are presented in Tables H.1, H.2, H.3 and H.4, respectively. One multi-arm study [
      • Anderson D.J.
      • Chen L.F.
      • Weber D.J.
      • Moehring R.W.
      • Lewis S.S.
      • Triplett P.F.
      • et al.
      Enhanced terminal room disinfection and acquisition and infection caused by multidrug-resistant organisms and Clostridium difficile (the Benefits of Enhanced Terminal Room Disinfection study): a cluster-randomised, multicentre, crossover study.
      ] was included in several of the NMAs. This study compared four treatments: UV-C after bleach disinfection; UV-C after standard manual cleaning/disinfection; bleach disinfection; and standard manual cleaning/disinfection. Where relevant, the NMAs incorporated adjustments for correlations between IRRs involving the three alternatives to standard manual cleaning/disinfection in this study. Another study [
      • McMullen K.
      • Guth R.M.
      • Wood H.
      • Mueller C.
      • Dunn G.
      • Wade R.
      • et al.
      Impact of no-touch ultraviolet light room disinfection systems on Clostridioides difficile infections.
      ] evaluated both UV-C and PX-UV through comparisons with manual cleaning/disinfection, but these comparisons were conducted in different hospitals and contributed statistically independent IRRs to the relevant NMA. One study comparing different approaches to manual cleaning/disinfection [
      • Boyce J.M.
      • Guercia K.A.
      • Sullivan L.
      • Havill N.L.
      • Fekieta R.
      • Kozakiewicz J.
      • et al.
      Prospective cluster controlled crossover trial to compare the impact of an improved hydrogen peroxide disinfectant and a quaternary ammonium-based disinfectant on surface contamination and health care outcomes.
      ] was also included in several of the NMAs. This ensured that the approaches to manual cleaning/disinfection represented in the analyses reflected the wide variety of approaches that might be used in practice (see Appendix H). However, the purpose of the analyses was not to compare the effectiveness of different approaches to manual cleaning/disinfection per se.
      Network diagrams corresponding to each NMA are presented in Figure H.1. The automated approaches to decontamination represented in the NMAs were UV-C, PX-UV, HPV, and AHP. The total number of patient-days represented in the networks of evidence was greatest for C. difficile (more than 3.5 million patient-days), lower for MRSA and VRE (∼1.5 million patient-days each), and lowest for Acinetobacter spp. (∼170,000 patient-days). None of the included studies involved head-to-head comparisons between different approaches to automated decontamination. The resulting star networks, in which the only direct comparisons were those between automated decontamination systems and the reference treatment (manual cleaning/disinfection), did not allow investigation of the consistency assumption underpinning each NMA, but the width of the 95% CrIs for indirect treatment effect estimates was taken into account when determining GRADE quality ratings for the domain of imprecision (see Appendix H). The transitivity assumption that also underpins each NMA was expected to hold because it was plausible that any automated decontamination system could have been implemented in any of the study settings represented in the networks of evidence.
      The numerical results (IRRs for relevant treatment contrasts and SUCRA scores) from the NMAs are presented in Tables H.5, H.6, H.7, H.8 and H.9. The IRRs (and 95% CrIs) are presented graphically in Figure 1.
      Figure 1
      Figure 1Forest plots for network meta-analysis of the clinical outcomes of infection or acquisition. IRRs with 95% CrIs that do not cross the line of no effect (IRR = 1) shown in red; upper limits of 95% CrIs truncated at ∼2 (see Appendix H for exact results); dashed vertical lines show thresholds for defining imprecision (IRR = 0.8 and IRR = 1.25). AHP, aerosolized hydrogen peroxide; CrI, credible interval; HPV, hydrogen peroxide vapour; IRR, incidence rate ratio; MRSA, meticillin-resistant Staphylococcus aureus; PX-UV, pulsed-xenon ultraviolet; UV-C, ultraviolet C; VRE, vancomycin-resistant enterococcus.

      GRADE tables

      A separate GRADE table was constructed for each type of evidence identified for the clinical outcomes of infection or acquisition. Thus, GRADE tables were produced for infection or acquisition due to the specific micro-organisms Acinetobacter spp., C. difficile, Klebsiella pneumoniae, MRSA, Pseudomonas aeruginosa, Stenotrophomonas maltophilia, and VRE (see Tables I.1, I.2, I.3, I.4, I.5, I.6 and I.7, respectively) and for the groups of micro-organisms Enterobacterales (including extended-spectrum β-lactamase-producing (ESBL) Entero-bacterales see Table I.8), multidrug-resistant Gram-negative rods (MDR-GNR; see Table I.9), extended-spectrum β-lactamase-producing Gram-negative bacteria (ESBL-GNB; see Table I.10), and multidrug-resistant Gram-negative bacteria (MDR-GNB; see Table I.11). The GRADE tables related to infection or acquisition due to specific micro-organisms refer to NMA results where relevant. Further GRADE tables summarize evidence for surgical site infection (see Table I.12), device-associated infection (CAUTI, CLABSI, and ventilator-associated pneumonia; see Table I.13), and infection of specific body organs or systems (enteric infection, respiratory infection, skin and soft tissue infections, and urinary tract infection (UTI); see Table I.14).
      All the evidence was assigned an overall quality rating of very low or low. Frequently occurring reasons for downgrading the quality of individual outcomes were serious risk of bias (as identified through the methodological quality assessments based on study design referred to above) and serious (or very serious) imprecision (where 95% CrIs or CIs for relative treatment effects such as IRRs or odds ratios (ORs) crossed one (or both) prespecified thresholds of 0.8 and 1.25).

      Evidence statements

      Acinetobacter spp. infection or acquisition

      There was weak evidence from an NMA based on a controlled trial [
      • Anderson D.J.
      • Chen L.F.
      • Weber D.J.
      • Moehring R.W.
      • Lewis S.S.
      • Triplett P.F.
      • et al.
      Enhanced terminal room disinfection and acquisition and infection caused by multidrug-resistant organisms and Clostridium difficile (the Benefits of Enhanced Terminal Room Disinfection study): a cluster-randomised, multicentre, crossover study.
      ], a controlled before–after study [
      • Morikane K.
      • Suzuki S.
      • Yoshioka J.
      • Yakuwa J.
      • Nakane M.
      • Nemoto K.
      Clinical and microbiological effect of pulsed xenon ultraviolet disinfection to reduce multidrug-resistant organisms in the intensive care unit in a Japanese hospital: a before-after study.
      ] and two uncontrolled before–after studies [
      • Napolitano N.A.
      • Mahapatra T.
      • Tang W.
      The effectiveness of UV-C radiation for facility-wide environmental disinfection to reduce health care-acquired infections.
      ,
      • Raggi R.
      • Archulet K.
      • Haag C.W.
      • Tang W.
      Clinical, operational, and financial impact of an ultraviolet-C terminal disinfection intervention at a community hospital.
      ] that using UV-C or PX-UV in addition to manual cleaning/disinfection might reduce the incidence of Acinetobacter spp. infection or acquisition in subsequent patients compared with manual cleaning/disinfection alone. However, the reductions were not statistically significant, nor was the difference in effectiveness between UV-C and PX-UV (UV-C versus manual cleaning/disinfection, IRR = 0.376 (95% CrI 0.068 to 1.638); PX-UV versus manual cleaning/disinfection, IRR = 0.370 (95% CrI 0.023 to 5.755); PX-UV versus UV-C, IRR = 0.987 (95% CrI 0.046 to 26.310)). Treatment rankings (from best to worst) based on SUCRA scores were as follows: UV-C (SUCRA = 36%); PX-UV (SUCRA = 34%); and manual cleaning/disinfection (SUCRA = 11%).

