Skip directly to search Skip directly to A to Z list Skip directly to navigation Skip directly to site content Skip directly to page options
CDC Home

Guideline for the Prevention and Control of Norovirus Gastroenteritis Outbreaks in Healthcare Settings, 2011

VIII. Evidence Review

Question 1: What host, viral or environmental characteristics increase or decrease the risk of norovirus infection in healthcare settings?

To answer this question, the quality of evidence was evaluated among risk factors identified in 57 studies.  In areas for which the outcome of symptomatic norovirus infection was available, this was considered the critical outcome in decision-making.  The evidence for this question consisted of one systematic review,[56 51] observational,[57-62,62-64,64-77,77-107] and 4 descriptive studies,[108-111] as well as one basic science study.[112]  The paucity of randomized controlled trials (RCT) and the large number of observational studies greatly influenced the quality of evidence supporting the conclusions in the evidence review.  Based on the available evidence, the risk factors were categorized as host, viral or environmental characteristics. Host characteristics were further categorized into demographics, clinical characteristics, and laboratory characteristics.  Environmental characteristics were further categorized into institution, pets, diet, and exposure.  The findings of the evidence review and the grades for all clinically relevant outcomes are shown in Evidence and Grade Table 1.


Q1.A Person characteristics

Q1.A.1 Demographic characteristics

Low-quality evidence was available to support age as a risk factor for norovirus infection,[57-60,62-64] and very low-quality evidence to support black race as a protective factor.[64]  Three studies indicated that persons over the age of 65 may be at greater risk than younger patients for prolonged duration and recovery from diarrhea in healthcare settings.[57-59]  Studies including children under the age of five showed an increased risk of household transmission as well as asymptomatic infection compared with older children and adults.[60,62] 

A single but large-scale observational study among military personnel found blacks to be at lower risk of infection than whites.[64]  Very low-quality evidence failed to demonstrate meaningful differences in the risk of infection corresponding to strata on the basis of educational background (in the community setting).[61]  Based upon very low-quality evidence, outbreaks originating from patients were more likely to affect a large proportion of patients than were outbreaks originating from staff.[56]  Exposure to vomitus and patients with diarrhea increased the likelihood that long-term care facility staff would develop norovirus infection.[66] 

The search did not identify studies that established a clear association between sex and symptomatic norovirus infection or complications of norovirus infection.[57,59, 79, 98]  Low-quality evidence from one prospective controlled trial did not identify sex as a significant predictor of symptomatic norovirus in univariate analyses.[57] There is low-quality evidence suggesting that sex is not a risk factor for protracted illness or complications of norovirus infection including acute renal failure and hypokalemia.[57]

Q1.A.2 Clinical characteristics

Review of the available studies revealed very low-quality evidence identifying clinical characteristics as risk factors for norovirus infection.[57,60,65,68]  One small study found hospitalized children with human immunodeficiency virus (HIV) and chronic diarrhea were more likely to have symptomatic infection with small round structured virus (SRSV) than those without HIV and affected with chronic diarrhea.[65,68]  Adult patients with symptomatic norovirus receiving immunosuppressive therapy or admitted with underlying trauma were at risk for a greater than 10% rise in their serum creatinine.[57]  Norovirus-infected patients with cardiovascular disease or having had a renal transplant were at greater risk for a decrease in their potassium levels by greater than 20%.[57]  Observational, univariate study data also supported an increased duration of diarrhea (longer than two days) among hospitalized patients of advanced age and those with malignancies.[57]  This search did not reveal data on the risk of norovirus acquisition among those co-infected with other acute gastrointestinal infections, such as C. difficile

Q1.A.3 Laboratory characteristics

Q1.A.3.a Antibody levels

There was very low-quality evidence to support limited protective effects of serum antibody levels against subsequent norovirus infection.74-76  In two challenge studies, adult and pediatric subjects with prior exposure to norovirus showed higher antibody titers than found in previously unexposed subjects after initial infection and after challenge.[74,76]  The detection of preexisting serum antibody does not appear to correlate with protection against subsequent norovirus challenge, nor did increasing detectable pre-existing antibody titres correlate with attenuations in the clinical severity of disease.[74,75]  In one study, symptoms such as vomiting, nausea, headaches, and arthralgia were correlated with increasing antibody titres.[74]  In a serial challenge study, 50% of participants (n=6) developed infection, and upon subsequent challenge 27-42 months later, only those same participants developed symptoms.  A third challenge 4-8 weeks after the second series resulted in symptoms in just a single volunteer.[76]  Pre-existing antibody may offer protection to susceptible persons only for a limited window of time, on the order of a few weeks. The search strategy did not reveal data  on the persistence of immunity to norovirus nor elevations in antibody titers that were consistently suggestive of immunity. 

