Health care associated infections were first described in the 19th century (Ayliff et al 1990). These are infections often identified in the hospital setting and remain a major cause of patient morbidity and mortality. In England , nosocomial infections have been estimated to cost the national health service (NHS) nearly a billion pounds annually and an additional 3.65 million bed days( Plowmann et al 2001) , while in the United States of America they have been found to cost hospitals an average of over $15,000(Roberts et al 2003)
Gastrointestinal disease are amongst the most frequent nosocomial acquired infection in both developing and developed countries (Zsolt 2003) and over the years it has become increasingly clear that gastroenteritis outbreaks are now a major contributions to nosocomial infections in the developed nations. Over the years these gastrointestinal outbreaks in Europe and America are more commonly being associated with a viral organism called Norovirus(Lopman et al 2004), which has now become the most common cause of outbreaks of gastroenteritis in hospitals ( Chadwick et al 2000). Norovirus has a high community prevalence and has become one of the most important aetiological agent to cause epidemic nonbacterial gastroenteritis globally (Atmar and Estes 2006) and the most common cause of outbreaks of gastroenteritis in hospitals (Chadwick et al 2000). However Hospital infections can be acquired or imported from the community or could be a true hospital acquired infection. In this context hospital acquired gastroenteritis must be differentiated from community acquired gastroenteritis. This is where norovirus is differentiated from other true nosocomial infections because the first case of norovirus because the first case of norovirus in the hospital setting usually comes from the community.
Globally the virus causes about 95% of non bacterial gastroenteritis outbreaks and 50% of all gastrointestinal outbreaks (Karst 2010). In the United States of America (U.S.A) it is estimated that there are about 23 million cases of norovirus gastroenteritis annually (Mead et al 1999). The Department of Health (DOH) in England and Wales suggest that there are about 600,000 cases of norovirus infections annually (Vipond et al 2004), although it has been appreciated that this number is an underestimation (Dedman et al 1998). The United Kingdom normally experiences 130-250 reported outbreaks of norovirus gastroenteritis annually and has now become the most common cause of gastrointestinal disease in the country (Lopman et al 2003).
In 2002-03 norovirus was found to be the cause of over 60% of gastroenteritis in three hospitals in England (Lopman et al 2004) while in other studies carried out by Meakins et al (2003) and Lopman et al (2003) on gastrointestinal outbreaks in England between 1992 an 2000 found that more than a quarter of the 5000 norovirus occurred in hospitals and of these more than half were due to norovirus. Similarly in Switzerland during the years 2002-03 more than half of the norovirus outbreaks occurred in health care settings which reflected the increase of the virus prevalence in the community during the same period (Fretz et al 2005). When such outbreaks occur they have an enormous impact on hospital services, not only in terms of staff shortages and additional days spent by patients, but also in an attempt to control spread of the virus, can lead to closure of medical units such as wards and bays or even entire hospitals (Sawyer et al 1998, Chadwick et al 1994). Nosocomial outbreaks have been recognised as one of the common causes of hospital ward closures (Hansen et al 2007). In a report published by the DOH in 2002-03, it was identified that the percentage of ward and bay closured in due to infections in acute trusts in England and Wales had risen from 66% to 80% between 2000and 2003 and many of these were mainly due to norovirus gastroenteritis (NAO 2004)
2. INTRODUCTION TO THE VIRUS
2.1 HISTORICAL BCKGROUND
Epidemic non bacterial gastroenteritis was first described in 1929 by Zaborsky in the United States of America (U.S.A) (Zaborsky 1929) and he named it as 'Hyperemesis hyemis'. A non-bacterial aetiology was described for this disease, characterised by sudden onset of diarrhoea and fever ( Vipond et al 2008), causing outbreaks mainly in the winter season, thereby its name changed to winter vomiting disease( Vipond et al 1999). Different studies carried out in which human volunteers were inoculated with bacterial free infiltrates resulted in infections with clinical symptoms similar to winter vomiting disease (Reiman et al 1945, Gordon et al 1947 and Jordan et al 1953). These studies strengthened the argument of a viral aetiology as the cause of winter vomiting disease, but inability to isolate viruses at that time made if difficult to identify a specific viral agent.