      Clostridioides difficile infection or acquisition

      There was moderate evidence from an NMA based on two controlled trials [
      • Anderson D.J.
      • Chen L.F.
      • Weber D.J.
      • Moehring R.W.
      • Lewis S.S.
      • Triplett P.F.
      • et al.
      Enhanced terminal room disinfection and acquisition and infection caused by multidrug-resistant organisms and Clostridium difficile (the Benefits of Enhanced Terminal Room Disinfection study): a cluster-randomised, multicentre, crossover study.
      ,
      • Boyce J.M.
      • Guercia K.A.
      • Sullivan L.
      • Havill N.L.
      • Fekieta R.
      • Kozakiewicz J.
      • et al.
      Prospective cluster controlled crossover trial to compare the impact of an improved hydrogen peroxide disinfectant and a quaternary ammonium-based disinfectant on surface contamination and health care outcomes.
      ], two controlled before–after studies [
      • Passaretti C.L.
      • Otter J.A.
      • Reich N.G.
      • Myers J.
      • Shepard J.
      • Ross T.
      • et al.
      An evaluation of environmental decontamination with hydrogen peroxide vapor for reducing the risk of patient acquisition of multidrug-resistant organisms.
      ,
      • Sampathkumar P.
      • Folkert C.
      • Barth J.E.
      • Nation L.
      • Benz M.
      • Hesse A.
      • et al.
      A trial of pulsed xenon ultraviolet disinfection to reduce Clostridioides difficile infection.
      ], three interrupted time series [
      • Brite J.
      • McMillen T.
      • Robilotti E.
      • Sun J.
      • Chow H.Y.
      • Stell F.
      • et al.
      Effectiveness of ultraviolet disinfection in reducing hospital-acquired Clostridium difficile and vancomycin-resistant Enterococcus on a bone marrow transplant unit.
      ,
      • McCord J.
      • Prewitt M.
      • Dyakova E.
      • Mookerjee S.
      • Otter J.A.
      Reduction in Clostridium difficile infection associated with the introduction of hydrogen peroxide vapour automated room disinfection.
      ,
      • Pegues D.A.
      • Han J.
      • Gilmar C.
      • McDonnell B.
      • Gaynes S.
      Impact of ultraviolet germicidal irradiation for no-touch terminal room disinfection on Clostridium difficile infection incidence among hematology–oncology patients.
      ], and 11 uncontrolled before–after studies [
      • Attia F.
      • Whitener C.
      • Mincemoyer S.
      • Houck J.
      • Julian K.
      The effect of pulsed xenon ultraviolet light disinfection on healthcare-associated Clostridioides difficile rates in a tertiary care hospital.
      ,
      • Boyce J.M.
      • Havill N.L.
      • Otter J.A.
      • McDonald L.C.
      • Adams N.M.T.
      • Cooper T.
      • et al.
      Impact of hydrogen peroxide vapor room decontamination on Clostridium difficile environmental contamination and transmission in a healthcare setting.
      ,
      • Green C.
      • Pamplin J.C.
      • Chafin K.N.
      • Murray C.K.
      • Yun H.C.
      Pulsed-xenon ultraviolet light disinfection in a burn unit: Impact on environmental bioburden, multidrug-resistant organism acquisition and healthcare associated infections.
      ,
      • Haas J.P.
      • Menz J.
      • Dusza S.
      • Montecalvo M.A.
      Implementation and impact of ultraviolet environmental disinfection in an acute care setting.
      ,
      • Horn K.
      • Otter J.A.
      Hydrogen peroxide vapor room disinfection and hand hygiene improvements reduce Clostridium difficile infection, methicillin-resistant Staphylococcus aureus, vancomycin-resistant enterococci, and extended-spectrum beta-lactamase.
      ,
      • Levin J.
      • Riley L.S.
      • Parrish C.
      • English D.
      • Ahn S.
      The effect of portable pulsed xenon ultraviolet light after terminal cleaning on hospital-associated Clostridium difficile infection in a community hospital.
      ,
      • Manian F.A.
      • Griesnauer S.
      • Bryant A.
      Implementation of hospital-wide enhanced terminal cleaning of targeted patient rooms and its impact on endemic Clostridium difficile infection rates.
      ,
      • McMullen K.
      • Guth R.M.
      • Wood H.
      • Mueller C.
      • Dunn G.
      • Wade R.
      • et al.
      Impact of no-touch ultraviolet light room disinfection systems on Clostridioides difficile infections.
      ,
      • Miller R.
      • Simmons S.
      • Dale C.
      • Stachowiak J.
      • Stibich M.
      Utilization and impact of a pulsed-xenon ultraviolet room disinfection system and multidisciplinary care team on Clostridium difficile in a long-term acute care facility.
      ,
      • Napolitano N.A.
      • Mahapatra T.
      • Tang W.
      The effectiveness of UV-C radiation for facility-wide environmental disinfection to reduce health care-acquired infections.
      ,
      • Vianna P.G.
      • Dale C.R.
      • Simmons S.
      • Stibich M.
      • Licitra C.M.
      Impact of pulsed xenon ultraviolet light on hospital-acquired infection rates in a community hospital.
      ] that using UV-C, PX-UV or HPV in addition to manual cleaning/disinfection reduced the incidence of C. difficile infection compared with manual cleaning/disinfection alone (UV-C versus manual cleaning/disinfection, IRR = 0.822 (95% CrI 0.525 to 1.258); PX-UV versus manual cleaning/disinfection, IRR = 0.761 (95% CrI 0.571 to 0.972); HPV versus manual cleaning/disinfection, IRR = 0.532 (95% CrI 0.372 to 0.755); PX-UV versus UV-C, IRR = 0.925 (95% CrI 0.553 to 1.531); HPV versus UV-C, IRR = 0.646 (95% CrI 0.373 to 1.145); HPV versus PX-UV, IRR = 0.699 (95% CrI 0.458 to 1.108)). Treatment rankings (from best to worst) based on SUCRA scores were as follows: HPV (SUCRA = 72%); PX-UV (SUCRA = 42%); UV-C (SUCRA = 32%); and manual cleaning/disinfection (SUCRA = 6%).
      There was inconsistent evidence regarding C. difficile infection from an interrupted time series [
      • Murphy P.
      • Kang L.
      • Fleming M.
      • Atkinson C.
      • Pryor R.
      • Cooper K.
      • et al.
      Effect of ultraviolet-C light disinfection at terminal patient discharge on hospital-acquired infections in bone marrow transplant and oncology units.
      ] comparing UV-C at every terminal discharge to UV-C only at terminal discharge of patients with C. difficile infection. In a bone marrow transplant unit, using UV-C at every terminal discharge reduced the baseline incidence of C. difficile infection and slowed the rate of increase over time compared with using UV-C only at terminal discharge of patients with C. difficile infection. The change in baseline incidence was statistically significant, but not the change in the rate of increase over time (segmented regression change in intercept, P = 0.044; change in slope, P = 0.417).

      Klebsiella pneumoniae infection or acquisition

      There was weak evidence regarding K. pneumoniae infection from two uncontrolled before–after studies [
      • Napolitano N.A.
      • Mahapatra T.
      • Tang W.
      The effectiveness of UV-C radiation for facility-wide environmental disinfection to reduce health care-acquired infections.
      ,
      • Raggi R.
      • Archulet K.
      • Haag C.W.
      • Tang W.
      Clinical, operational, and financial impact of an ultraviolet-C terminal disinfection intervention at a community hospital.
      ] comparing UV-C to manual cleaning/disinfection. There were no statistically significant differences between UV-C and manual cleaning/disinfection.

      Meticillin-resistant Staphylococcus aureus infection or acquisition

      There was weak evidence from an NMA based on two controlled trials [
      • Anderson D.J.
      • Chen L.F.
      • Weber D.J.
      • Moehring R.W.
      • Lewis S.S.
      • Triplett P.F.
      • et al.
      Enhanced terminal room disinfection and acquisition and infection caused by multidrug-resistant organisms and Clostridium difficile (the Benefits of Enhanced Terminal Room Disinfection study): a cluster-randomised, multicentre, crossover study.
      ,
      • Boyce J.M.
      • Guercia K.A.
      • Sullivan L.
      • Havill N.L.
      • Fekieta R.
      • Kozakiewicz J.
      • et al.
      Prospective cluster controlled crossover trial to compare the impact of an improved hydrogen peroxide disinfectant and a quaternary ammonium-based disinfectant on surface contamination and health care outcomes.
      ], three controlled before–after studies [
      • Kitagawa H.
      • Mori M.
      • Kawano R.
      • Hara T.
      • Kashiyama S.
      • Hayashi Y.
      • et al.
      Combining pulsed xenon ultraviolet disinfection with terminal manual cleaning helps reduce the acquisition rate of methicillin-resistant Staphylococcus aureus.
      ,
      • Morikane K.
      • Suzuki S.
      • Yoshioka J.
      • Yakuwa J.
      • Nakane M.
      • Nemoto K.
      Clinical and microbiological effect of pulsed xenon ultraviolet disinfection to reduce multidrug-resistant organisms in the intensive care unit in a Japanese hospital: a before-after study.
      ,
      • Passaretti C.L.
      • Otter J.A.
      • Reich N.G.
      • Myers J.
      • Shepard J.
      • Ross T.
      • et al.
      An evaluation of environmental decontamination with hydrogen peroxide vapor for reducing the risk of patient acquisition of multidrug-resistant organisms.
      ], an interrupted time series [
      • Mitchell B.G.
      • Digney W.
      • Locket P.
      • Dancer S.J.
      Controlling methicillin-resistant Staphylococcus aureus (MRSA) in a hospital and the role of hydrogen peroxide decontamination: an interrupted time series analysis.
      ], and six uncontrolled before–after studies [
      • Green C.
      • Pamplin J.C.
      • Chafin K.N.
      • Murray C.K.
      • Yun H.C.
      Pulsed-xenon ultraviolet light disinfection in a burn unit: Impact on environmental bioburden, multidrug-resistant organism acquisition and healthcare associated infections.
      ,
      • Haas J.P.
      • Menz J.
      • Dusza S.
      • Montecalvo M.A.
      Implementation and impact of ultraviolet environmental disinfection in an acute care setting.
      ,
      • Horn K.
      • Otter J.A.
      Hydrogen peroxide vapor room disinfection and hand hygiene improvements reduce Clostridium difficile infection, methicillin-resistant Staphylococcus aureus, vancomycin-resistant enterococci, and extended-spectrum beta-lactamase.
      ,
      • Napolitano N.A.
      • Mahapatra T.
      • Tang W.
      The effectiveness of UV-C radiation for facility-wide environmental disinfection to reduce health care-acquired infections.
      ,
      • Raggi R.
      • Archulet K.
      • Haag C.W.
      • Tang W.
      Clinical, operational, and financial impact of an ultraviolet-C terminal disinfection intervention at a community hospital.
      ,
      • Vianna P.G.
      • Dale C.R.
      • Simmons S.
      • Stibich M.
      • Licitra C.M.
      Impact of pulsed xenon ultraviolet light on hospital-acquired infection rates in a community hospital.
      ] that using UV-C, PX-UV, HPV, or AHP in addition to manual cleaning/disinfection reduced the incidence of MRSA infection or acquisition compared with manual cleaning/disinfection alone. The reduction was statistically significant with PX-UV, but not with UV-C, HPV, or AHP, nor when comparing differences in effectiveness between UV-C, PX-UV, HPV, and AHP (UV-C versus manual cleaning/disinfection, IRR = 0.838 (95% CrI 0.656 to 1.052); PX-UV versus manual cleaning/disinfection, IRR = 0.760 (95% CrI 0.621 to 0.966); HPV versus manual cleaning/disinfection, IRR = 0.554 (95% CrI 0.272 to 1.150); AHP versus manual cleaning/disinfection, IRR = 0.701 (95% CrI 0.170 to 2.677). Treatment rankings (from best to worst) based on SUCRA scores were as follows: HPV (SUCRA = 81%); PX-UV (SUCRA = 60%); AHP (SUCRA = 56%); UV-C (SUCRA = 44%); and manual cleaning/disinfection (SUCRA = 11%).

      Pseudomonas aeruginosa infection or acquisition

      There was moderate evidence from an uncontrolled before–after study [
      • Raggi R.
      • Archulet K.
      • Haag C.W.
      • Tang W.
      Clinical, operational, and financial impact of an ultraviolet-C terminal disinfection intervention at a community hospital.
      ] that using UV-C in addition to manual cleaning/disinfection did not significantly reduce the incidence of P. aeruginosa infection compared with manual cleaning/disinfection alone (IRR = 0.871 (95% CI 0.634 to 1.197)).
      There was weak evidence from an uncontrolled before–after study [
      • Green C.
      • Pamplin J.C.
      • Chafin K.N.
      • Murray C.K.
      • Yun H.C.
      Pulsed-xenon ultraviolet light disinfection in a burn unit: Impact on environmental bioburden, multidrug-resistant organism acquisition and healthcare associated infections.
      ] that using PX-UV in addition to manual cleaning/disinfection did not significantly reduce the incidence of multidrug-resistant P. aeruginosa acquisition compared with manual cleaning/disinfection alone (IRR = 0.670 (95% CI 0.032 to 13.947)).

