Cohorting Inpatients with Omicron and Delta Variants of Sars-CoV-2 Does Not Increase Rates of Mixed Infection

Main textTo the editor,We read with interest the letter by Zhang and colleagues1Zhang Y. Jiang N. Qi W. Li T. Zhang Y. Zhang H. et al.Intra-host SARS-CoV-2 single-nucleotide variants emerged during the early stage of COVID-19 pandemic forecast population fixing mutations.J Infect [Internet]. Jan 2022; ([cited 2022 Apr 4]; Available from https://pubmed.ncbi.nlm.nih.gov/35041922/)Abstract Full Text Full Text PDF Scopus (1) Google Scholar which reported intra host gene variants occurring during COVID-19 infection. The co-circulation of novel variants raises the possibility of mixed infections, leading to a theoretical risk of generating interlineage recombinants within individual hosts. This of particular concern in the hospital setting, where multiple infectious patients are in close proximity.At the time of writing, the WHO has recognised five variants of concern (VOCs), demonstrating variable pathogenicity, transmissibility, and potential for immune evasion. The Omicron variant (B.1.1.529) is the most recently characterised, and was designated as a novel variant of concern on 26 November 2021.2WHO. Classification of Omicron (B.1.1.529): sARS-CoV-2 Variant of Concern [Internet]. 2021 [cited 2022 Mar 28]. Available from: https://www.who.int/news/item/26-11-2021-classification-of-omicron-(b.1.1.529)-sars-cov-2-variant-of-concernGoogle Scholar This variant harbors a large number of mutations in the genome including in the S gene, particularly focused around the region encoding the receptor binding motif.3Mannar D. Saville J.W. Zhu X. Srivastava S.S. Berezuk A.M. Tuttle K.S. et al.SARS-CoV-2 Omicron variant: antibody evasion and cryo-EM structure of spike protein–ACE2 complex.Science 80-[Internet]. 2022 Feb 18; ([cited 2022 Mar 28];375(6582):760–4. Available from https://www.science.org/doi/full/10.1126/science.abn7760)Google Scholar,4McCallum M. Czudnochowski N. Rosen L.E. Zepeda S.K. Bowen J.E. Walls A.C. et al.Structural basis of SARS-CoV-2 Omicron immune evasion and receptor engagement.Science [Internet]. 2022 Feb 25; ([cited 2022 Mar 28];eabn8652. Available from http://www.ncbi.nlm.nih.gov/pubmed/35076256)Crossref Scopus (125) Google Scholar The omicron variant has demonstrated higher rates of reinfection when compared with other variants. This likely arises from partial immune evasion, particularly reduced serum antibody neutralization.5Garcia-Beltran W.F. Denis-KJ St. Hoelzemer A. Lam E.C. Nitido A.D. Sheehan M.L. mRNA-based COVID-19 vaccine boosters induce neutralizing immunity against SARS-CoV-2 Omicron variant.Cell [Internet]. 2022 Feb 3; ([cited 2022 Mar 28];185(3):457–466.e4. Available from https://pubmed.ncbi.nlm.nih.gov/34995482/)Abstract Full Text Full Text PDF Scopus (332) Google Scholar, 6Dejnirattisai W. Shaw R.H. Supasa P. Liu C. Stuart A.S. Pollard A.J. et al.Reduced neutralisation of SARS-CoV-2 omicron B.1.1.529 variant by post-immunisation serum.Lancet [Internet]. 2022 Jan 15; ([cited 2022 Mar 28];399(10321):234–6. Available from https://pubmed.ncbi.nlm.nih.gov/34942101/)Abstract Full Text Full Text PDF Scopus (139) Google Scholar, 7Rössler A. Riepler L. Bante D. von Laer D. Kimpel J. SARS-CoV-2 Omicron variant neutralization in serum from vaccinated and convalescent persons.N Engl J Med [Internet]. 2022 Feb 17; ([cited 2022 Mar 28];386(7):698–700. Available from https://www.nejm.org/doi/full/10.1056/NEJMc2119236)Crossref Scopus (139) Google ScholarIn acute hospital settings in the UK, patients testing positive for SARS-CoV-2 are either isolated in a side-room or admitted to a cohort bay alongside other positive patients.8UKHSA. COVID-19: infection prevention and control (IPC) - GOV.UK [Internet]. [cited 2022 Mar 24]. Available from: https://www.gov.uk/government/publications/wuhan-novel-coronavirus-infection-prevention-and-controlGoogle Scholar Cohort arrangements are variant agnostic, as genotyping is either not-available, not-performed or the results are not available within a clinically meaningful timeframe. The cohorting of patients infected with different VOCs could be associated with a theoretical risk of re-infection or super-infection -- particularly when there is limited cross-immunity between variants. In addition there is a theoretical risk of generating interlineage recombinants when mixing patients infected with different