Noroviruses (NV) are a diverse group of non-enveloped RNA viruses belonging to a genus within the family Caliciviridae. Previous names for NV infection have included: winter vomiting, Small Round Structured Virus (SRSV) infection and Norwalk-like virus (NLV) infection. The virus was first recognised more than 25 years ago, from a outbreak of gastro-enteritis in a school in the town of Norwalk, Ohio, USA. Vomiting is the prominent symptom and is often spectacularly projectile. The vomiting may not be preceded by any prodromal symptoms, and cases cannot therefore be isolated in anticipation of its occurrence. Diarrhoea tends to be short-lived and less severe than with other forms of gastro-enteritis. Other symptoms include nausea, abdominal cramps, headache, myalgia, chills and fever. Symptoms typically last between one and three days and recovery is usually rapid. NV affects people of all age groups, although in Scotland 75% of laboratory reports are from cases aged over 65 years. Humans are the only known hosts of NV. NV is highly infectious and may be acquired through consuming contaminated food or water, or more commonly from an infected person via the faeco-oral route, through fomites, or by aerosol spread, particularly when a case vomits. Semi-closed communities such as hotels, cruise ships and hospitals are particularly prone to large outbreaks.
Two surveillance systems operate in Scotland for NV infection: one for laboratory identifications of NV and the second for general outbreaks of infectious intestinal disease.
Surveillance in Scotland (and the rest of the UK) is based upon the voluntary reporting of identifications of the virus by clinical diagnostic laboratories to the national surveillance centre. Until recently, the main method by which NV was identified was by electron microscopy (EM) of stools. This is a relatively insensitive method and is reliant upon expensive and rare equipment and considerable expertise on the part of the virologist. In recent years, some laboratories have been using polymerase chain reaction (PCR) methods for identification, which has probably resulted in many more specimens being examined in Scotland. From laboratory surveillance, SCIEH cannot reliably distinguish between cases from outbreaks and sporadic cases. The annual number of reports to SCIEH of NV identifications has been increasing year on year since 1997. A fourfold increase, from 348 to 1477, occurred between 2001 and 2002. It is noteworthy that the number of laboratory reports has declined in Scotland for the first 16 weeks of 2004, compared to the corresponding period in 2003, from 1035 to 372. The seasonality of NV infection is well documented, and is confirmed by the laboratory surveillance in Scotland in most years.
It should be noted however that this seasonality is not as pronounced every year: in 2002 there was a lesser decline in the summer months, similarly in England and Wales a summer peak of NV reports was observed in 2002 (1).
In 1996, the SCIEH established ‘ObSurv’, a surveillance system for general outbreaks of infectious intestinal disease (IID) in Scotland. For the purpose of ObSurv an outbreak is defined as an incident in which two or more linked cases experience the same illness, or when the observed number of cases unaccountably exceeds the expected number. The system seeks information on general outbreaks, defined as outbreaks affecting members of more than one household or residents of an institution. Since September 2002, outbreaks of IID in healthcare settings have also been included by SCIEH in a surveillance system for outbreaks of Healthcare Associated Infection (HAI) that has sought to collect additional information. Whenever SCIEH becomes aware of a potential outbreak of IID, by whatever route, it dispatches a standard form to the appropriate Consultant in Public Health Medicine, to be completed at the end of the outbreak’s investigation. This IID surveillance system while it may be subject to varying levels of ascertainment resulting, for example, from an increased media profile of ‘winter vomiting’ will be unaffected by technological improvements in the identification of NV, although the proportion of outbreaks identified that are confirmed as NV may increase. The total number of general outbreaks of IID due to confirmed and suspected NV infection, suspected viral infection, and unknown aetiology (i.e. all potentially NV infection outbreaks) rose from 28 in 1996 to 98 in 2000. A fall to 56 in 2001 has been followed by a continuing increase to 260 in 2002, and by a much smaller margin to 267 in 2003.
The seasonality of outbreaks reflects that of laboratory reports, outbreaks being less common in the third quarter, and most common in the first and fourth. 2002 was an exception to this rule with reports in the third quarter being higher than in quarters one or two. The continuing incidence of outbreaks from the winter of 2001/2002 through the summer of 2002 was overshadowed by the increase in the number of outbreaks reported in the winter of 2002/2003. It is noteworthy that the leveling off of laboratory reports in 2003 is not reflected in the outbreak surveillance data. The number of outbreaks reported as NV infection, in hospitals has risen from 7 in 1996 to 90 in 2003. The proportion in hospitals has shown no trend, remaining between 38% and 56%, implying that hospitals are not a special case, but reflect the experience in the rest of the community. Most residential institutions reported are care homes for the elderly. ‘Other’ is mainly accounted by the leisure industry, particularly hotels. It is likely that cases that occur in hospitals and residential institutions are more readily identified as part of an outbreak than cases in the community.