      Stenotrophomonas maltophilia infection or acquisition

      There was weak evidence from an uncontrolled before–after study [
      • Green C.
      • Pamplin J.C.
      • Chafin K.N.
      • Murray C.K.
      • Yun H.C.
      Pulsed-xenon ultraviolet light disinfection in a burn unit: Impact on environmental bioburden, multidrug-resistant organism acquisition and healthcare associated infections.
      ] that using PX-UV in addition to manual cleaning/disinfection did not significantly change the incidence of S. maltophilia acquisition compared with manual cleaning/disinfection alone (IRR = 2.511 (95% CI 0.562 to 11.218)).

      Vancomycin-resistant enterococcus infection or acquisition

      There was moderate evidence from an NMA based on two controlled trials [
      • Anderson D.J.
      • Chen L.F.
      • Weber D.J.
      • Moehring R.W.
      • Lewis S.S.
      • Triplett P.F.
      • et al.
      Enhanced terminal room disinfection and acquisition and infection caused by multidrug-resistant organisms and Clostridium difficile (the Benefits of Enhanced Terminal Room Disinfection study): a cluster-randomised, multicentre, crossover study.
      ,
      • Boyce J.M.
      • Guercia K.A.
      • Sullivan L.
      • Havill N.L.
      • Fekieta R.
      • Kozakiewicz J.
      • et al.
      Prospective cluster controlled crossover trial to compare the impact of an improved hydrogen peroxide disinfectant and a quaternary ammonium-based disinfectant on surface contamination and health care outcomes.
      ], two controlled before–after studies [
      • Passaretti C.L.
      • Otter J.A.
      • Reich N.G.
      • Myers J.
      • Shepard J.
      • Ross T.
      • et al.
      An evaluation of environmental decontamination with hydrogen peroxide vapor for reducing the risk of patient acquisition of multidrug-resistant organisms.
      ,
      • Sampathkumar P.
      • Folkert C.
      • Barth J.E.
      • Nation L.
      • Benz M.
      • Hesse A.
      • et al.
      A trial of pulsed xenon ultraviolet disinfection to reduce Clostridioides difficile infection.
      ], an interrupted time series [
      • Brite J.
      • McMillen T.
      • Robilotti E.
      • Sun J.
      • Chow H.Y.
      • Stell F.
      • et al.
      Effectiveness of ultraviolet disinfection in reducing hospital-acquired Clostridium difficile and vancomycin-resistant Enterococcus on a bone marrow transplant unit.
      ], and five uncontrolled before–after studies [
      • Haas J.P.
      • Menz J.
      • Dusza S.
      • Montecalvo M.A.
      Implementation and impact of ultraviolet environmental disinfection in an acute care setting.
      ,
      • Horn K.
      • Otter J.A.
      Hydrogen peroxide vapor room disinfection and hand hygiene improvements reduce Clostridium difficile infection, methicillin-resistant Staphylococcus aureus, vancomycin-resistant enterococci, and extended-spectrum beta-lactamase.
      ,
      • Napolitano N.A.
      • Mahapatra T.
      • Tang W.
      The effectiveness of UV-C radiation for facility-wide environmental disinfection to reduce health care-acquired infections.
      ,
      • Raggi R.
      • Archulet K.
      • Haag C.W.
      • Tang W.
      Clinical, operational, and financial impact of an ultraviolet-C terminal disinfection intervention at a community hospital.
      ,
      • Vianna P.G.
      • Dale C.R.
      • Simmons S.
      • Stibich M.
      • Licitra C.M.
      Impact of pulsed xenon ultraviolet light on hospital-acquired infection rates in a community hospital.
      ] that using UV-C, PX-UV or HPV in addition to manual cleaning/disinfection reduced the incidence of VRE infection or acquisition compared with manual cleaning/disinfection alone. The reduction was statistically significant with HPV, but not with UV-C or PX-UV; HPV also reduced the incidence of VRE infection or acquisition compared with UV-C and PX-UV and these reductions were statistically significant (UV-C versus manual cleaning/disinfection, IRR = 0.626 (95% CrI 0.376 to 1.075); PX-UV versus manual cleaning/disinfection, IRR = 0.740 (95% CrI 0.427 to 1.139); HPV versus manual cleaning/disinfection, IRR = 0.180 (95% CrI 0.060 to 0.482)). Treatment rankings (from best to worst) based on SUCRA scores were as follows: HPV (SUCRA = 74%); UV-C (SUCRA = 42%); PX-UV (SUCRA = 31%); and manual cleaning/disinfection (SUCRA = 3%).

      Enterobacterales infection or acquisition

      There was weak evidence from an uncontrolled before–after study [
      • Green C.
      • Pamplin J.C.
      • Chafin K.N.
      • Murray C.K.
      • Yun H.C.
      Pulsed-xenon ultraviolet light disinfection in a burn unit: Impact on environmental bioburden, multidrug-resistant organism acquisition and healthcare associated infections.
      ] that using PX-UV in addition to manual cleaning/disinfection did not significantly change the incidence of ESBL Enterobacterales acquisition compared with manual cleaning/disinfection alone (IRR = 1.674 (95% CI 0.152 to 18.460)).

      Multidrug-resistant Gram-negative rod infection or acquisition

      There was weak evidence from an uncontrolled before–after study [
      • Green C.
      • Pamplin J.C.
      • Chafin K.N.
      • Murray C.K.
      • Yun H.C.
      Pulsed-xenon ultraviolet light disinfection in a burn unit: Impact on environmental bioburden, multidrug-resistant organism acquisition and healthcare associated infections.
      ] that using PX-UV in addition to manual cleaning/disinfection did not significantly change the incidence of MDR-GNR acquisition compared with manual cleaning/disinfection alone (IRR = 1.674 (95% CI 0.504 to 5.559)).
      There was weak evidence from a controlled before–after study [
      • Passaretti C.L.
      • Otter J.A.
      • Reich N.G.
      • Myers J.
      • Shepard J.
      • Ross T.
      • et al.
      An evaluation of environmental decontamination with hydrogen peroxide vapor for reducing the risk of patient acquisition of multidrug-resistant organisms.
      ] that using HPV in addition to manual cleaning/disinfection did not significantly change the incidence of MDR-GNR acquisition compared with manual cleaning/disinfection alone (IRR = 0.715 (95% CI 0.307 to 1.667)).

      Extended-spectrum β-lactamase-producing Gram-negative bacterial infection or acquisition

      There was moderate evidence from an uncontrolled before–after study [
      • Horn K.
      • Otter J.A.
      Hydrogen peroxide vapor room disinfection and hand hygiene improvements reduce Clostridium difficile infection, methicillin-resistant Staphylococcus aureus, vancomycin-resistant enterococci, and extended-spectrum beta-lactamase.
      ] that using HPV in addition to manual cleaning/disinfection reduced the incidence of ESBL-GNB acquisition compared with manual cleaning/disinfection alone. The reduction was statistically significant (IRR = 0.063 (95% CI 0.008 to 0.500)).

      Multidrug-resistant Gram-negative bacterial infection or acquisition

      There was moderate evidence from an uncontrolled before–after study [
      • Haas J.P.
      • Menz J.
      • Dusza S.
      • Montecalvo M.A.
      Implementation and impact of ultraviolet environmental disinfection in an acute care setting.
      ] that using PX-UV in addition to manual cleaning/disinfection reduced the incidence of MDR-GNB acquisition compared with manual cleaning/disinfection alone. The reduction was statistically significant (IRR = 0.81 (95% CI 0.66 to 0.98)).

      Surgical site infection

      There was inconsistent evidence regarding surgical site infection from an uncontrolled before–after study [
      • Catalanotti A.
      • Abbe D.
      • Simmons S.
      • Stibich M.
      Influence of pulsed-xenon ultraviolet light-based environmental disinfection on surgical site infections.
      ] comparing PX-UV to manual cleaning/disinfection. Among class 1 (clean wound) surgical procedures, using PX-UV in addition to manual cleaning/disinfection reduced the incidence of surgical site infection. The reduction was statistically significant (IRR = 0.553 (95% CI 0.334 to 0.918)). However, among class 2 (clean contaminated wound) procedures, using PX-UV in addition to manual cleaning/disinfection increased the incidence of surgical site infection. The increase was not statistically significant (IRR = 1.230 (95% CI 0.632 to 2.393)).
      There was moderate evidence from a controlled before–after study [
      • Murrell L.J.
      • Hamilton E.K.
      • Johnson H.B.
      • Spencer M.
      Influence of a visible-light continuous environmental disinfection system on microbial contamination and surgical site infections in an orthopedic operating room.
      ] that using visible (indigo and white) light in addition to manual cleaning/disinfection reduced the incidence of surgical site infection compared with manual cleaning/disinfection alone. The reduction was statistically significant (adjusted OR = 0.22 (95% CI 0.05 to 0.90)).

      Device-associated infection

      There was weak evidence regarding device-associated infection from an uncontrolled before–after study [
      • Green C.
      • Pamplin J.C.
      • Chafin K.N.
      • Murray C.K.
      • Yun H.C.
      Pulsed-xenon ultraviolet light disinfection in a burn unit: Impact on environmental bioburden, multidrug-resistant organism acquisition and healthcare associated infections.
      ]. Using PX-UV in addition to manual cleaning/disinfection, there was no statistically significant change in the incidence of CAUTI or CLABSI compared with manual cleaning/disinfection alone (P = 0.23 and P = 0.20, respectively).
      There was weak evidence regarding CLABSI from an interrupted time series [
      • Murphy P.
      • Kang L.
      • Fleming M.
      • Atkinson C.
      • Pryor R.
      • Cooper K.
      • et al.
      Effect of ultraviolet-C light disinfection at terminal patient discharge on hospital-acquired infections in bone marrow transplant and oncology units.
      ] comparing UV-C at every terminal discharge to UV-C only at terminal discharge of patients with C. difficile infection. In a bone marrow transplant unit, using UV-C at every terminal discharge reduced the baseline incidence of CLABSI and slowed the rate of increase over time compared with using UV-C only at terminal discharge of patients with C. difficile infection. The change in baseline incidence was statistically significant, but not the change in the rate of increase over time (segmented regression change in intercept, P = 0.048 and change in slope, P = 0.204).
      There was weak evidence regarding ventilator-associated pneumonia from an uncontrolled before–after study [
      • Green C.
      • Pamplin J.C.
      • Chafin K.N.
      • Murray C.K.
      • Yun H.C.
      Pulsed-xenon ultraviolet light disinfection in a burn unit: Impact on environmental bioburden, multidrug-resistant organism acquisition and healthcare associated infections.
      ]. Using PX-UV in addition to manual cleaning/disinfection, there was no statistically significant change in the incidence of ventilator-associated pneumonia compared with manual cleaning/disinfection alone (P = 0.12).