It was four decades later, after the original description by Zaborsky in 1929, in 1968, during an outbreak of gastroenteritis in a school in Norwalk Ohio, U.S.A that a viral agent was identified by Kapikian et al (1972). With the help of an Electron Microscope (EM) they were able to identify a 32nm agent and it was called the Norwalk virus, named after the place it was discovered. In 1982 under a new classification scheme brought the birth of the term Small Round Structured Virus (SRSV). This new name was used to describe viruses that were structurally indistinguishable from the original Norwalk virus discovered by Kapikan and his co-workers (Vipond et al 1999). Finally the international committee on taxonomy of viruses settled that the current name for winter vomiting disease should be Norovirus and classified it as a member of the Caliciviridae family (Lopman et al 2008).
Since its discovery, the norovirus has becomes an important cause of sporadic and epidemic gastroenteritis globally (Anne et al 2003) and the most common cause of acute gastroenteritis in humans globally (Lopman et al 2003, Huston et al 2004, Atmer and Ester 2006, Koo et al 2010) .In the United states It is responsible for more than two thirds of all food borne gastroenteritis (Breese et al 2002) causing up to 23 million cases annually (Mead et al 1999). Outbreaks of gastroenteritis due to this virus have been identified in populations including the elderly (Green et al 2002), children (Patel et al 2008), the immuno-suppressed (Roddie et al 2009) , travellers (Koo et al 2010b), cruise ships (Widdowson et al 2004), aeroplanes (Widdowson et al 2005), hotels (Marks et al 2000), Schools ( Marks et al 2003),Nursing homes (Cooper et al 2002, Calderon et al 2005) and Hospitals (Lears et al 2007, Tseng et al 2008, Ohiwaka et al 2009). Norovirus outbreaks in hospital settings have been becoming more common over the past few years and this has been attributed to the high community prevalence of the virus, thereby making it difficult to prevent its introduction in to the health care setting (Lopman et al 2004)
For a long period the electron microscope had remained they main diagnostic tool for norwalk virus detection, but with the with advent of the polymerase chain reaction(PCR) the sensitivity of its detection has increased (Jiang et al 1992) very useful in norovirus detection in clinical settings(Koopmans 2008).
The norovirus belong to the caliciviridae family and have a genome composed a single stranded plus sense Ribonucleic acid (RNA) that is protected by a non-enveloped protein capsid (Wu et al 2005) . The presence of this protective capsid confers resistance of environmental degradation due to elevated temperature and chemical disinfection, a feature which accounts for high transmissibility of the virus. The virus is classified in to five genogroups GI- GV and each of these genogroups is further divided into genotypes. Norovirus strains are typically named after the location in which they were first identified and are commonly referred to by their genogroups and genotype. For instance the prototype norovirus , Norwalk virus, discovered by Kapikan et al (1972), was named after Norwalk Ohio, belongs to geno-group I type I and is represented as GI.I Nov (Bull et al 2006). Of the Five geno-groups known only GI, G II and GIV are known to affect humans while GIII are bovine and GV being murine strains (Zheng et al 2006). The great diversity exhibited by the virus is due to its ability to undergo point mutations and recombination of its genome, which has led to the emergence of novel strains (Bull et al 2006). Currently the most predominant circulating genotype identified is the GII.IV Nov which accounts for 60-90% of reported cases of norovirus outbreaks in the United Kingdom and globally (Gallimore et al 2007, Glass et al 2005). Since the 1990's the pandemic outbreaks that have occurred have been due to the emergence of different variants of this GII.IV strain. Such variants include the, US95/96 strain(Noel et al 1999, White et al 2002) Grimsby cluster between 1995-2002(), Framington hills clusters 2002-2004(Widdowson et al 2004), the Hunter cluster between 2004-2006(Bull et al 2006) and the Lordsdale cluster ( Noel et al 1999). Till 2002 the Lordsdale cluster (Vipond et al 2000) and more recently the Grimsby strain (Koopmans et al 2002) have been the most common cause outbreaks in health care settings. The ability of these variants to successfully cause nosocomial outbreaks have been linked to their improved biological fitness, a larger viral shedding in patients and diverse human receptor specificity ( Lopman et al 2004).