VOC.9Jackson B. Boni M.F. Bull M.J. Colleran A. Colquhoun R.M. Darby A.C. et al.Generation and transmission of interlineage recombinants in the SARS-CoV-2 pandemic.Cell. 2021 Sep 30; 184 (.e8): 5179-5188Abstract Full Text Full Text PDF PubMed Scopus (53) Google ScholarWe assessed the practicalities and safety of cohorting patients at a time of co-circulation of delta (B.1.617.2) and omicron variants. We retrospectively assessed a series of patients admitted to a designated COVID-19 ward over a three-week period. Patients admitted to hospital with a positive lateral flow test, a positive PCR test, or a positive test in the community within 14 days preceding admission and triaged on clinical grounds to one of the general medical wards were included. These patients underwent twice-weekly PCR testing for SARS-CoV-2. Positive samples were genotyped based on SNP typing.A total of 76 patients were treated as inpatients on the COVID ward over the 22 day period between 16th December 2021 and 7th January 2022. Of these 72/76 (95%) were tested on admission and 4/76 (5%) were moved to a COVID cohort ward pending a confirmatory PCR test having tested positive in the community. The median age was 69 years (range 18 to 101). The median length of stay was 8 days (range 1 to 42), contributing a total of 875 patient days. Of the 72 patients tested on admission, 66/72 (92%) tested positive, of which 58/72 (81%) were genotyped. Of the samples genotyped during the study period, 27/58 (47%) were classified as delta, and 27/58 (47%) were classified as omicron. Genotyping was indeterminate in 4/58 samples.Serial naso-pharyngeal swab sampling was performed twice weekly for the duration of their hospital stay. Paired samples were available for 36/72 (50%) patients, of which 27 (38%) were genotyped. Of these, 15 patients were infected with the Delta variant, 11 with Omicron and 2 were indeterminate. In all instances, the genotype identified on second swab was congruent with that identified on initial testing (Fig. 1). Three patients who had a positive result on initial swabbing were found to be negative on repeat PCR testing, whilst five did not undergo variant sequencing.Eight patients had a third serial swab genotyped. In all instances, the variant identified on repeat sampling corresponded to that initially identified (5/8 Delta infection, 3/8 Omicron - (Fig. 1). Two patients had a positive fourth swab genotyped, both of which were congruent with initial (2/2 Delta infection).Through repeated swabbing of inpatients on a COVID ward, we have attempted to evaluate the risk of co-infection and serial infection with different SARS-CoV-2 variants of concern inherent to current cohorting practices. Despite a sample size of 76 patients, we did not see any instances of co-infection with distinct variants of SARS-Cov-2, nor did we see any acquisition of alternative genotype during inpatient stay. These results suggest that current infection control practices – where bed management is guided by positive swab status rather than genotyping – are adequate for patient safety. This appears to apply during the acute stage of infection and convalescence despite uncertainty around the degree of antibody cross-neutralisation between different VOCs.We acknowledge the limitations of our study. In particular, the number of patients undergoing sequencing following serial sampling was limited, as this was an opportunistic study. In addition, SARS-CoV-2 variants were identified based on SNP typing as opposed to whole genome sequencing, limiting the resolution of our analysis. Future studies should include structured sampling timepoints, alongside sampling of healthcare workers and the environment and supported by whole genome sequencing.These findings are relevant to inform infection control procedures during times when different VOCs are co-circulating, such as during the delta and omicron wave. These findings give a degree of re-assurance that the cohorting of SARS-CoV-2 positive patients is safe at the point of admission in the absence of a genotype result. They are likely to be of increasing importance as novel variants of concern emerge where the degree of cross-immunity remains uncertain. We would advocate for continued active genomic surveillance of patients hospitalized with SARS-CoV-2 should continue, as there remains a risk of generating new interlineage SARS-CoV-2 recombinants when cohorting patients within