      Infection of specific body organs or systems

      There was weak evidence regarding infection of specific body organs or systems from an uncontrolled before–after study [
      • Kovach C.R.
      • Taneli Y.
      • Neiman T.
      • Dyer E.M.
      • Arzaga A.J.A.
      • Kelber S.T.
      Evaluation of an ultraviolet room disinfection protocol to decrease nursing home microbial burden, infection and hospitalization rates.
      ] comparing PX-UV to manual cleaning/disinfection. Using PX-UV in addition to manual cleaning/disinfection, there were no inferential analyses reported regarding the incidence of enteric infection compared with manual cleaning/disinfection alone. However, there were statistically significant differences in the incidence of respiratory system infections, skin and soft tissue infections, and UTIs (P = 0.017, P = 0.014, and P = 0.014, respectively).
      There was weak evidence regarding respiratory viral infection from an interrupted time series [
      • Murphy P.
      • Kang L.
      • Fleming M.
      • Atkinson C.
      • Pryor R.
      • Cooper K.
      • et al.
      Effect of ultraviolet-C light disinfection at terminal patient discharge on hospital-acquired infections in bone marrow transplant and oncology units.
      ] comparing UV-C at every terminal discharge to UV-C only at terminal discharge of patients with C. difficile infection. In a bone marrow transplant unit and an oncology unit, using UV-C at every terminal discharge did not significantly change the baseline incidence of respiratory viral infection or the rate of increase over time compared with using UV-C only at terminal discharge of patients with C. difficile infection.

      Interpretation of the evidence

      Outcomes that matter most

      The Working Party focused on the clinical outcomes of infection or acquisition as these reflect the direct impact on patients in healthcare settings when exposed to microbiological environmental contamination. In many studies infection and acquisition were not distinguished, although all included studies that evaluated C. difficile reported infection as the clinical outcome. The Working Party included evidence regarding surgical site infection, device-associated infection and infection of specific body organs or systems while noting that this does not identify specific target micro-organisms.

      Quality of the evidence

      All the evidence was of very low or low quality, reflecting potential for bias in the design, analysis and reporting of individual studies, and in many cases imprecise estimates of treatment effects (as reflected by wide CrIs/CIs or those that crossed predetermined thresholds for precision). The Working Party emphasized the potential for bias when reporting study results in research funded by manufacturers of devices being evaluated.
      One study evaluated the effectiveness of AHP compared with manual cleaning/disinfection [
      • Mitchell B.G.
      • Digney W.
      • Locket P.
      • Dancer S.J.
      Controlling methicillin-resistant Staphylococcus aureus (MRSA) in a hospital and the role of hydrogen peroxide decontamination: an interrupted time series analysis.
      ], but AHP was used only in single-occupancy rooms while hydrogen peroxide was applied manually in shared rooms. The quality of the evidence from this study was, therefore, downgraded for indirectness; the consequence of the indirectness would be to dilute any real effect of AHP compared with manual cleaning/disinfection alone.
      In accordance with the overall quality of the evidence, the Working Party formulated weak/conditional recommendations for practice (that is, starting with the verb ‘consider’).

      Benefits and harms

      The NMA results suggest that using the different forms of automated decontamination (UV-C, PX-UV, HPV and AHP) in addition to manual cleaning/disinfection have some benefits compared with manual cleaning/disinfection alone.
      There was moderate evidence of benefit against C. difficile infection, with HPV and PX-UV having statistically significant effects; the effect of UV-C was also in the direction of benefit but was not statistically significant.
      There was moderate evidence of benefit against VRE infection or acquisition, with HPV having a statistically significant effect; the effects of UV-C and PX-UV were also in the direction of benefit although not statistically significant. HPV was associated with statistically significant reductions in the incidence of VRE infection or acquisition compared with UV-C and PX-UV. This might reflect the persistence of VRE in clinical environments and the ability of HPV to reach all surfaces whereas UV light might be subject to shadowing effects, etc.
      No evidence was identified in relation to the effectiveness of HPV when considering Acinetobacter spp. infection or acquisition. The evidence identified for Acinetobacter spp. was regarded as weak: the included studies were small and this resulted in effect estimates for UV-C and PX-UV being very imprecise and not statistically significant.
      There was weak evidence of benefit against MRSA infection or acquisition, with PX-UV having a statistically significant effect; the effects of UV-C, HPV and AHP were also in the direction of benefit but were not statistically significant.
      The Working Party's overall conclusion from the NMAs was that, where evidence was available, HPV was consistently most effective based on SUCRA scores; conversely, UV-C was generally the least effective of the automated approaches to decontamination, while still providing a marginal reduction in infection or acquisition of clinically relevant micro-organisms compared with manual cleaning/disinfection alone. The Working Party noted that the findings were consistent with in-vitro research demonstrating that ultraviolet light delivers a lower log10-kill rate than does hydrogen peroxide [
      • Weber D.J.
      • Rutala W.A.
      • Anderson D.J.
      • Chen L.F.
      • Sickbert-Bennett E.E.
      • Boyce J.M.
      Effectiveness of ultraviolet devices and hydrogen peroxide systems for terminal room decontamination: focus on clinical trials.
      ]. However, the bacterial load used in these studies far exceeds that likely to be encountered in the environment [
      • Yui S.
      • Ali S.
      • Muzslay M.
      • Jeanes A.
      • Wilson A.P.R.
      Identification of Clostridium difficile reservoirs in the patient environment and efficacy of aerial hydrogen peroxide decontamination.
      ].
      Most of the single-study analyses related to clinical outcomes with UV-C, PX-UV, or HPV, although one study evaluated the effectiveness of visible (indigo and white) light. The incidence of clinical events was reduced in some single-study reports, but the reductions were statistically significant in very few cases (for example, infection or acquisition due to ESBL-GNB, MDR-GNB and surgical site infection). A recent systematic review with 43 included articles found insufficient assessment of patient outcome because many were before–after studies and sponsored by industry; most were confounded by other infection control or audit interventions [
      • Dancer S.J.
      • King M.-F.
      Systematic review on use, cost and clinical efficacy of automated decontamination devices.
      ].
      The Working Party emphasized that the theoretical superiority of HPV reflected the increased effectiveness in killing spores demonstrated in laboratory studies [
      • Weber D.J.
      • Rutala W.A.
      • Anderson D.J.
      • Chen L.F.
      • Sickbert-Bennett E.E.
      • Boyce J.M.
      Effectiveness of ultraviolet devices and hydrogen peroxide systems for terminal room decontamination: focus on clinical trials.
      ], but that practical considerations might outweigh the theoretical advantages, for example, room turnaround times and training of operational staff. Use of lower concentrations of hydrogen peroxide might be attractive in practice, as would use of an alternative automated decontamination method allowing rapid re-entry of patient rooms. These considerations are explored further in Sections Procuring an automated room decontamination device and Using an automated room decontamination device.

      Cost-effectiveness and resource use

      The implementation of automated decontamination devices to enhance terminal cleaning of patient rooms and other clinical areas will have a cost impact. The exact costs will be dependent on the particular type of device.
      The cost consequences of healthcare-associated infection might be reduced using automated decontamination, particularly in elderly populations or clinical groups with weakened immune systems, and this might influence settings in which automated decontamination devices are recommended. Although the Working Party did not undertake a formal economic evaluation, the cost-effectiveness of automated decontamination is explored further in Section How cost-effective are automated room decontamination devices.

      Other considerations

      The Working Party highlighted the scarcity of evidence regarding the effectiveness of automated decontamination using AHP and visible (indigo and white) light. Although the literature searches were broad enough to identify clinical evidence related to high-intensity narrow-spectrum (HINS) light, steam, and ozone, no such evidence was identified.
      The evidence for Acinetobacter spp. displayed moderate heterogeneity, possibly due to the clinical outcomes of infection or acquisition being defined somewhat differently across the included studies. By contrast, the definitions of C. difficile, MRSA, and VRE infections or acquisitions were more consistent across studies (see Appendix H for further details).
      The evidence identified by the Working Party involved direct (head-to-head) comparisons only with manual cleaning/disinfection. Comparisons between different automated room decontamination devices were, therefore, made through indirect comparisons in the corresponding NMAs; the resulting effect estimates were subject to much uncertainty (as reflected in the GRADE imprecision domain). The study designs encountered in the evidence might reflect difficulties in implementing different automated decontamination systems in the same setting.

      Recommendations

      Consider use of an automated decontamination device as a supplement to manual cleaning in the context of rising or high prevalence of nosocomial infection, such as C. difficile, MRSA, or VRE.
      Consider use of hydrogen peroxide vapour or pulsed-xenon ultraviolet light in room surface decontamination during an outbreak of C. difficile infection when other modalities have failed to reduce acquisition.

      Procuring an automated room decontamination device

      How cost effective are automated room decontamination devices?

      The decision on which equipment to purchase or rent will depend on user needs, budget and service contract provisions. Contractors commonly have a preferred system, but the choice needs to be reviewed for effectiveness by the client. Cost is likely to be a key influencer, as well as the profile and reputation of the manufacturer. Some devices cost in excess of £50,000 to purchase outright from larger suppliers. Service contracts are an added cost but might often be attractive as the supplier company will risk assess use, train staff and include the cost of consumables. There is insufficient published information to recommend specific decontamination systems on the basis of cost savings due to reduced infections [
      • Dancer S.J.
      • King M.-F.
      Systematic review on use, cost and clinical efficacy of automated decontamination devices.
      ].

      What are the considerations for the relative benefits of hiring versus buying a device?

      The relative cost of purchase compared with lease should be investigated with prospective contractors. Options might include outright purchase, long-term hire or loan, rental or rental period with the option to purchase at the end. There may be benefits in terms of mitigating breakdown costs between the different purchase options. Differences between the different service options and related response times to breakdowns or repairs should be ascertained. The contractor should offer potential upgrades to the system as new developments occur but there may be additional costs best managed with a rental arrangement. However, in a hire or lease arrangement, there need to be adequate safeguards in place if the company goes into liquidation. The expected life-span of the equipment is an important consideration in the case of purchase.