After its discovery in the early 70's by Kapikian et al (1972), the only means of detecting norovirus was by an electron microscope. The EM can be used to detect norovirus particles in stool, but using the EM requires that there is a viral load of at least a million viral particles per gram of stool (10/gm)(). This means that it can only be used in the very early part of the illness (Atmar and Estes 2001). The consequent inability to grow that virus in cell culture made in difficult to develop reactive antibodies for immunological methods of detection and when eventually developed, the first immunological assays (Greenberg et al 1978) had similar low sensitivities as compared to the EM. However with the development of the Polymerase chain reaction ( PCR) in the early 1990's (Jiang et al 1992) that sensitivity of detection increased, and with innovation of real time PCR (RT-PCR) which added more sensitivity (Pang et al 2005, Nordgren et al 2008), the RT-PCR has become the gold standard for the diagnosis and detection of norovirus.
3. EPIDEMIOLOGY OF NOROVIRUS
3.1 HOST SUSCEPTIBILITY
Furthermore human susceptibility to norovirus infection showed that study participants were repeatedly has been related to certain antigens the human host may possess. The potential role of genetic resistance was first recognised more than three decades ago when experimental human infection studies with norovirus showed that study participants were repeatedly susceptible or resistant to symptomatic infection following repeated viral challenge (Parrino et al 1977). The first of these is the Histo Blood Group Antigen (HBGA) H1 which is coded by the secretor gene FUT 2( ). Non secretors of FUT2 have been found to be resistant to infections by GI and GII norovirus strains (Huston et al 2002, Ko et al 2010). In a study carried out by Guyader et al (2010) on an outbreak of norovirus affecting 33 people, non secretors FUT2 were found to have less significant symptoms. In addition to the HBG antigen, studies carried out by Prasad et al (1999), Hugg et al (2002) and Huston et al (2002) on norwalk-like virus particles with red blood cells show that viral particles agglutinate blood cells of type O and not type B. Similar protective effects of type B blood group have been described by Rockx et al (2005) where it was found that individuals with blood group B were protected against GI and not GII Norovirus. Contrary to this Guyader et al (2010) and Halpan et al (2005) have described in separate studies that genetic susceptibility does not differ with different blood groups. While different norovirus strains have been shown to display distinct HBGA binding properties, collectively they can infect nearly all individuals due to their high genetic variability (Karst 2010), a feature that highlights its highly adaptive nature (Le Pendu et al 2006).
In addition to the genetic diversity exhibited by norovirus, the inability of human infections to initiate a long lasting immunity, has contributed to its ability to cause global epidemics. Approximately half of the people exposed to the virus develop a short term homologous immunity that last 4-6 months to 2-3 years (Dolin 2007). Early volunteer challenge studies have shown that susceptible individuals were protected from disease when re-challenged soon after secondary exposure (<6months) in contrast to a more distant re-challenge (>6 months) (Parrino et al 1977, Johnson et al 1990), suggesting that short term immunity is elicited but not long term immunity.
Acquired immunity usually does not last until a subsequent norovirus season because of the antigenic shifts and drifts exhibited by the virus. Such drifts and shifts have been likened to those exhibited by the influenza virus, and have recently been described to be occurring more frequently with new norovirus variants being identified in the United Kingdom ( HPA 2008).
Norovirus are highly transmissible and can spread through various media (Karst et al 2010), and factors such as a low infectious dose of only 10-100 viral particles needed to cause clinical infection (Kaplan et al 1979), high resistance to temperatures from 0 to 60 degrees and ability to survive in food and water for 10days and 2 months respectively (D'Souza et al 2006), it can easily be transmitted for days after initial contamination of the environment. Due to these reasons, when outbreaks of norovirus gastroenteritis do occur in hospitals they can be quite difficult to control, This coupled with its high prevalence in the community (Lopman et al 2003) It is difficult to prevent the introduction of the virus into a hospital setting and when introduced the control of the resulting infection and outbreak can be quite difficult to control. Understanding the various ways in which norovirus is transmitted and the different settings they are more implicated in, is necessary for adequate control of a nosocomial outbreak. Due to its extreme stability in the environment, its highly contagious nature and resistance to disinfectants, the virus is classified as a Category B agent (Karst et al 2010).