healthcare settings. Main textTo the editor,We read with interest the letter by Zhang and colleagues1Zhang Y. Jiang N. Qi W. Li T. Zhang Y. Zhang H. et al.Intra-host SARS-CoV-2 single-nucleotide variants emerged during the early stage of COVID-19 pandemic forecast population fixing mutations.J Infect [Internet]. Jan 2022; ([cited 2022 Apr 4]; Available from https://pubmed.ncbi.nlm.nih.gov/35041922/)Abstract Full Text Full Text PDF Scopus (1) Google Scholar which reported intra host gene variants occurring during COVID-19 infection. The co-circulation of novel variants raises the possibility of mixed infections, leading to a theoretical risk of generating interlineage recombinants within individual hosts. This of particular concern in the hospital setting, where multiple infectious patients are in close proximity.At the time of writing, the WHO has recognised five variants of concern (VOCs), demonstrating variable pathogenicity, transmissibility, and potential for immune evasion. The Omicron variant (B.1.1.529) is the most recently characterised, and was designated as a novel variant of concern on 26 November 2021.2WHO. Classification of Omicron (B.1.1.529): sARS-CoV-2 Variant of Concern [Internet]. 2021 [cited 2022 Mar 28]. Available from: https://www.who.int/news/item/26-11-2021-classification-of-omicron-(b.1.1.529)-sars-cov-2-variant-of-concernGoogle Scholar This variant harbors a large number of mutations in the genome including in the S gene, particularly focused around the region encoding the receptor binding motif.3Mannar D. Saville J.W. Zhu X. Srivastava S.S. Berezuk A.M. Tuttle K.S. et al.SARS-CoV-2 Omicron variant: antibody evasion and cryo-EM structure of spike protein–ACE2 complex.Science 80-[Internet]. 2022 Feb 18; ([cited 2022 Mar 28];375(6582):760–4. Available from https://www.science.org/doi/full/10.1126/science.abn7760)Google Scholar,4McCallum M. Czudnochowski N. Rosen L.E. Zepeda S.K. Bowen J.E. Walls A.C. et al.Structural basis of SARS-CoV-2 Omicron immune evasion and receptor engagement.Science [Internet]. 2022 Feb 25; ([cited 2022 Mar 28];eabn8652. Available from http://www.ncbi.nlm.nih.gov/pubmed/35076256)Crossref Scopus (125) Google Scholar The omicron variant has demonstrated higher rates of reinfection when compared with other variants. This likely arises from partial immune evasion, particularly reduced serum antibody neutralization.5Garcia-Beltran W.F. Denis-KJ St. Hoelzemer A. Lam E.C. Nitido A.D. Sheehan M.L. mRNA-based COVID-19 vaccine boosters induce neutralizing immunity against SARS-CoV-2 Omicron variant.Cell [Internet]. 2022 Feb 3; ([cited 2022 Mar 28];185(3):457–466.e4. Available from https://pubmed.ncbi.nlm.nih.gov/34995482/)Abstract Full Text Full Text PDF Scopus (332) Google Scholar, 6Dejnirattisai W. Shaw R.H. Supasa P. Liu C. Stuart A.S. Pollard A.J. et al.Reduced neutralisation of SARS-CoV-2 omicron B.1.1.529 variant by post-immunisation serum.Lancet [Internet]. 2022 Jan 15; ([cited 2022 Mar 28];399(10321):234–6. Available from https://pubmed.ncbi.nlm.nih.gov/34942101/)Abstract Full Text Full Text PDF Scopus (139) Google Scholar, 7Rössler A. Riepler L. Bante D. von Laer D. Kimpel J. SARS-CoV-2 Omicron variant neutralization in serum from vaccinated and convalescent persons.N Engl J Med [Internet]. 2022 Feb 17; ([cited 2022 Mar 28];386(7):698–700. Available from https://www.nejm.org/doi/full/10.1056/NEJMc2119236)Crossref Scopus (139) Google ScholarIn acute hospital settings in the UK, patients testing positive for SARS-CoV-2 are either isolated in a side-room or admitted to a cohort bay alongside other positive patients.8UKHSA. COVID-19: infection prevention and control (IPC) - GOV.UK [Internet]. [cited 2022 Mar 24]. Available from: https://www.gov.uk/government/publications/wuhan-novel-coronavirus-infection-prevention-and-controlGoogle Scholar Cohort arrangements are variant agnostic, as genotyping is either not-available, not-performed or the results are not available within a clinically meaningful timeframe. The cohorting of patients infected with different VOCs could be associated with a theoretical risk of re-infection or super-infection -- particularly when there is limited cross-immunity between variants. In addition there is a theoretical risk of generating interlineage recombinants when mixing patients infected with different VOC.9Jackson B. Boni M.F. Bull M.J. Colleran A. Colquhoun R.M. Darby A.C. et al.Generation