      Does the system come with guarantees, service agreements and recommendations on whether it requires recalibration, and, if so, the frequency of this?

      In the procurement process, the lengths of time that standard guarantees can be extended should be determined with the associated costs. The breadth of the guarantee (parts, labour, manufacture fault and operator fault) should be ascertained. If there any disputes, the resolution process and the cover – for example, misuse by operators – is important. The published response time for callouts and supportive data on average response times is essential information during an outbreak or equipment failure. The requirements for the client when calling out an engineer must be clearly understood, particularly whether an order number is required to cover those items that might not be covered by a guarantee or warranty. The provisions for the replacement of faulty equipment must be agreed. The client will need to confirm the recommended period or number of uses after which the system should be recalibrated or checked by the provider and the associated cost.

      What key information should be requested from the manufacturer?

      Medical engineering, infection control, facilities management and procurement services should be consulted in all decisions regarding adoption of new devices as most information will be gathered by those departments as part of due diligence in the tendering process. All responsible suppliers/manufacturers should be able to provide the following information to the end user for any purchased system.
      • Written instructions on the correct use of the system, including safety-related information on the storage and handling of any hazardous chemicals required for filling the machine.
      • A full demonstration of the safe operation of the machine. A certificate for those receiving training should be provided as a record.
      • Electrical safety information.
      • Validation information relating to any known levels of efficacy of the system against defined groups of micro-organisms. National standards will be available shortly (see https://standardsdevelopment.bsigroup.com/projects/2019-00332#/section).
      • For the device, ease of use, quality control and functionality and reliability.
      In the case of fumigants, the supplier should provide a monitoring device for end users to check residual levels of fumigant in treated areas prior to full room re-entry. Such monitors should be calibrated against the fumigant in use and should be accompanied by a calibration certificate. Premature entry by cleaning staff may result in adverse effects.
      Regarding device maintenance and product life, the following should be established:
      • whether maintenance and repair are undertaken by the supplier or a third party
      • data showing the frequency of breakdown (on an annual basis) following purchase of equipment and itemized into the types of faults with the system
      • availability and source of spare parts, and the timeframe for accessing spare parts
      • whether the provider has systems in place for decontamination of any equipment or materials used in the maintenance of equipment during visits to other organizations
      • the shelf life of any consumables used with the system and the length of time after manufacturing consumables are stored prior to shipping.
      The proposed method of monitoring effectiveness should enable comparison between each treatment and should automatically record treatments carried out. The following information should be provided:
      • the means of recording within the system and how data are transferred and to whom
      • the security measures in place to safeguard any data
      • the ease of use, required education of users, supply of training (users vary but may be dedicated staff from the supplier or client staff who have been trained in the use of the equipment)
      • an operational manual for ongoing reference
      • safety measures such as lock or tamper-proofing, physical or software setting of the system, adjustment of the programme for room size, recording of every operation and setting, means of monitoring any leakage from the room or to monitor the system while in use
      • accessories supplied or recommended such as air vent covers, door seals, tape, hand-held hydrogen peroxide vapour gas monitors.
      Room-sealing devices are essential for both fumigant and ultraviolet-light systems, with signage to warn patients and staff that an area or room is being treated. Monitoring devices to determine the level of active agent and any leakage are desirable. Accessories should be provided to allow the system to be transported easily.

      What are the limitations of automated room decontamination devices?

      Automated room decontamination can assist IPC activities within the hospital setting by adding assurance to existing cleaning/disinfection procedures. They can treat areas that might not otherwise receive regular cleaning attention, for example, the treatment of high-level surfaces and, using fumigants, undersides of furnishings [
      • Mitchell B.G.
      • Digney W.
      • Locket P.
      • Dancer S.J.
      Controlling methicillin-resistant Staphylococcus aureus (MRSA) in a hospital and the role of hydrogen peroxide decontamination: an interrupted time series analysis.
      ,
      • Dancer S.J.
      Controlling hospital-acquired infection: focus on the role of the environment and new technologies for decontamination.
      ].
      In addition, when used at higher intensity, some of these systems might offer effective interventions during outbreaks, allowing hospital areas to be treated and returned to normal operational conditions more rapidly than might otherwise be possible. There is greater confidence in the standards of surface disinfection achieved [
      • McCord J.
      • Prewitt M.
      • Dyakova E.
      • Mookerjee S.
      • Otter J.A.
      Reduction in Clostridium difficile infection associated with the introduction of hydrogen peroxide vapour automated room disinfection.
      ].
      Automated room decontamination systems are not designed to be used in isolation or as a standalone solution to hard surface disinfection of hospital ward and clinical treatment areas. Surfaces such as floors, tables, wash areas or toilets still require physical cleaning. This basic requirement removes organic soiling, making surfaces appear acceptably clean and facilitates a more effective fumigation process or permits better access of ultraviolet light. Biocidal processes using chemicals or ultraviolet light are more effective when associated organic soiling is removed or reduced. Due to toxicity, both hydrogen peroxide and ultraviolet systems are used in single patient rooms and not in shared ward bays. Although temporary single-bed enclosures have been tested, none has been widely adopted.
      For systems using a fumigant, performance is affected by [
      • Dancer S.J.
      Controlling hospital-acquired infection: focus on the role of the environment and new technologies for decontamination.
      ]:
      • the size and sealability of the treated area – leaks may result in failure to maintain effective levels of fumigant in the decontamination process and threaten the safety of those potentially exposed in surrounding areas
      • the type and levels of fumigant delivered into the treated area
      • the physical nature of the gas, vapour, or aerosolization – for example, small droplets will only reach shadowed areas if the particles generated are small enough to be buoyant and to mix well in the treated room air; a truly gaseous product will not suffer such problems
      • the nature of challenge micro-organisms (and concentration) that require eradication; for example, C. difficile spores are likely to be more difficult to kill than P. aeruginosa vegetative cells; however, some vegetative micro-organisms might be more resistant than spores due to physiological characteristics such as catalase production
      • the presence of cold surfaces in rooms, such as outside windows and doors, might increase localized condensation of vapourized product, limiting deposition (micro-condensation) of fumigant on to other surfaces in the room
      • some equipment in a treated area might be sensitive to the effects of raised chemical and moisture levels, limiting the areas of use – for example, avoiding microelectronics, mild steel, and porous surfaces such as wood and textiles
      • presence of organic matter can reduce penetration of the fumigant or break it down.
      Similarly, ultraviolet light systems have limitations, which include [
      • Dancer S.J.
      Controlling hospital-acquired infection: focus on the role of the environment and new technologies for decontamination.
      ]:
      • shadowing effects produced by surface soiling or furniture and room geometry
      • reduced biocidal effect due to distance of the contaminant from the ultraviolet light source
      • type and energy level of ultraviolet light delivered
      • position and number of units used to treat the area (that is, triangulation and overlap effects of three systems enhance effectiveness)
      • age of the units, which will affect the energy level of the ultraviolet light emission
      • presence/absence of low-level light units which can influence floor-level energy delivery – a luminometer can be used for quality purposes as an independent measure
      • penetration of glass might be uncertain.

      Using an automated room decontamination device

      What standard of manual cleaning/disinfection should be evident before using an automated device?

      The use of automated decontamination equipment does not negate the need for physical cleaning of healthcare surfaces. When used alongside traditional (manual) cleaning/disinfection methods the two represent distinct but recognized hygiene approaches that seek to achieve the same endpoint – a safe, clean hospital environment. Both are generally used to deliver effective IPC [
      • Dancer S.J.
      Controlling hospital-acquired infection: focus on the role of the environment and new technologies for decontamination.
      ].
      The purpose of environmental cleaning is to remove debris and organic matter. For a fully effective physical process, a properly prepared environment, trained and competent staff, ergonomically designed equipment, and chemical contact are all required. The time taken to clean a patient room between successive occupations depends on the space, en-suite facilities, ventilator grills, radiators, and the general condition of the area. Turnaround can be reduced by effective collaboration between clinical and domestic staff. A standard operating procedure detailing staff responsibilities will improve efficiency and set expectations of management.
      A visual inspection undertaken jointly by clinical and domestic supervisors who are competent at inspections will ensure the standard is achieved consistently. Routine ward-based cleaning of solid surfaces, such as bed frames, bedside cabinets, over-bed tables and bathroom areas, is typically undertaken using approaches specified in national guidance. For example, in the UK, professionals should refer to National Standards of Healthcare Cleanliness 2021 [
      NHS England and NHS Improvement
      National standards of healthcare cleanliness 2021.
      ] and the standard for providing a clean and safe hospital environment [
      BSI Standards Limited
      Specification for the planning, application, measurement and review of cleanliness services in hospitals.
      ]. An initial physical cleaning step, using detergent and warm water, is followed in high-risk areas by wiping over the same areas with a disinfectant, either using disposable wipes or a freshly prepared solution of a chlorinated disinfectant.
      The standard achieved by manual cleaning can be assessed by a rapid test for adenosine triphosphate (ATP), which is present in all organic material but not a direct measure of bacterial, viral, or sporal loading on a surface. Presence of ATP can be linked to improvements in training in cleaning processes and improving the cleanliness of surfaces.
      Automated decontamination systems, whether chemical or ultraviolet light-based, are likely to work less effectively if surfaces are visibly dirty [
      • Weber D.J.
      • Rutala W.A.
      • Anderson D.J.
      • Chen L.F.
      • Sickbert-Bennett E.E.
      • Boyce J.M.
      Effectiveness of ultraviolet devices and hydrogen peroxide systems for terminal room decontamination: focus on clinical trials.
      ,
      • Dancer S.J.
      Controlling hospital-acquired infection: focus on the role of the environment and new technologies for decontamination.
      ]. This is because, in the case of chemical fumigants, organic residues will react with micro-condensed disinfectant on material surfaces and might neutralize it, reducing its activity before it can impact on contaminating micro-organisms. Physical soiling can shield micro-organisms exposed to ultraviolet light, masking the contaminants from full exposure to the treatment and again reducing the beneficial impact. Treated surfaces should first be cleaned to maximize the enhanced hygiene offered by automated decontamination [
      • Dancer S.J.
      Controlling hospital-acquired infection: focus on the role of the environment and new technologies for decontamination.
      ].

      Good practice point

      • Manual cleaning should be completed to the same high standard regardless of the subsequent use of automated cleaning devices.

      Is there a potential risk of bacterial resistance to the decontamination method?