Similar to other gastrointestinal pathogens norovirus are also transmitted via the faecal -oral route. Human volunteer studies in the 1950's established this as a major route for its transmission (Caul 1995), and since then it has been considered as being the most common method by which the virus is transmitted (Owen et al 1994, Frankhauser et al 2002) accounting for up to 85% of nosocomial outbreaks (Mattner et al 2005). In healthcare environments such faecal-oral transmission has been associated with environmental persistence of the virus and thereby infecting patients and hospital staff via contaminated fomites.
Faecal carriage of the virus by asymptomatic patients or hospital staff in the hospital has also been shown to play a major role in transmission of the virus in these settings (Gallimore et al 2004). Experiments carried out on the feline calicivirus, a closely related virus to norovirus have shown the routine antiseptics and hand hygiene products (Lages et al 2008) or use of cloths to clean surfaces (Gould 2009) do not destroy the virus. A human challenge study carried out by Barker et al (2004) showed that the virus could be consistently transferred via contaminated fingers to surfaces such as toilet tops door handles and telephone receivers. Widespread environmental contamination of hospital rooms of ill patient has been described (Weber et al 2010) with toilet tops being the most frequently contaminated site Furthermore the virus has been shown to survive for up to 12 days hospital environments (Malik et al 2006) and infect carpet fitters up to 17 days after an outbreak in a hospital (Cheesebrough et al 1997). Studies have shown that the viral RNA can be detected on environmental surfaces such as sinks, commodes, carpets and door handles for several days after contamination (Green et al 2008, Green et al 1999, Liu et al 2003, Cheesebrough et al 1997). This promotes faecal- oral and person to person transmission which have become the most common methods of secondary transmission of norovirus in healthcare settings such as nursing homes ( Calderon- Margalit et al 2004) , hospitals ( Khanna et al 2003, Leers et al 1987) and long term care facilities ( Cooper et al 2005, Khanna et al 2003, Said et al 2008).
Secondly, vomiting has also been recognised as an important means of norovirus transmission in heath care settings. It has been shown that a millilitre of vomitus can contain up to a thousand viral particles (Kaplan et al 1979). In a study by Caul (1994) it was estimated that a bout of vomitus by a patient liberated up to 30 million viral particles. Apart from causing outbreaks in hospitals and elderly care units (Chadwick and McCann 1994), transmission of the virus through vomitus has also been described in outbreaks of gastroenteritis in Aeroplanes (Caul 1994 ) , Hotels (Marks and Vipond 2000) and theatres ( Mark et al 2003)
Thirdly airborne transmission of the virus during outbreaks in health care settings has been suggested. Inhalation or ingestion of norovirus particles aerosolized of viral particles has been suggested to occur during vomiting virus during an episode of projectile vomiting which promotes secondary spread of the infection through contamination of the surrounding environment or food (Gilbert et al 1990). Spread by this route is said to be more likely when the vomiting occurs in a confined space such as a hospital ward or bay in the ward (Chadwick and Mc Cann 1994). Such transmission by aerosalization has been reported in an outbreak in an Accident and Emergency (A&E) of a hospital in Toronto (Sawyer et al 1988), where it was found that the risk of becoming ill was four times greater for people who walked through the A&E than for those who did not. Similar findings were found by Chadwick and McCann (1994) when investigating an outbreak in a ward in a long term care facility. Airborne transmission has also been documented in cruise ships (Ho et al 1989), aeroplanes (Widdowson et al 2005) and Hotels (Marks and Vipond 2003)
In a study by Keshwich et al (1982) norovirus was shown to be resistant to chlorination a feature that maybe explain its ability to be transmitted by through this medium. Water borne outbreaks are more common in the community or related to recreational water use. Such recreational water outbreaks were seen in a swimming pool (Podiwels et al 2009), where people who visited the pool over a 3 days period developed diarrhoea and vomiting and school children playing in recreational fountain (Hoebe et al 2004). There has been one reported outbreak in a hospital reported by Schvoerer et al (1999) where the virus was believed to have come from contaminated tap water.