Q1.A.3.b Secretor genotype

Review of the outlined studies demonstrated high-quality evidence to support the protective effects of human host non-secretor genotypes against norovirus infection.[70-72,113]  Two observational studies and one intervention study examined volunteers with and without the expression of the secretor (FUT2) genotype after norovirus challenge.[70-72]  Statistically significant differences were reported with secretor-negative persons demonstrating a greater likelihood of protection against, or innate resistance to symptomatic and asymptomatic norovirus infection than seen in persons with secretor-positive genotypes.  This search did not reveal data on the dose-response effects of norovirus in persons with homozygous and heterozygous secretor genotypes.  Because the FUT2-mediated secretor positive phenotype appears to confer susceptibility to subsequent norovirus infection following challenge, there is an association between this phenotype and measurable circulating antibody (suggesting prior infection) in the population.  One study estimated that 80% of the population is secretor-positive (or susceptible to norovirus) and 20% is secretor-negative (resistant to norovirus challenge independent of inoculum dose).  Among susceptible persons, approximately 35% are protected from infection.  This protection is potentially linked to a memory-mediated rapid mucosal IgA response to norovirus exposure that is not seen in the other 45% of susceptibles, who demonstrate delayed mucosal IgA and serum IgG responses.[72]  Although elevated antibody levels following infection appear to confer some protective immunity to subsequent challenge, paradoxically, measurable antibody titers in the population may be a marker of increased susceptibility to norovirus because of the association between such antibodies and FUT2-positive status.      

Q1.A.3.c ABO phenotype

There was low-quality evidence suggesting any association of ABO blood type with the risk of norovirus infection.[69,72,73,77,78,114,115]  An RCT  suggested that persons with histo-blood group type O was associated with an  increased risk of symptomatic or asymptomatic norovirus infection among secretor-positive patients.72  Binding of norovirus to the mucosal epithelium may be facilitated by ligands associated with type-O blood.  The other blood types—A, B, and AB—were not associated with norovirus infection after controlling for secretor status.  Three studies showed no protective effect of any of the blood types against norovirus.[69,77,78]  The search strategy did not reveal prospective cohort data to correlate the role of ABO blood types with risk of norovirus infection. 


Q1.B Viral characteristics

There was very low-quality evidence to suggest an association of virus characteristics with norovirus infection.[57,108-110]  Very low-quality descriptive evidence suggested that increases in overall norovirus activity may result from the emergence of new variants among circulating norovirus strains, and strains may differ in pathogenicity, particularly among GII.3 and GII.4 variants.[108-110]  In recent years, GII.4 strains are increasingly reported in the context of healthcare-associated outbreaks, but further epidemiologic and laboratory studies are required to expand on this body of information.  This search did not identify studies examining genotypic characteristics of viruses associated with healthcare-acquired norovirus infection.


Q1.C Environmental characteristics

Q1.C.1 Institutional characteristics

Very low-quality evidence was available to support the association of institutional characteristics with symptomatic norovirus infection.[82,99]  Among two observational studies, the number of beds within a ward, nurse understaffing, admission to an acute care hospital (compared to smaller community-based facilities), and having experienced a prior outbreak of norovirus gastroenteritis within the past 30 days were all possible risk factors for new infections.[82,99]  These increased institutional risks were identified from univariate analyses in pediatric and adult hospital populations.  There were statistically significant, increased risks of infection among those admitted to geriatric, mental health, orthopedic, and general medicine wards.  The review process did not reveal data on the comparative risks of infection among those admitted to private and shared patient rooms. 

Q1.C.2 Pets

Review of the outlined studies demonstrated very low-quality evidence to support exposure to pets (e.g., cats and dogs) as a risk factor for norovirus infection.[61]  One case-control study examined pet exposure among households in the community and concluded that the effect of cats was negligible.[61]  The single study did not demonstrate any evidence of transmission between pets and humans of norovirus infection.  This search strategy did not reveal studies that evaluated the impact of therapy pets in healthcare settings during outbreaks of norovirus gastroenteritis or data examining domestic animals as reservoirs for human infection. 