      Fumigant automated decontamination systems may deliver: a true gas such as ozone or chlorine dioxide; heat-generated vapour (for example, hydrogen peroxide or formaldehyde); or cold-generated fogs or dry mists from aqueous liquid disinfectants such as hydrogen peroxide, quaternary ammonium compounds and peracetic acid [
      • Beswick A.J.
      • Farrant J.
      • Makison C.
      • Gawn J.
      • Frost G.
      • Crook B.
      • et al.
      Comparison of multiple systems for laboratory whole room fumigation.
      ,
      • Otter J.A.
      • Mepham S.
      • Athan B.
      • Mack D.
      • Smith R.
      • Jacobs M.
      • et al.
      Terminal decontamination of the Royal Free London's high-level isolation unit after a case of Ebola virus disease using hydrogen peroxide vapor.
      ,
      • Pottage T.
      • Richardson C.
      • Parks S.
      • Walker J.T.
      • Bennett A.M.
      Evaluation of hydrogen peroxide gaseous disinfection systems to decontaminate viruses.
      ,
      • Pottage T.
      • Macken S.
      • Walker J.T.
      • Bennett A.M.
      Meticillin-resistant Staphylococcus aureus is more resistant to vaporized hydrogen peroxide than commercial Geobacillus stearothermophilus biological indicators.
      ]. The delivery system should ensure that the disinfectant chemicals reach the targeted area, with active agents typically designed to give a powerful oxidizing effect on pathogenic micro-organisms. This effect damages microbial cell structures, including cell membranes and internal cellular components such as nucleic acids. The levels of chemical delivery are high and even spores are not immune to most oxidizing treatments. Some protection might be conferred where micro-organisms are associated with high levels of organic soiling (for example, faeces or blood), but this should not occur if physical cleaning of surfaces has been undertaken. Penetration of any liquid contaminant is challenging for most fumigation systems. In addition, complex room structures/furnishings and tubular (hollow) objects limit fumigant ingress and result in reduced effectiveness.
      Ultraviolet-based systems rely on energy delivery to surfaces, rather than chemical action, to disrupt target cells that lie within the line of sight of the treatment unit(s) [
      • Weber D.J.
      • Rutala W.A.
      • Anderson D.J.
      • Chen L.F.
      • Sickbert-Bennett E.E.
      • Boyce J.M.
      Effectiveness of ultraviolet devices and hydrogen peroxide systems for terminal room decontamination: focus on clinical trials.
      ]. All ultraviolet-based biocidal treatments are therefore limited by the amount of shadowing of surfaces intended for treatment. The evidence for successfully treating surfaces not in line of sight or for using methods of ultraviolet reflection under or around static objects remains mixed. As with fumigation delivery, any organic co-contaminants that confer physical protection to the micro-organisms will shield them from ultraviolet exposure, thus reducing the effectiveness of the treatment. The positioning of the emitters accounts for most differences in bactericidal efficacy between systems [
      • Ali S.
      • Yui S.
      • Muzslay M.
      • Wilson A.P.R.
      Comparison of two whole-room ultraviolet irradiation systems for enhanced disinfection of contaminated hospital patient rooms.
      ]. Physical cleaning therefore remains an important prerequisite to environmental decontamination involving ultraviolet-light treatment delivery.

      What are the health and safety considerations of using automated room decontamination devices?

      Automated room decontamination devices comprise fumigation systems and devices used to deliver biocidal ultraviolet light (UV-C and PX-UV). Most published research focuses on treatment efficacy, rather than safety. However, there are health and safety issues that require consideration prior to deploying the equipment.
      The chemicals delivered by fumigation systems are harmful if exposure occurs during use. Products may be based on hydrogen peroxide, ozone, or chlorine dioxide. Hydrogen peroxide can be present as the only active substance or in combination with other components such as silver or peracetic acid. Systems delivering this chemical may achieve hydrogen peroxide concentrations of several hundred parts per million (ppm) in air, whereas the workplace exposure limit (WEL) for this chemical in the UK is just 1 ppm for long-term exposure and 2 ppm for short-term exposure [
      • Ali S.
      • Yui S.
      • Muzslay M.
      • Wilson A.P.R.
      Response to letter of K. Singh, ‘Role of silver nitrate in the efficacy of hydrogen peroxide aerial decontamination systems’ regarding S. Ali et al. ‘Efficacy of two hydrogen peroxide vapour aerial decontamination systems for enhanced disinfection of meticillin-resistant Staphylococcus aureus, Klebsiella pneumoniae and Clostridium difficile in single isolation rooms.
      ,]. Similarly, other widely used fumigants, such as ozone, have a low WEL (ozone = 0.2 ppm). The potential for harmful chemical exposure is, therefore, clear and should be controlled. The whole process should be risk assessed, with procedures in place to ensure the room is sealed to retain the fumigant, preventing room entry during treatment, monitoring for fumigant leaks during and after treatment, and effective fumigant removal and room aeration to complete the process. There is no justification for unnecessarily exposing staff or patients to harmful chemicals during fumigation treatments.
      The availability of the correct gas monitoring equipment is crucial to ensuring appropriate measurements can be made during or after the treatment process [
      • Murdoch L.E.
      • Bailey L.
      • Banham E.
      • Watson F.
      • Adams N.M.T.
      • Chewins J.
      Evaluating different concentrations of hydrogen peroxide in an automated room disinfection system.
      ]. These items typically cost a fraction of the price of the fumigation equipment and their use is central to safe working with any fumigant [
      • Ali S.
      • Yui S.
      • Muzslay M.
      • Wilson A.P.R.
      Response to letter of K. Singh, ‘Role of silver nitrate in the efficacy of hydrogen peroxide aerial decontamination systems’ regarding S. Ali et al. ‘Efficacy of two hydrogen peroxide vapour aerial decontamination systems for enhanced disinfection of meticillin-resistant Staphylococcus aureus, Klebsiella pneumoniae and Clostridium difficile in single isolation rooms.
      ].
      As with other forms of high-intensity light, biocidal ultraviolet light is potentially harmful and can damage the eyes and skin if they are exposed [
      • Anderson D.J.
      • Knelson L.P.
      • Lewis S.S.
      • Chen L.F.
      • Sexton D.J.
      • Weber D.J.
      • et al.
      Implementation lessons learned from the Benefits of Enhanced Terminal Room (BETR) Disinfection Study: process and perceptions of enhanced disinfection with ultraviolet disinfection devices.
      ,
      • Masse V.
      • Hartley M.J.
      • Edmond M.B.
      • Diekema D.J.
      Comparing and optimizing ultraviolet germicidal irradiation systems use for patient room terminal disinfection: an exploratory study using radiometry and commercial test cards.
      ]. Ultraviolet-light carousels vary in terms of their energy delivery, but all are designed to cause cellular damage to micro-organisms and should, therefore, be used with control measures in place to prevent human exposure. As with fumigation systems, it is important to avoid entry of the area during treatment. Modern carousels may be fitted with motion sensors that immediately turn off the system if any motion is detected in the room. In addition, ultraviolet light does not usually penetrate double-glazed windows, although any assumption that glass is protecting an observer outside the room should be checked with a light meter capable of measuring ultraviolet emissions between 100 and 280 nm. There have been anecdotal reports of some ultraviolet-light systems generating localized ozone around high-energy lamps, but this effect would need to be investigated if suspected. Ozone could, if present even at low residual levels after treatment, potentially cause respiratory irritation for those exposed.

      Good practice point

      • On first use of a fumigant or ultraviolet light in a specific room design, efficacy of sealing should be monitored to ensure safety.

      How easy is the equipment to use, what standard of education are users expected to have, and what training and training materials are supplied?

      The operation of earlier systems could be complex and sometimes required open handling of potentially harmful chemicals. However, in modern machine designs, disinfectants are often supplied in sealed or smart cartridges, or decanting of chemicals is minimized with use of protective gloves and eye protection [
      Health and Safety Executive
      Manual cleaning and disinfecting surfaces: harm via skin or eye contact, SR4, COSHH essentials for service and retail.
      ].
      The operational interfaces on both fumigation and ultraviolet-light systems may range between ‘on–off’ switch activation with a timer delay to allow staff to safely leave the room, to more complex touchscreen interfaces with multiple programmes. Regardless of the method, all systems should be provided with effective training from the supplier.

      What indicators should trigger use of an automated device?

      IPC teams in collaboration with domestic services will determine the appropriate use of hydrogen peroxide and ultraviolet light systems in addition to standard manual cleaning. The indication to use hydrogen peroxide, ultraviolet light or other systems in addition to manual cleaning will depend on the terms of the contract agreed with the suppliers, cost and availability of staff, as well as a risk assessment of the pathogenicity of the organisms that caused infection in the last occupant of the room. Prevalence of, or outbreaks due to, certain pathogens may be deemed higher priority for additional room disinfection by local IPC teams. Automated devices are not usually practicable in shared bays, unless the area is free of patients and staff and can be sealed. Where available, hydrogen peroxide cleaning may be used following discharge of patients with C. difficile, norovirus, multi-resistant organisms, such as acinetobacter, carbapenem-resistant Enterobacterales and tuberculosis, and viral haemorrhagic fever [
      • Passaretti C.L.
      • Otter J.A.
      • Reich N.G.
      • Myers J.
      • Shepard J.
      • Ross T.
      • et al.
      An evaluation of environmental decontamination with hydrogen peroxide vapor for reducing the risk of patient acquisition of multidrug-resistant organisms.
      ,
      • Yui S.
      • Ali S.
      • Muzslay M.
      • Jeanes A.
      • Wilson A.P.R.
      Identification of Clostridium difficile reservoirs in the patient environment and efficacy of aerial hydrogen peroxide decontamination.
      ]. Turnaround times are usually between 3 and 4 h depending on local circumstances. An ultraviolet-light cleaning system may be used following occupation by a patient with norovirus or rotavirus, COVID-19, MRSA, Streptococcus group A, extended-spectrum β-lactamase producers, VRE and during outbreaks of infection not successfully managed by increased manual cleaning (or any of the above if hydrogen peroxide is not available) [
      • Anderson D.J.
      • Chen L.F.
      • Weber D.J.
      • Moehring R.W.
      • Lewis S.S.
      • Triplett P.F.
      • et al.
      Enhanced terminal room disinfection and acquisition and infection caused by multidrug-resistant organisms and Clostridium difficile (the Benefits of Enhanced Terminal Room Disinfection study): a cluster-randomised, multicentre, crossover study.
      ,
      • Anderson D.J.
      • Moehring R.W.
      • Weber D.J.
      • Lewis S.S.
      • Chen L.F.
      • Schwab J.C.
      • et al.
      Effectiveness of targeted enhanced terminal room disinfection on hospital-wide acquisition and infection with multidrug-resistant organisms and Clostridium difficile: a secondary analysis of a multicentre cluster randomised controlled trial with crossover design (BETR Disinfection).
      ,
      • Ali S.
      • Yui S.
      • Muzslay M.
      • Wilson A.P.R.
      Comparison of two whole-room ultraviolet irradiation systems for enhanced disinfection of contaminated hospital patient rooms.
      ]. Turnaround times are shorter (1.5–2 h) but depend on local circumstances. In the absence of either fumigant or ultraviolet-light systems, a second manual clean can reduce environmental contamination and transient flora acquired on the hands of staff to a similar degree [
      • Wilson A.P.R.
      • Smyth D.
      • Moore G.
      • Singleton J.
      • Jackson R.
      • Gant V.
      • et al.
      The impact of enhanced cleaning within the intensive care unit on contamination of the near-patient environment with hospital pathogens: a randomized crossover study in critical care units in two hospitals.
      ].