Food borne outbreaks of norovirus gastroenteritis are most common outside the hospital setting. In Europe it has been described as an important food borne pathogen in the community (Vasickova et al 2005) and the Centre for disease Control (CDC) Atlanta estimates that up to half of food borne gastroenteritis are due to norovirus (Dolin et al 2007). In another study in the U.S.A it was estimated that it is responsible for up to 93% of food related outbreaks of gastroenteritis (Frankhauser et al 2002). Any food item can be contaminated with norovirus through airborne or faecal-oral transmission.
3.3 CLINICAL FEATURES
Norovirus circulates freely in the population and infections are observed in all age groups (Glass et al 1). Once infected, the incubation period is usually dose dependent (Caul 1994) and usually lies in the range of 12-4 hours, but can last as long as 77 hours (Kaplan et al 1982). In immuno-competent individuals the resulting infections is self limiting and mild with resolution occurring within 12-72 hours (Estes et al 2006). Asymptomatic infections do occur and it is estimated that up to a third of people exposed to the virus do not develop symptoms (Graham et al 1994).The most common symptoms of norovirus gastroenteritis are diarrhoea, vomiting, nausea, low grade fever, abdominal cramps, headaches, chills and myalgia (Kaplan et al 1982). Vomiting has been reported as a more prominent symptom in children, unlike in adults who experience more of diarrhoea (Chang et al 2006, Kaplan et al 1982.) The virus is non invasive, thereby the diarrhoeal stools are usually non bloody. Diarrhoea is usually explosive and the vomiting which is projectile in nature is pathognomic of norovirus infection and in the absence of other causes is sufficient of a presumptive diagnosis. However protracted length of illness has been described in children and the elderly with symptoms being more protracted and severe (Koopmans 2009). Community based studies in the Netherlands (Rockx et al 2002) have shown that up to a quarter of cases develop symptoms that last more than a week. Similar extended duration of symptoms have been seen in an outbreak in a paediatric ward (Chang et al 2006) and a teaching hospital ( Mattner et al 2006) where symptoms lasted up to a week in some patients. Furthermore symptoms have been seen to be more severe in patients with underlying disease (Lopman et al 2004, Seiberger et al 2008, Mattner et al 2006). Symptoms lasting up to eight weeks have been reported in children (Murata et al 2007, Rockx et al 2002), while in transplant and immunosuppressed they have been reported to have lasted for as long as two years(Gallimore et al 2004, Morotti et al 2004).
Despite the short duration of illness patients can continue to shed the norovirus for extended periods after the resolution of symptoms. This gives an opportunity for the transmission of the virus to other people and thereby initiating new outbreaks. In experimental studies carried out by Atmar et al (2005) on 16 people inoculated with norovirus , it was found that virus shedding in some of them continued for a duration of eight weeks .Other volunteer studies have described post symptomatic viral shedding in healthy adults for to three weeks (Rockx et al 2002). In other studies viral excretion has been seen to occur for up to 47 days in children and six weeks in infants (Mutaro et al 2007) and for 45 days in the elderly (Tu et al 2008) 140 days in a paediatric oncology unit (Simon et al 2006) and two years in an immuno-suppressed patient (Kaufman et al 2008).
A misconception about norovirus is that apart from causing an acute gastroenteritis results to no serious complications. More common complications seen as a result of the diarrhoea and vomiting are severe dehydration and renal impairment usually affecting elderly patients (Gould 2006), which can sometimes be severe and resulting in death (Lopman et al 2003, Lopman et al 2004). However with the improvement in diagnostics, the virus is now being identified in other clinical outcomes other that gastroenteritis. However in children there have been reports of more serious but rare complications of norovirus infection. Such include norovirus associated encephalopathy (Obinata et al 2010, Shuichi et al 2006), benign infantile seizures(Chen et al 2009), irritable bowel syndrome (Marshal et al 2007), acute renal failure(Kanai et al 2010 ), Necrotizing enterocolitis (Turcois-Ruiz et al 2008) and pneumatosis intestinalis (Chen et al 2009). In the United Kingdom there has been one reported case of spontaneous bowel perforation in an 83 year old woman secondary to norovirus acute gastroenteritis (Pawa et al 2009).