and transmission of interlineage recombinants in the SARS-CoV-2 pandemic.Cell. 2021 Sep 30; 184 (.e8): 5179-5188Abstract Full Text Full Text PDF PubMed Scopus (53) Google ScholarWe assessed the practicalities and safety of cohorting patients at a time of co-circulation of delta (B.1.617.2) and omicron variants. We retrospectively assessed a series of patients admitted to a designated COVID-19 ward over a three-week period. Patients admitted to hospital with a positive lateral flow test, a positive PCR test, or a positive test in the community within 14 days preceding admission and triaged on clinical grounds to one of the general medical wards were included. These patients underwent twice-weekly PCR testing for SARS-CoV-2. Positive samples were genotyped based on SNP typing.A total of 76 patients were treated as inpatients on the COVID ward over the 22 day period between 16th December 2021 and 7th January 2022. Of these 72/76 (95%) were tested on admission and 4/76 (5%) were moved to a COVID cohort ward pending a confirmatory PCR test having tested positive in the community. The median age was 69 years (range 18 to 101). The median length of stay was 8 days (range 1 to 42), contributing a total of 875 patient days. Of the 72 patients tested on admission, 66/72 (92%) tested positive, of which 58/72 (81%) were genotyped. Of the samples genotyped during the study period, 27/58 (47%) were classified as delta, and 27/58 (47%) were classified as omicron. Genotyping was indeterminate in 4/58 samples.Serial naso-pharyngeal swab sampling was performed twice weekly for the duration of their hospital stay. Paired samples were available for 36/72 (50%) patients, of which 27 (38%) were genotyped. Of these, 15 patients were infected with the Delta variant, 11 with Omicron and 2 were indeterminate. In all instances, the genotype identified on second swab was congruent with that identified on initial testing (Fig. 1). Three patients who had a positive result on initial swabbing were found to be negative on repeat PCR testing, whilst five did not undergo variant sequencing.Eight patients had a third serial swab genotyped. In all instances, the variant identified on repeat sampling corresponded to that initially identified (5/8 Delta infection, 3/8 Omicron - (Fig. 1). Two patients had a positive fourth swab genotyped, both of which were congruent with initial (2/2 Delta infection).Through repeated swabbing of inpatients on a COVID ward, we have attempted to evaluate the risk of co-infection and serial infection with different SARS-CoV-2 variants of concern inherent to current cohorting practices. Despite a sample size of 76 patients, we did not see any instances of co-infection with distinct variants of SARS-Cov-2, nor did we see any acquisition of alternative genotype during inpatient stay. These results suggest that current infection control practices – where bed management is guided by positive swab status rather than genotyping – are adequate for patient safety. This appears to apply during the acute stage of infection and convalescence despite uncertainty around the degree of antibody cross-neutralisation between different VOCs.We acknowledge the limitations of our study. In particular, the number of patients undergoing sequencing following serial sampling was limited, as this was an opportunistic study. In addition, SARS-CoV-2 variants were identified based on SNP typing as opposed to whole genome sequencing, limiting the resolution of our analysis. Future studies should include structured sampling timepoints, alongside sampling of healthcare workers and the environment and supported by whole genome sequencing.These findings are relevant to inform infection control procedures during times when different VOCs are co-circulating, such as during the delta and omicron wave. These findings give a degree of re-assurance that the cohorting of SARS-CoV-2 positive patients is safe at the point of admission in the absence of a genotype result. They are likely to be of increasing importance as novel variants of concern emerge where the degree of cross-immunity remains uncertain. We would advocate for continued active genomic surveillance of patients hospitalized with SARS-CoV-2 should continue, as there remains a risk of generating new interlineage SARS-CoV-2 recombinants when cohorting patients within healthcare settings. To the editor, We read with interest the letter by Zhang and