      Good practice point

      • Prioritize different cleaning systems to the type of infection of the most recent room occupant by use of a red/amber/green rating based on local nosocomial infection rates.

      What are the requirements in terms of engineer audit and user audit?

      Prior to any work commencing, it is important to audit the information provided with the device, which should include the engineering and calibration protocols/results that the generator has undergone, with any corrective measures.
      Validation data should be generated to demonstrate that an effective cycle can be completed in each enclosure prior to the generator's use for decontamination. Biological indicators should be positioned around the enclosure and an effective fumigation process determined by the inactivation of the indicators. For validation purposes, the enclosure should be set up to replicate actual use, allowing optimization of cycle conditions before use in the target area. The room design and set-up should be as similar as possible to that intended for use, or a room in the target area temporarily set aside for the test. Usually a 3–5 log10 reduction in bacterial numbers is required in the healthcare environment [
      • Dancer S.J.
      Controlling hospital-acquired infection: focus on the role of the environment and new technologies for decontamination.
      ]. As the devices provide disinfection – not sterilization – survival of some micro-organisms may be acceptable, provided any residual levels can be tolerated.
      Data generated from the automated decontamination device can be used to document an effective cycle, preferably every time, together with surface cultures (before and after) or biological indicators if available (see Section How often should testing be performed?) [
      • Ali S.
      • Muzslay M.
      • Bruce M.
      • Jeanes A.
      • Moore G.
      • Wilson A.P.R.
      Efficacy of two hydrogen peroxide vapour aerial decontamination systems for enhanced disinfection of meticillin-resistant Staphylococcus aureus, Klebsiella pneumoniae and Clostridium difficile in single isolation rooms.
      ,
      • Ali S.
      • Yui S.
      • Muzslay M.
      • Wilson A.P.R.
      Comparison of two whole-room ultraviolet irradiation systems for enhanced disinfection of contaminated hospital patient rooms.
      ,
      • Beal A.
      • Mahida N.
      • Staniforth K.
      • Vaughan N.
      • Clarke M.
      • Boswell T.
      First UK trial of Xenex PX-UV, an automated ultraviolet room decontamination device in a clinical haematology and bone marrow transplantation unit.
      ]. Some devices provide a printout of the different parameters used during a cycle as a record to form part of audit to ensure that the decontamination process is effective and repeatable. Other devices provide real-time feedback on a display screen that can be checked against records to ensure that the correct parameters have been achieved. For devices that need to be inside the enclosure and do not have a visible display during operation, it is important to ensure that there is a method of accurately determining that a cycle has been completed successfully – for example, a download from the device to a computer to check that the cycle has been completed. All cycle-monitoring data should be adequately organized (by reporting date, location, cycle type, etc.) to allow processes to be audited. Some providers have maintenance audit and calibration checks within their contract with the client. Fumigant chemicals should be used within date and some smart systems may not accept out-of-date cartridges.

      Can extra equipment be placed in the room during decontamination?

      Any future standard fumigation or ultraviolet-light test is likely to avoid the placement of additional equipment or furnishings in the treated area, other than the fumigation machine and test coupons. In practice, other items such as furniture will be present and other equipment needing decontamination may be placed in the room [
      • Andersen B.M.
      • Rasch M.
      • Hochlin K.
      • Jensen F.H.
      • Wismar P.
      • Fredriksen J.E.
      Decontamination of rooms, medical equipment and ambulances using an aerosol of hydrogen peroxide disinfectant.
      ]. The number of items should be controlled to a reasonable level following the supplier's advice. However, some items are difficult to disinfect because of their shape – for example, convoluted or tubular – or, if too many items are moved into the room, then the surface area to treat may exceed the dose of fumigant supplied or in the case of UV-C result in excessive shadowing. For fumigant systems, porous materials will absorb the active agent and off-gas (see Section 8.3.8).

      How does effectiveness of decontamination compare for hard and soft surfaces?

      During automated chemical disinfection processes, porous materials should be removed from rooms for reasons related to effectiveness and safety. If common porous materials such as textiles, foams, and cardboard are left in a room they can absorb fumigant chemicals during treatment. These are released slowly afterwards, a process known as off-gassing. This phenomenon might influence fumigant performance on surfaces, with overall effectiveness becoming less certain than if all surfaces are non-porous. If all surfaces in a room are smooth and impervious, then this problem will be avoided because fumigant levels in the treated area will be more predictable [
      • Beswick A.J.
      • Farrant J.
      • Makison C.
      • Gawn J.
      • Frost G.
      • Crook B.
      • et al.
      Comparison of multiple systems for laboratory whole room fumigation.
      ]. Seemingly protected items such as foam-filled mattresses with waterproof covers might still absorb chemical fumigant via zip closure [
      • Poppendieck D.
      • Hubbard H.
      • Corsi R.L.
      Hydrogen peroxide vapor as an indoor disinfectant: removal to indoor materials and associated emissions of organic compounds.
      ]. Foam-filled items are best removed from the room and treated separately with wet surface disinfectant.
      Even when a fumigation system has a chemical removal step this might not remove all fumigant from porous materials. These can continue to off-gas chemicals beyond treatment completion and chemical levels might rise above WELs. For this reason, it is recommended that a portable sensor is used to check fumigant levels at the point of room re-entry, even when such levels are expected to be safe. In some situations, small ‘pockets’ of the fumigant may remain (for example, under solid surfaces, where aeration might not fully take place).
      In contrast to fumigant chemicals, ultraviolet light does not penetrate porous surfaces such as sheets, upholstery and curtains. In addition, any shadowing effect caused by a material's porosity, shape, and softness is likely to inhibit the exposure of contaminants to the full ultraviolet dose [
      • Nerandzic M.M.
      • Cadnum J.L.
      • Pultz M.J.
      • Donskey C.J.
      Evaluation of an automated ultraviolet radiation device for decontamination of Clostridium difficile and other healthcare-associated pathogens in hospital rooms.
      ,
      • Nottingham M.
      • Peterson G.
      • Doern C.
      • Doll M.
      • Masroor N.
      • Sanogo K.
      • et al.
      Ultraviolet-C light as a means of disinfecting anesthesia workstations.
      ]. This will in turn cause uncertainty in machine performance and would require site-specific validation to confirm that required microbiological kill is being achieved. There is no reason why softer or more porous materials cannot be left in a room being treated with ultraviolet light, but the success of the treatment would be dependent on the amount of light energy hitting exposed surfaces; harder, more even surfaces are always easier to treat with light-based technology.

      Good practice point

      • Remove foam materials from the room if fumigant is used unless sealed under an impervious cover.

      How many times can the device be used in a room?

      There is no imposed or recognized limit on how many times a device can be used in a particular room or treatment space. Multiple treatments would be expected, depending on the nature of the room's use. The room and its resident equipment should be able to tolerate the intended treatment, preferably be free from porous or absorbent materials, and any ‘leaky’ areas of the room should be sealed if fumigant chemicals are to be used. Losses might not only reduce the effectiveness of the treatment but might also allow the seepage of fumigant into sensitive areas outside the intended treatment location. Such considerations should be built into an appropriate risk assessment for the treatment of any room, healthcare or otherwise.
      There are reports of certain types of fumigant chemicals damaging surfaces after only small numbers of treatments, and the powerful oxidizing effect of the chemicals is the most likely cause [
      • Kimura T.
      • Yahata H.
      • Uchiyama Y.
      Examination of material compatibilities with ionized and vaporized hydrogen peroxide decontamination.
      ]. HPV and chlorine dioxide gas have been implicated especially because of their corrosive properties, especially when unwanted condensate pools on surfaces or inside sensitive equipment [
      Reliability of electronic equipment exposed to chlorine dioxide used for biological decontamination.
      ,
      • Girouard D.J.
      • Czarneski M.A.
      Room, suite scale, class III biological safety cabinet, and sensitive equipment decontamination and validation using gaseous chlorine dioxide.
      ,
      • Hall L.
      • Otter J.A.
      • Chewins J.
      • Wengenack N.L.
      Use of hydrogen peroxide vapor for deactivation of Mycobacterium tuberculosis in a biological safety cabinet and a room.
      ]. Consideration should also be given to seals around doors, utilities such as pipes and cables passing through walls, etc., where sealant or bespoke seals around these items will be required to contain fumigant chemicals. Such seals may deteriorate over multiple treatments, requiring examination and testing to ensure that no leaks have appeared in the fabric of the room. For non-specialized areas that are treated repeatedly, the risk is greater than in laboratories where the fabric of the room has been designed to tolerate such treatments.
      Similar principles can be applied to ultraviolet-light treatments, in that multiple treatments can be delivered if required, but it should be confirmed that materials in a room can tolerate the energy delivery over time. Ultraviolet light would not be expected to have major cumulative effects on room infrastructure and integrity (wall and ceiling materials, seals around doors, windows, etc.), although some polymers (including some plastics) might be affected after prolonged exposure.

      Does decontamination degrade room equipment?