Control of NV infection in healthcare settings
Cowden published a summary of control measures for NV infection in hospitals in 2002 (2). This made no attempt to provide an evidence base for the measures suggested. Chadwick et al attempted a more detailed review of the evidence base for cardinal issues in control of outbreaks of NV infection in healthcare settings. The paper attempted to categorise the strength of evidence for each of the control measures it recommends as:
- Category I: Strongly recommended and strongly supported by well designed experimental epidemiological studies
- Category II: Strongly recommended and viewed as effective by experts in the field and by the working group, based on strong rationale and suggestive evidence even though definitive studies may not have been done.
- No Category: an unresolved issue as there is insufficient evidence or consensus regarding efficacy
Chadwick et al (3) noted that in principal, there are three points at which NV transmission may be controlled: the introduction of the NV into the hospital or other healthcare environment; containment measures at individual ward level; and measures to prevent spread to other wards. Controlling introduction of NV into healthcare settings is difficult, and measures are therefore best directed at the latter two points. Despite the attempt at categorisation, the paper however did not provide the details upon which each measure is assessed. For example, it does not reference the ‘well designed experimental epidemiological studies’ which support ‘Category I’ measures, or name the ‘experts in the field [who] strongly recommend’ ‘Category II’ measures. The paper, while recommending hand hygiene, failed to highlight the importance of the correct technique for hand decontamination that is required to control the spread of all micro-organisms including NV. Although the paper has these limitations, it does reflect reasonably accurately the current views on specific control measures for NV infection in healthcare settings.
Specific control measures
Recommended control measures and their stated categories of supporting evidence are: (Tables reproduced from Chadwick et al. (3))
Recommended control measures for SRSVs within affected clinical areas:
- Cohort nurse or isolate symptomatic individuals (Category II).
- Wear gloves and apron for contact with an affected patient or environment (Category II).
- Wash hands with soap and water after contact with an affected patient or environment, after removing gloves and apron (Category I).
- Remove exposed food such as fruit (No Category).
- Consider use of antiemetics for patients with vomiting (No Category).
- Exclude affected staff from the ward immediately and until 48hrs symptom-free (Category II).
- Close the ward to prevent the introduction of new susceptibles. Avoid transfer to unaffected wards or departments (unless medically urgent and after consultation with infection control staff). The priority is to stop spread of the virus to other areas (Category II).
- Exclude non-essential personnel from the ward (Category II).
- Caution visitors and emphasize hand hygiene (Category II).
- Clean and disinfect vomit and faeces spillages promptly (Category II).
- Increase the frequency of routine ward, bathroom and toilet cleaning (Category II).
- Use freshly prepared 0.1% (1000ppm) hypochlorite to disinfect hard surfaces after cleaning (Category II).
- The ward should not be re-opened until 72hrs after the last new case and 72hrs after uncontained vomiting and diarrhoea (Category II).
- Thoroughly clean the ward and change the bed curtains before re-opening (Category II).
- Clean carpets and soft furnishings with hot water and detergent, or steam clean. Vacuum cleaning is not recommended (No Category).
Recommendations for preventing spread of SRSVs to unaffected areas:
- Staff working in affected areas must not then work in unaffected areas for 48h (includes agency and bank staff) (Category II).
- Avoid patient movements to unaffected areas (unless medically urgent and after consultation with infection control staff) and other institutions (Category II).
Other specific control measures are discussed within the text of the paper but did not appear in the tables themselves. These include:
- Suspected point source food or waterborne outbreaks should be managed appropriately.
- Staff movement from affected to unaffected areas should be limited. If this is unavoidable staff should visit the affected area last, and they must leave an area immediately if nausea occurs.
- Block bookings should be used for bank/agency staff and movement between areas avoided.
- Patients in affected areas should not normally be discharged to other institutions, such as residential and nursing homes until 72hrs after the last documented case (unless they have actually recovered from the virus).
- Non-essential visitors, especially children should be discouraged.
- Terminal cleaning should not commence until 72hrs after the onset of last case and 72 hrs since uncontrolled vomiting or diarrhoea with contamination of the ward area (if patients present with faeces unformed but contained – can carry on with terminal clean).
- Communication, which could stop spread is essential – e.g. to the Infection Control Team, all areas, occupational health, managers.
An additional point not specifically detailed previously, was recognised as important while preparing an Aide Memoir for Managing NV outbreaks in NHS Hospitals (in draft – SCIEH). The point, ‘guidelines and decontamination facilities should be available to staff who become grossly contaminated from body fluids’ was specifically drawn for this memoir from the Watt Report (2002).
In 2003, SCIEH produced guidelines for the management of NV infections in the leisure industry setting, which were well received. (http://www.show.scot.nhs.uk/scieh/documents/NV_Guide/NV_Guide.pdf). A second edition of these guidelines is currently being produced.