colleagues1Zhang Y. Jiang N. Qi W. Li T. Zhang Y. Zhang H. et al.Intra-host SARS-CoV-2 single-nucleotide variants emerged during the early stage of COVID-19 pandemic forecast population fixing mutations.J Infect [Internet]. Jan 2022; ([cited 2022 Apr 4]; Available from https://pubmed.ncbi.nlm.nih.gov/35041922/)Abstract Full Text Full Text PDF Scopus (1) Google Scholar which reported intra host gene variants occurring during COVID-19 infection. The co-circulation of novel variants raises the possibility of mixed infections, leading to a theoretical risk of generating interlineage recombinants within individual hosts. This of particular concern in the hospital setting, where multiple infectious patients are in close proximity. At the time of writing, the WHO has recognised five variants of concern (VOCs), demonstrating variable pathogenicity, transmissibility, and potential for immune evasion. The Omicron variant (B.1.1.529) is the most recently characterised, and was designated as a novel variant of concern on 26 November 2021.2WHO. Classification of Omicron (B.1.1.529): sARS-CoV-2 Variant of Concern [Internet]. 2021 [cited 2022 Mar 28]. Available from: https://www.who.int/news/item/26-11-2021-classification-of-omicron-(b.1.1.529)-sars-cov-2-variant-of-concernGoogle Scholar This variant harbors a large number of mutations in the genome including in the S gene, particularly focused around the region encoding the receptor binding motif.3Mannar D. Saville J.W. Zhu X. Srivastava S.S. Berezuk A.M. Tuttle K.S. et al.SARS-CoV-2 Omicron variant: antibody evasion and cryo-EM structure of spike protein–ACE2 complex.Science 80-[Internet]. 2022 Feb 18; ([cited 2022 Mar 28];375(6582):760–4. Available from https://www.science.org/doi/full/10.1126/science.abn7760)Google Scholar,4McCallum M. Czudnochowski N. Rosen L.E. Zepeda S.K. Bowen J.E. Walls A.C. et al.Structural basis of SARS-CoV-2 Omicron immune evasion and receptor engagement.Science [Internet]. 2022 Feb 25; ([cited 2022 Mar 28];eabn8652. Available from http://www.ncbi.nlm.nih.gov/pubmed/35076256)Crossref Scopus (125) Google Scholar The omicron variant has demonstrated higher rates of reinfection when compared with other variants. This likely arises from partial immune evasion, particularly reduced serum antibody neutralization.5Garcia-Beltran W.F. Denis-KJ St. Hoelzemer A. Lam E.C. Nitido A.D. Sheehan M.L. mRNA-based COVID-19 vaccine boosters induce neutralizing immunity against SARS-CoV-2 Omicron variant.Cell [Internet]. 2022 Feb 3; ([cited 2022 Mar 28];185(3):457–466.e4. Available from https://pubmed.ncbi.nlm.nih.gov/34995482/)Abstract Full Text Full Text PDF Scopus (332) Google Scholar, 6Dejnirattisai W. Shaw R.H. Supasa P. Liu C. Stuart A.S. Pollard A.J. et al.Reduced neutralisation of SARS-CoV-2 omicron B.1.1.529 variant by post-immunisation serum.Lancet [Internet]. 2022 Jan 15; ([cited 2022 Mar 28];399(10321):234–6. Available from https://pubmed.ncbi.nlm.nih.gov/34942101/)Abstract Full Text Full Text PDF Scopus (139) Google Scholar, 7Rössler A. Riepler L. Bante D. von Laer D. Kimpel J. SARS-CoV-2 Omicron variant neutralization in serum from vaccinated and convalescent persons.N Engl J Med [Internet]. 2022 Feb 17; ([cited 2022 Mar 28];386(7):698–700. Available from https://www.nejm.org/doi/full/10.1056/NEJMc2119236)Crossref Scopus (139) Google Scholar In acute hospital settings in the UK, patients testing positive for SARS-CoV-2 are either isolated in a side-room or admitted to a cohort bay alongside other positive patients.8UKHSA. COVID-19: infection prevention and control (IPC) - GOV.UK [Internet]. [cited 2022 Mar 24]. Available from: https://www.gov.uk/government/publications/wuhan-novel-coronavirus-infection-prevention-and-controlGoogle Scholar Cohort arrangements are variant agnostic, as genotyping is either not-available, not-performed or the results are not available within a clinically meaningful timeframe. The cohorting of patients infected with different VOCs could be associated with a theoretical risk of re-infection or super-infection -- particularly when there is limited cross-immunity between variants. In addition there is a theoretical risk of generating interlineage recombinants when mixing patients infected with different VOC.9Jackson B. Boni M.F. Bull M.J. Colleran A. Colquhoun R.M. Darby A.C. et al.Generation and transmission of interlineage recombinants in the SARS-CoV-2 pandemic.Cell. 2021 Sep 30; 