      There have been reports of equipment and surfaces being damaged by repeated automated disinfection treatments. Electronic equipment, stainless steel, powder-coated paint, anodized metals and enamelled surfaces have reportedly suffered damage. Metal corrosion, surface tarnishing, material colour ‘bleaching’ and enamel loss have all been described, with hydrogen peroxide and chlorine dioxide treatments implicated most frequently.
      These effects have been an end-user concern since fumigation treatments became more commercially focused in the 1990s, and system suppliers have sought to demonstrate safety and reproducibility, publishing a number of articles to support their use with sensitive materials [
      • Girouard D.J.
      • Czarneski M.A.
      Room, suite scale, class III biological safety cabinet, and sensitive equipment decontamination and validation using gaseous chlorine dioxide.
      ]. The potential for bias in such articles should, however, be considered, especially in terms of funding from or association with device manufacturers. Most chemical fumigation systems operate at relative humidity levels in excess of 65%, and this level of moisture alone might be incompatible with some types of complex electrical equipment [
      • Krause J.
      • McDonnell G.
      • Riedesel H.
      Biodecontamination of animal rooms and heat-sensitive equipment with vaporized hydrogen peroxide.
      ]. Where high humidity meets cold surfaces, such as external walls and windows in a treated room, condensation can pool in a form that contains the concentrated active chemical [
      • Beswick A.J.
      • Farrant J.
      • Makison C.
      • Gawn J.
      • Frost G.
      • Crook B.
      • et al.
      Comparison of multiple systems for laboratory whole room fumigation.
      ]. This might in turn cause damage to paint work and some metals, particularly after repeated treatments [
      • Pottage T.
      • Lewis S.
      • Lansley A.
      • Fraser S.
      • Hendon-Dunn C.
      • Bacon J.
      • et al.
      Hazard Group 3 agent decontamination using hydrogen peroxide vapour in a class III microbiological safety cabinet.
      ].
      For biocidal ultraviolet light, the rate and extent of any material degradation is also likely to relate to the levels of ultraviolet energy delivered, usually measured in Joules (J) or mJ/cm2 of surface treated. Prolonged exposure has been associated with damage in endoscope storage facilities [
      • Irving D.
      • Lamprou D.A.
      • Maclean M.
      • MacGregor S.J.
      • Anderson J.G.
      • Grant M.H.
      A comparison study of the degradative effects and safety implications of UVC and 405 nm germicidal light sources for endoscope storage.
      ].
      The compatibility of any surface materials or equipment should be discussed with the equipment suppliers prior to procurement or embarking on room treatments.

      Are there any limitations on use of the active agent on some materials (including compatibility with equipment used in intensive care units)?

      In areas where microelectronic circuitry is present, such as critical care equipment, there is potential for damage to metals and delicate circuit boards by the fumigant [
      • Girouard D.J.
      • Czarneski M.A.
      Room, suite scale, class III biological safety cabinet, and sensitive equipment decontamination and validation using gaseous chlorine dioxide.
      ,
      • Krause J.
      • McDonnell G.
      • Riedesel H.
      Biodecontamination of animal rooms and heat-sensitive equipment with vaporized hydrogen peroxide.
      ]. Proving that this will not occur requires validation and prior testing on similar equipment and may not be feasible. The integrity of plastic and rubber used in seals on or within equipment might be susceptible to degradation because of chemical exposure. If there is any doubt about the impact of fumigant chemicals on sensitive and important equipment, the automated device supplier should be consulted before any treatment is undertaken.
      High-intensity ultraviolet light is less likely to damage internal electronic componentry but might affect surface finishes such as whitened PVC [
      • Irving D.
      • Lamprou D.A.
      • Maclean M.
      • MacGregor S.J.
      • Anderson J.G.
      • Grant M.H.
      A comparison study of the degradative effects and safety implications of UVC and 405 nm germicidal light sources for endoscope storage.
      ,
      • Kauffman R.
      • Wolf J.
      Study of the degradation of typical HVAC materials, filters, and components irradiated by UVC energy – Part II: Polymers (RP-1509).
      ]. The flexibility of some softer materials such as sealants might also be affected, resulting in shrinkage and cracking. The extent of this damage will depend on the material composition, the intensity of ultraviolet energy delivery and the frequency of treatments. If there is any doubt, the automated device supplier should be consulted before any treatment is undertaken.

      Good practice point

      • After purchasing an ultraviolet-light decontamination system, consider the impact on surface finishes such as whitened PVC before purchasing other equipment, and ask the equipment supplier to confirm compatibility.

      Do damaged surfaces need to be sealed before use of the device?

      Normally impervious surfaces that already show signs of cracks, fissures, and flaking may be further damaged by fumigation. Surface imperfections can allow colonization by micro-organisms that avoid exposure to the active agent. Removal of damaged items from the area to be treated is advisable. Trying to seal off surfaces with plastic or tape will trap fumigant or the contamination.
      For ultraviolet-light treatment, sealing of surfaces is not necessary and the problem relates more to the ability of light energy to penetrate a breach or depression in an otherwise smooth surface. Unevenness due to pitting and flaking is likely to cause small shadowing effects, which means that full ultraviolet energy delivery to the damaged region cannot be assumed.

      Can a device be used in rooms with positive or negative pressure ventilation?

      Under normal circumstances a ventilation system that maintains a positive or negative air pressure in a clinical workspace would be switched off during fumigation treatment. Most ventilation control is based on total air loss, that is, feeding air in and out of the room space at a pre-set flow rate to refresh and mix the air and achieve the required air flow and pressure conditions. The movement of air can rapidly dilute fumigant in the room and would reduce treatment effectiveness. In addition, a room maintained at positive pressure would be at higher risk of leaking toxic fumigant chemicals. Keeping the ventilation running and simply blocking off the vents could adversely affect air pressurization in other critical areas such as isolation rooms [
      • Byrns G.
      • Fuller T.P.
      The risks and benefits of chemical fumigation in the health care environment.
      ,
      • Doll M.
      • Morgan D.J.
      • Anderson D.
      • Bearman G.
      Touchless technologies for decontamination in the hospital: a review of hydrogen peroxide and UV devices.
      ]. For these reasons, and to control the fumigation process, the room ventilation should be turned off during these procedures, unless a room air recirculation function is available.
      For ultraviolet-light systems any air flow or air pressure changes should not undermine the performance of the light-based treatment, although if expected air pressure changes are well above or below ambient then it would be prudent to check with the equipment supplier to ensure no damage to the system electronics is likely.

      What measurements are required when calculating the dose of active chemical for the room size? Is this carried out by the supplier/manufacturer or can the machine be programmed by the user?

      Many modern automated fumigation systems have a user-control panel (touchscreen) or remote PC/tablet operation that allows the machine to be programmed with the required treatment conditions. The machine control software might already be pre-programmed with a variety of cycle conditions. Although different manufacturers take different approaches to machine set-up, the volume of the room to be treated is often the basis for calculating fumigant chemical delivery, with the machine automatically calculating the required dosage once these dimensions have been entered into the software. For some simpler and less expensive machines, where software programming might be absent, an ‘on–off’ approach is often used, with a set volume of liquid disinfectant added to the machine prior to treatment. The amount of disinfectant would typically be informed by the supplier, a user manual, or both. In the event of a purely gaseous product being used, such as ozone generated from ambient air, there is no requirement for liquid disinfectant calculations and the duration of delivery is again likely to be calculated from treated area volume. Before using any fumigation system, detailed advice should be sought from the supplier regarding the set-up and use of the system. This is likely to vary between different rooms, especially if room geometry varies. The cycle should be validated when used for the first time using representative micro-organisms if possible.
      The amount of energy delivered by different ultraviolet-light devices is likely to vary and will be dependent on the energy output from ultraviolet-light units, the number of units present on the device, the number of devices used to treat a room and the distance of the ultraviolet-light emission system from the target surface. Detailed advice should be sought from the suppliers regarding the positioning and duration of use for these systems, which is likely to vary from room to room.

      What is the cycle time?

      The term ‘cycle time’ is often used to describe the time required for an automated decontamination device to go from start to finish of its treatment, at the point where staff and patients can safely re-enter the treated area.
      For chemical systems this normally includes clearing and sealing of the room, followed by placement of the system in an agreed position and:
      • a room air-conditioning step – where the room air might be treated in some way before delivery of the disinfectant (for example, reduction of humidity); this step is often absent for smaller devices
      • the fumigant delivery step – where the fumigant is sprayed or pumped into the room: some systems will inject until a set point is reached; others will inject a certain volume of the fumigant
      • the dwell or exposure time – where the fumigant chemical remains in the air, or is deposited on to surfaces, often mixed by the delivery system fan or some other means; some advanced systems will constantly ‘top up’ the fumigant levels during this period, to maximize treatment effectiveness
      • the removal or aeration step – this might be an active chemical removal phase completed by the same machine that delivers the chemical or it might involve mechanical ventilation being reactivated to dilute and remove the chemical and promote its breakdown; indeed it might involve a suitably protected operator entering the treated area and opening external windows to facilitate natural aeration of the room (this option requires the use of appropriate personal protective equipment, namely respirator, skin and eye protection, and should be avoided if possible or used only as a last resort or emergency procedure).
      Cycle times might vary from ∼1 h to several hours, or even overnight, depending on the circumstances and the type of chemical in use. The aeration period may need to be extended due to incomplete removal of the fumigant from the room by the automated device in the time allowed, therefore it is necessary to monitor the level prior to entry with a calibrated handheld monitor.
      For ultraviolet-light systems the cycle steps are far simpler and generally shorter. There are normally no chemical residues generated and so no requirement for a ‘removal’ step. Typical treatment steps would be:
      • placement of the ultraviolet-light unit(s) in the room and covering of windows/closure of doors to prevent any risk of human exposure for those outside the treated area
      • safe activation of the system once the room is clear of staff and patients; this is normally achieved using a timer delay of 1 min or more, to allow machine activation and then time for room clearing
      • the delivery step – this typically lasts 15–30 min, but might involve stopping the system thereafter, relocating it and then repeating the treatment, to make sure all areas receive equal coverage
      • room re-entry once the treatment is complete, removal of machines and re-occupation; most modern ultraviolet-light systems have motion sensors attached as a safety feature, such that the machine will shut down if any movement is detected in the room during treatment.

      For a single-occupancy patient room of typical size, what is the time requirement for the process from completing manual cleaning/disinfection to being able to enter the room to set up for the next patient?

      Following physical cleaning of a room, the duration of the fumigation process might vary considerably between systems and is influenced by the type of room [
      • Beswick A.J.
      • Farrant J.
      • Makison C.
      • Gawn J.
      • Frost G.
      • Crook B.
      • et al.
      Comparison of multiple systems for laboratory whole room fumigation.
      ,