Norovirus outbreaks occur mostly in semi enclosed communities, such as nursing homes(Lopman et al 2004) ,hospitals (Chadwick et al 2000), cruise ships (Widdowson et al 2002), military facilities ( McCarthy et al 2000, Thornton et al 2005). The virus can be introduced into the health setting through various routes. Introduction of norovirus into long term care facilities by employees has been described (Wu et al 2005, Milazzo et al 2002, Rodriguez et al 1996). However more common routes of introduction are usually a patient being admitted in the hospital (Khanna et al 2003).
Preventing introduction of norovirus into health care settings such as hospitals and thereby causing outbreaks can be very challenging. This is because of its high prevalence in the community and also because of the asymptomatic viral shedding exhibited by people. Lapses in hospital hygiene, increasing age of the population number of susceptible individuals in the community have been proposed as reasons for the increased incidence of norovirus outbreaks seen in hospitals nowadays (Gould 2008). Other institutional factors such as greater number of beds in a ward, being a geriatric or medical ward and short duration of stay of a patient have been proposed to increase the risk of norovirus gastroenteritis occurring in a hospital ward (Vardy et al 2007).
4.1 OUTBREAK DEFINITION
The safest definition of an outbreak if viral diarrhoea in a hospital ward is when the observed incidence in the ward is greater than the expected incidence provided all other causes of diarrhoea have been excluded. In the case of norovirus there are set definitions put forward by the Health protection Agency (HPA 2010), that are used in all acute trusts in the United Kingdom to define norovirus outbreaks.
i.) Suspected case:
-Diarrhoea: two or more loose stools in a 24 hour period
-Vomiting:- two or more episodes of vomiting of suspected infectious cause occurring in a 24 hour period
-Diarrhoea and vomiting - one or more episodes of both occurring in a 24 hour period.
To be linked to Norovirus infection these symptoms must not be associated with prescribed drugs or treatments, not be due to a reaction to an anaesthetic or due to an underlying medical condition.
ii.) Suspected outbreak: A suspected outbreak occurs when two or more cases fitting the above definition occur in a open ward caring for patients in a hospital, without lab confirmation.
iii.)Confirmed outbreak: When there is laboratory confirmation of the causative agent being norovirus then a suspected outbreak becomes a confirmed outbreak.
Rapid recognition of an outbreak is essential if rapid control is desired. This can usually be done through enhanced awareness of the clinical characteristics of the disease. Kaplan et al (1982) established a clinical and epidemiological criterion to define norovirus outbreaks in settings where microbiological confirmation was not possible. These criteria called Kaplan's criteria were based on an in depth analysis of 38 Norwalk virus acute gastroenteritis outbreaks occurring between1976-80 and was used to distinguish norovirus outbreaks from other enteric pathogens.
These criteria are
A negative stool culture for bacteria
Greater than 50% of cases with vomiting
Mean /median duration of illness lasting 12-60 hours
Mean /median incubation period of 24-48 hours( If available)
Subsequent studies also showed that this set of criteria is highly specific ( 99%) and sensitive(68%) (Turcois et al 2006). The applicability of these criteria to define outbreaks in hospitals has been questioned, since it was created based on community and not hospital outbreaks (Said et al 2008, Lopman et al 2004). Furthermore its relative insensitivity; norovirus cannot be excluded as the etiological agent if outbreak characteristics fail to meet the criteria.
Nevertheless this criterion has been used to define norovirus outbreaks in hospitals (Billgreen et al 2002, Lopman et al 2004) and in long-term care facilities (Navarro et al 2005). Other clinical criteria proposed by Osterhelm (1993) have been fever to vomiting ratio, to differentiate norovirus from other bacterial outbreaks and the diarrhoea to vomiting ratio by Dalton et al (1999) used to distinguish Enterotoxigenic Escherichia coli from norovirus outbreaks. These criteria have been found to have sensitivities of 90.1% and 96.6%, and specificities of 46.6% and 44.5% respectively ( MacCannal et al 2010).The applicability of these other criteria in clinical settings have however not been documented.
Chadwick et al (2000) noted that there are three aspects applicable to the control of norovirus infection transmission in hospitals : the introduction of the virus into the hospital and other health care settings; containment of infection at ward level; and measures to prevent spread to other wards. However due to factors such as the high community prevalence of norovirus, extended periods of asymptomatic viral shedding by individuals, low infectious dose and lack of any recognised infection prodrome, controlling introduction of norovirus into health care settings is very difficult. Therefore control measures are more focused on minimizing the spread of the virus within and between hospital units, with the aim of shortening the duration of an outbreak.