Despite possible artefacts in the surveillance data, NV infection appears to have increased in Scotland between 1996 and 2003 (despite a small decline in laboratory reports in 2003). It is unclear how much of the large increase in NV infection reported since 2001 was due to improved identification and reporting and how much to a real increase. The view of clinicians, virologists, public health doctors and infection control nurses is pretty consistent, albeit anecdotal and not amenable to rigorous confirmation: is that at least part of the increase is real. Even against the background of declining laboratory reports of NV in 2003, the number of outbreaks of NV infection rose from 187 to 205. The total number of outbreaks of potentially NV infection also rose from 260 to 267. The outbreak surveillance data therefore cast doubt on the conclusion that might be drawn from the laboratory surveillance that the burden of illness from NV infection fell in 2003.
Laboratory surveillance substantially underestimates the true burden of NV infection as, due to the relatively mild and brief nature of illness, cases in the community rarely seek medical attention. A study in England and Wales in the early 1990s concluded that as many as 1500 cases occur in the community for every one that is confirmed and reported (4). It is unlikely that the situation in Scotland is very different, thus emphasising that the incidence of NV infection is far greater than that captured by surveillance.
The incidence of NV in England and Wales also showed a large increase from 2001 to 2002, and also a decrease in 2003 (5,6). Interestingly the rate of laboratory reports in Scotland is noticeably higher than in England and Wales. In 2003, there were 2041 laboratory reports in England and Wales, compared to 1443 in Scotland, despite the population in England and Wales being approximately ten times greater than that of Scotland.
The increasing incidence of NV infection in Scotland is not dissimilar to the situation elsewhere in the UK, Europe (7) and Hong Kong (8). Data from the USA are less helpful, as laboratory reporting is not universal and no surveillance system for non-foodborne outbreaks exists. Increased norovirus activity was, however, reported in the USA during 2002 (9).
The exact reasons for this widespread increase in NV infection is not understood. Lopman et al (7) in reporting on outbreaks in Europe documented the emergence, and subsequent predominance of a new strain within the genotype II4 genotype. They postulate that this new variant might be more virulent or environmentally stable than previous genotype II4 virus, and consequently might have increased risk of epidemic spread.
Outbreaks of NV infection can place a considerable burden on the health service through restricted admissions, ward or hospitals closures, delayed discharge, cancelled operations and staff shortages, despite this there is hardly any published information on the cost of such outbreaks to the health service or indeed to other establishments in the tourist or educational sphere.
Very few of the measures recommended for the control of outbreaks of NV infection in hospital are evidence based, although the consensus is emerging about what measures are appropriate. In addition, other guidance documents produced outwith the UK feature similar control measures. Contributions from healthcare settings attempting different measures to control NV infection are important in order that evaluation can be conducted and sharing of new information can be facilitated. Particularly in outbreak situations, it is essential that specific control measures such as hand decontamination and environmental cleaning techniques are publicised and awareness heightened through the use of appropriate communication and poster campaigns.
Clear guidance on control measures must also be provided and publicised in all other settings where NV infection continues to cause concern e.g. the leisure industry.
In addition to continuing to develop its surveillance and providing operational support and training to those responsible for managing outbreaks, SCIEH is planning a project to better define not only the size and nature of the problem, but to identify the most effective control measures.
- Lopman BA, Reacher M, Gallimore C, Adak GK, Gray JJ, Brown DWG. A summertime peak of “winter vomiting disease”: Surveillance of noroviruses in England and Wales, 1995 to 2002. BMC Public Health 2003; 3: 13
- Cowden JM Winter Vomiting. (Editorial) BMJ 2002; 324: 249-250
- Chadwick PR, Beards G, Brown D, Caul EO, Cheesbrough J, Clark I, Curry A, O’Brien S, Quigley K, Sellwood J and Westmoreland D. Management of hospital outbreaks of gastroenteritis due to small round structured virus. Journal of Hospital Infection 2000; 45: 1-10
- Study team for the IID study A Report of the study of Infectious Intestinal Disease in England. Food Standards Agency, London. 2000.
- Communicable Disease Report: 9 Jan 2004 http://www.hpa.org.uk/cdr/PDFfiles/2004/cdr/0204.pdf
- Lopman B, Vennema H, Kohli E, Pothier P, Sanchez A, Negredo A, Buesa J, Schreier E, Reacher M, Brown D, Gray J, Iturriza M, Gallimore C, Bottiger B, Hedlund K-O, Torven M, von Bonsdorff C-H, Maunula L, Poljsak-Prijatelj M, Zimsek J, Reuter G, Szuces G, Melegh B, Svennson L, van Duijnhoven Y, Koopmans M for the European Food-borne Viruses Network . Increase in viral gastroenteritis outbreaks in Europe and epidemic spread of new norovirus variant. Lancet 2004; 363: 682-688.
- Leung KM, Lam R, Tsang T. (2004) Epidemiology of noroviruses – associated gastroenteritis institutional outbreaks in Hong Kong 2001-2003. Public Health and Epidemiology Bulletin. Department of Health Hong King, China. 2004; 13: 1-6. (http://www.info.gov.hk/dh/diseases/public.htm)
- Anderson L, et al Norovirus activity – United States, 2002. JAMA 2003; 289: 693-696.