184 (.e8): 5179-5188Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar We assessed the practicalities and safety of cohorting patients at a time of co-circulation of delta (B.1.617.2) and omicron variants. We retrospectively assessed a series of patients admitted to a designated COVID-19 ward over a three-week period. Patients admitted to hospital with a positive lateral flow test, a positive PCR test, or a positive test in the community within 14 days preceding admission and triaged on clinical grounds to one of the general medical wards were included. These patients underwent twice-weekly PCR testing for SARS-CoV-2. Positive samples were genotyped based on SNP typing. A total of 76 patients were treated as inpatients on the COVID ward over the 22 day period between 16th December 2021 and 7th January 2022. Of these 72/76 (95%) were tested on admission and 4/76 (5%) were moved to a COVID cohort ward pending a confirmatory PCR test having tested positive in the community. The median age was 69 years (range 18 to 101). The median length of stay was 8 days (range 1 to 42), contributing a total of 875 patient days. Of the 72 patients tested on admission, 66/72 (92%) tested positive, of which 58/72 (81%) were genotyped. Of the samples genotyped during the study period, 27/58 (47%) were classified as delta, and 27/58 (47%) were classified as omicron. Genotyping was indeterminate in 4/58 samples. Serial naso-pharyngeal swab sampling was performed twice weekly for the duration of their hospital stay. Paired samples were available for 36/72 (50%) patients, of which 27 (38%) were genotyped. Of these, 15 patients were infected with the Delta variant, 11 with Omicron and 2 were indeterminate. In all instances, the genotype identified on second swab was congruent with that identified on initial testing (Fig. 1). Three patients who had a positive result on initial swabbing were found to be negative on repeat PCR testing, whilst five did not undergo variant sequencing. Eight patients had a third serial swab genotyped. In all instances, the variant identified on repeat sampling corresponded to that initially identified (5/8 Delta infection, 3/8 Omicron - (Fig. 1). Two patients had a positive fourth swab genotyped, both of which were congruent with initial (2/2 Delta infection). Through repeated swabbing of inpatients on a COVID ward, we have attempted to evaluate the risk of co-infection and serial infection with different SARS-CoV-2 variants of concern inherent to current cohorting practices. Despite a sample size of 76 patients, we did not see any instances of co-infection with distinct variants of SARS-Cov-2, nor did we see any acquisition of alternative genotype during inpatient stay. These results suggest that current infection control practices – where bed management is guided by positive swab status rather than genotyping – are adequate for patient safety. This appears to apply during the acute stage of infection and convalescence despite uncertainty around the degree of antibody cross-neutralisation between different VOCs. We acknowledge the limitations of our study. In particular, the number of patients undergoing sequencing following serial sampling was limited, as this was an opportunistic study. In addition, SARS-CoV-2 variants were identified based on SNP typing as opposed to whole genome sequencing, limiting the resolution of our analysis. Future studies should include structured sampling timepoints, alongside sampling of healthcare workers and the environment and supported by whole genome sequencing. These findings are relevant to inform infection control procedures during times when different VOCs are co-circulating, such as during the delta and omicron wave. These findings give a degree of re-assurance that the cohorting of SARS-CoV-2 positive patients is safe at the point of admission in the absence of a genotype result. They are likely to be of increasing importance as novel variants of concern emerge where the degree of cross-immunity remains uncertain. We would advocate for continued active genomic surveillance of patients hospitalized with SARS-CoV-2 should continue, as there remains a risk of generating new interlineage SARS-CoV-2 recombinants when cohorting patients within healthcare settings. No conflicts of interest exist. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. M.M.G. is supported in part by the NIHR Imperial Biomedical Research Centre.