Chadwick et al (2000) attempted a detailed review of the evidence base for the cardinal issues in the control of outbreaks of norovirus infections in health care settings. These guidelines which were last published in 2000, put forward recommendations that were based into categories used by the combined working party of the British society of Antimicrobial Chemotherapy, the Hospital infection society and the Infection control Nurses association in the Guidelines for the control of Methicillin resistant staphylococcus aureus(MRSA) ( Duckworth et al 1998). These categories which were based on the strength of evidence for each of the recommended control measures are as follows:
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.
Currently in the United Kingdom most of the infection control policies for the management and control of Small Round Cell Structured Viruses (SRSV) in acute trusts and other health care settings are based on these guidelines.
The recommended control measures put forward by Chadwick et al (2000) were as follows
A.) Category I
- Wash hands with soap and water after contact with an affected patient or environment, after removing gloves and apron. (Category I)
B.) Category II
- Cohort nurse and isolate symptomatic patients. (Category II)
- Wear gloves and apron for contact with an affected patient or environment. (Category II)
- Exclude affected staff from ward immediately and until 48hours symptom -free. ( Category II)
- Close ward to prevent the introduction of new susceptible. 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 on 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 reopened until 72hours after the last new case and 72 hours after uncontained vomiting and diarrhoea. (Category II)
- Thoroughly clean the ward and change the bed curtains before re-opening (Category II)
C.) No Category
-Remove exposed food such as fruit. (No category)
-Consider use of antiemetics for patients with vomiting. (No category)
-Clean carpets and soft furnishings with hot water and detergent, or steam clean. Vacuum cleaning is not recommended. (No category).
Preventing spread of the virus through person to person contact and contaminated environmental surfaces and beddings is of great significance in halting hospital outbreaks. The CDC endorses standard enteric precautions with an emphasis on adequate hand hygiene. There have been various studies that demonstrate the role and effectiveness of hand and environmental hygiene in preventing and controlling norovirus in hospitals (Malik et al 2006, Weber et al 2010, Seigel et al 2007, Johnston et al 2007). Contaminated of hospital beds (Creamer and Humphreys 2008), pillows (Fretz et al 2005), carpets ( Cheesebrough et al 2000), door knobs and commodes ( Lopman et 2003) have all been shown to initiate outbreaks in different settings. Widespread environmental contamination has been shown to extend duration of outbreaks in health care settings (Greig and Lee 2009). Overcrowding in an emergency department with the build up of soiled bedpans and dirty linens was implicated in an outbreak in a hospital in Toronto (Sawyer et al 1988). In another study involving outbreaks on cruise ships, in which ships were scored based on the thorough disinfection of restrooms, as a measure of environmental hygiene, ships with higher scores were less likely to have subsequent norovirus outbreaks than cruise ships with poorer scores (Ho et al 1999, Carling et al 2010). Other studies have demonstrated the effectiveness of environmental cleaning as part of comprehensive control measures to reduce enteric infections (Amstrong-Evans et al 1999, Byers et al 2001, Markis et al 2000). A terminal deep clean of hospital wards should ideally be carried out at least 72 hours after the resolution of symptoms in the last case (Chadwick et al 2000 Barker et al 2004). This takes into account the period of maximum infectivity (48hrs) and the incubation period (24hrs) of the virus. Cleaning with simple detergents (Marks et al 2003) and use of disinfectants containing phenolic compounds have been found to be ineffective on norovirus (Greig and Lee 2009,Gould 2008) and it is thus recommended that hypochlorite solutions at 1000ppm be used (Barker et al). Using hypochlorite solutions in this concentration rather than 500ppm was found to be more effective in controlling an outbreak in a paediatric ward (Frankie et al 2006). Failure to carry out a deep clean may result in an outbreak restarting as soon as patients are readmitted (Dancer 2008)
Hand hygiene however is considered more critical to pathogen control than cleaning and disinfection of environmental surfaces (Dettenkofer and Spensor 2007). In the guidelines by Chadwick et al (2000) and the CDC it is the only recommendation that is based on evidence strongly supported by experimental epidemiological studies. A review related to infection control has shown that hand washing decreases diarrhoeal episodes by 30% ( Ejemot et al 2008). The use of alcohol based hand sanitizers in one hospital outbreak (Belliot et al 2008) and wipes to disinfect surfaces in a school (Sandora et al) were found to be useful in reducing contamination and spread. However the effectiveness alcohol based hand sanitizers as norovirus disinfectants have been questioned (Gehrke et al 2004) and recent studies have shown that they are not as effective against norovirus as basic hand hygiene with soap and water (Liu et al 2010) . As a result it been recommended that hand washing with soap and water should be carried out between patient visits during hospital outbreaks of norovirus gastroenteritis (Chadwick et al 2000). Hand washing for at least a minute has been suggested to be more effective in removing norovirus than hand washing for 10 to 20 seconds (Liu et al 2010, Barker et al 2010).