Q1.C.3 Diet

There was low-quality evidence to suggest that extrinsically contaminated food items are commonly implicated as vehicles of norovirus exposure in healthcare settings.[61,77,80,84,86,87,89-97,100-102,104-107,111]  Nineteen observational studies itemized statistically significant food sources implicated in community outbreaks.[80,81,84,86,87,89-97,100,101,104-106]  Common to most of these food sources was a symptomatic or asymptomatic food-handler.  Sauces, sandwiches, fruits and vegetables, salads, and other moisture-containing foods were most often cited as extrinsically contaminated sources of outbreaks of norovirus gastroenteritis.  Importantly, these data reflected the breadth of foods that can become contaminated. Tap water and ice were also associated with norovirus contamination during an outbreak with an ill food-handler.  This literature review did not identify studies that examined the introduction of intrinsically contaminated produce or meats as a nidus for norovirus infection and dissemination within healthcare facilities. 

Q1.C.4 Proximity to infected persons

This review demonstrated high-quality evidence to suggest that proximity to infected persons with norovirus is associated with increased risk of symptomatic infection.[61,62,64,79,83,88,98,103,111]  Eight observational studies found statistically significant factors such as proximate exposure to an infected source within households or in crowded quarters increased infection risk, as did exposures to any or frequent vomiting episodes [61,62,64,79,83,88,98,103].   These data suggest person-to-person transmission is dependent on close or direct contact as well as short-range aerosol exposures.  One observational study established a linear relationship between a point source exposure and attack rate based on proximity to an infected and vomiting source.[88]  This search process did not identify studies that quantified the spatial radius necessary for transmission to successfully occur.


Q1 Recommendations

1.A.1 Avoid exposure to vomitus or diarrhea.  Place patients on Contact Precautions in a single occupancy room if they have symptoms consistent with norovirus gastroenteritis. (Category IB)(Key Question 1A)

1.A.2.a Consider longer periods of isolation or cohorting precautions for complex medical patients (e.g., those with cardiovascular, autoimmune, immunosuppressive, or renal disorders) as they can experience protracted episodes of diarrhea and prolonged viral shedding.  Patients with these or other comorbidities have the potential to relapse and facilities may choose longer periods of isolation based on clinical judgment. (Category II) (Key Question 1A)

1.C.1  Consider the development and adoption of facility policies to enable rapid clinical and virological confirmation of suspected cases of symptomatic norovirus infection while implementing prompt control measures to reduce the magnitude of a potential norovirus outbreak. (Category II) (Key Question 1C)

1.C.3.a  To prevent food-related outbreaks of norovirus gastroenteritis in healthcare settings, food handlers must perform hand hygiene prior to contact with or the preparation of food items and beverages. For more information visit FDA Food Code website. (Category IC) (Key Question 1C)

1.C.3.b Personnel who work with, prepare or distribute food must be excluded from duty if they develop symptoms of acute gastroenteritis.  Personnel should not return to these activities until a minimum of 48 hours after the resolution of symptoms or longer as required by local health regulations. For more information visit FDA Food Code website. (Category IC) (Key Question 1C)

1.C.4 If norovirus infection is suspected, adherence to PPE use according to Contact and Standard Precautions is recommended for individuals entering the patient care area (i.e., gowns and gloves upon entry) to reduce the likelihood of exposure to infectious vomitus or fecal material. (Category IB)(Key Question 1C)

Contact Us:
  • Centers for Disease Control and Prevention
    1600 Clifton Rd
    Atlanta, GA 30333
  • 800-CDC-INFO
    TTY: (888) 232-6348
  • Contact CDC–INFO The U.S. Government's Official Web PortalDepartment of Health and Human Services
Centers for Disease Control and Prevention   1600 Clifton Rd. Atlanta, GA 30333, USA
800-CDC-INFO (800-232-4636) TTY: (888) 232-6348 - Contact CDC–INFO
A-Z Index
  1. A
  2. B
  3. C
  4. D
  5. E
  6. F
  7. G
  8. H
  9. I
  10. J
  11. K
  12. L
  13. M
  14. N
  15. O
  16. P
  17. Q
  18. R
  19. S
  20. T
  21. U
  22. V
  23. W
  24. X
  25. Y
  26. Z
  27. #