These control measures may not be sufficient to halt an ongoing outbreak and closure of a hospital unit such as a ward, or even a whole hospital may sometimes be necessary to stop replenishment of susceptible individuals. The spread of norovirus between wards during an outbreak can mean the difference between a small contained problem and a larger hospital wide problem. A ward closure can be due to staff shortages as experienced an outbreak in England in 1994 which closed a 220 bed hospital (Chadwick et al 1994) or may be an integral component of the control measures implemented to prevent spread of a norovirus infection.
The rationale for closures and cohort nursing is basically to prevent patient and ward staff from infected wards transmitting the virus to other units in the hospital. However the
recommended use of ward closures as a control measure is placed under category II in the guidelines by Chadwick et al (2000). Similarly ward closures were also placed in category II in recently published guidelines by the Centre of Disease Control (CDC) Atlanta (MacCannal et al 2010). In the CDC guidelines, control measures placed in this category are based on a 'weak recommendation supported by any quality evidence'. It has been recognised that there is limited data on measures such as patient cohorting and ward closures in controlling and reducing the duration of outbreaks caused by norovirus (Harris et al 2010). Closing wards within the first 4 days of onset of a norovirus outbreak has been shown to reduce the duration of the outbreak by about 7 days (Lopman et al 2004). Different studies on hospital outbreaks (Sawyer et al 1988, Russo et al 1997) have shown that spread of the norovirus infection through the hospitals was attributed to staff movements between infected and uninfected wards (Sawyer et al 1988, Russo et al 1997) which in one case led to closure of the 600 bed tertiary referral hospital (Sawyer et al 1988). Furthermore ward closure and cohort nursing have been used successfully in ending hospital outbreaks have been described (Cunney et al 2000, Johnston et al 2007). On the contrary in a outbreak in a 600 bed tertiary care facility it was found that instituting ward closure early in the outbreak did not prevent rapid spread of the virus across the hospital and thus outbreak duration ( Leers et al 1987). Augustin et al (1995) compared two outbreaks occurring in two health care centers. One center applied hygiene based control measures and the other center included additional measures such unit closures to new admission. The authors concluded that applying the additional measures had no impact on disease transmission. Similarly Harris et al (2010) found in a systematic review of 47 outbreaks that there was no difference in duration of outbreaks between health settings that did and did not use infection control measures. However it should be noted that ward closures were not among any of the control measures indentified in these 47 outbreaks.
In a report by the Department of Health (NAO 2004) it was identified that acute trusts in England and Wales operate at a high bed occupancy rate and are usually under pressure to meet performance targets. Therefore closure of an entire unit in an acute trust due to a norovirus outbreak can be a very expensive venture. Norovirus outbreaks in the United Kingdom have been estimated to cost about a million dollars per 1000 hospital beds(Lopman et al 2004) while In an outbreak in John Hopkins hospital it cost an estimated $650,000 ( Johnston et al 2007). The pressures of having to meet trust performance targets sometimes leads to the adoption of practices that may not be consistent with good infection control. Thus some trusts may adopt closures of bays within wards instead of whole ward closures as an initial response to norovirus outbreaks. This helps address the problem of bed pressures, especially in wards with high patient turnover, but whether bay closures are effective in shortening outbreak duration is not known.