Norovirus Transmission: How the Virus Spreads

According to the CDC, you get norovirus "by accidentally getting tiny particles of feces (poop) or vomit in your mouth from a person infected with norovirus." Every route below is a variation on that one mechanism — the virus leaves one person in stool or vomit and reaches another person's mouth, usually indirectly.

How Infectious Is Norovirus, Really?

The figure most often quoted — including on this site's homepage — is that as few as 18 viral particles can cause infection. It is worth knowing where that number comes from, because the CDC itself no longer uses it; its transmission page now says only that "it only takes a few norovirus particles to make you and other people sick." The 18 is a modeling result, and a later experiment produced a much larger number.

Two human-challenge studies, two very different answers

• Teunis et al., 2008 (Journal of Medical Virology, DOI 10.1002/jmv.21237) re-analyzed volunteer studies in which healthy adults swallowed measured doses of Norwalk virus (the prototype norovirus, genotype GI.1). Their dose–response model estimated that a single virus particle has close to a 50% probability of establishing infection — "exceeding that reported for any other virus studied to date." The widely cited median infectious dose of roughly 18 particles is derived from that model. Infected volunteers did not always get sick: the probability of illness rose with dose, from about 10% at 1,000 genome copies to about 70% at 100 million.
• Atmar et al., 2014 (Journal of Infectious Diseases, DOI 10.1093/infdis/jit620) ran a new randomized, placebo-controlled challenge in 57 adults and measured the 50% human infectious dose directly: about 1,320 genome equivalents for secretor-positive people with blood group O or A, and about 2,800 for all secretor-positive people. The authors describe this as "higher than previous estimates and similar to that of other RNA viruses." Of those infected, 67% developed gastroenteritis.

So the honest range for "how many particles does it take" spans from roughly 18 to roughly 1,300 — a gap of almost two orders of magnitude between the two best-known studies. The two research groups exchanged letters about it: Kirby, Teunis and Moe (J Infect Dis 2015, DOI 10.1093/infdis/jiu385) argued that both studies confirm "high infectivity" once viral clumping (aggregation) and host susceptibility are accounted for — much of the difference is in how you count the particles and who you count as susceptible. Either number is small. A gram of stool from an infected person at peak shedding carries billions of genome copies (see below), so the practical conclusion is the same under both estimates: a quantity you cannot see is enough.

Who Cannot Catch It: Genetic Resistance

The CDC's About Norovirus page notes that "your likelihood of getting a norovirus infection is also determined in part by your genes." The mechanism was identified in a 2003 challenge study by Lindesmith and colleagues (Nature Medicine, DOI 10.1038/nm860). About 29% of their study population carried two non-functional copies of the FUT2 gene — so-called "non-secretors" — and did not produce the sugar molecule (the H type-1 histo-blood group antigen) that Norwalk virus binds to in the gut. None of these non-secretors became infected after challenge, "regardless of dose." Among people with a working FUT2 gene, some were also resistant, which the authors attributed to prior immunity or an unidentified factor.

Two caveats that the headline finding usually loses: the study used Norwalk virus (GI.1), and resistance to one norovirus genotype is not the same as resistance to all of them — the CDC is explicit that "infection with one type of norovirus may not protect you against other types." The Atmar 2014 study above excluded secretor-negative volunteers from its dose calculation for exactly this reason, and all 21 of the people it infected were blood group O or A.

Airborne Transmission: Vomit as an Aerosol

Norovirus is not a respiratory virus, but it can travel through the air. The CDC's transmission page lists "tiny drops of vomit from a person with norovirus spray through the air, landing on surfaces or entering another person's mouth" as a route. The classic evidence is an outbreak investigated by Marks and colleagues (Epidemiology and Infection 2000, DOI 10.1017/S0950268899003805): after one diner vomited during a hotel meal, the attack rate at each table fell in step with its distance from that person, no food could be implicated, nobody in a separate restaurant got sick, and genetic sequencing found an identical strain in every case. The authors concluded the pattern was "consistent with airborne spread … with infection by inhalation with subsequent ingestion of virus particles."

The Atmar 2014 challenge study adds a number to that picture: Norwalk virus was detected in 56% of vomit samples at a median concentration of about 41,000 genome equivalents per millilitre. Set against the same study's 50% infectious dose of ~1,320 genome equivalents, one millilitre of vomit at that median concentration contains on the order of 30 infectious doses (41,000 ÷ 1,320 ≈ 31 — our arithmetic, from the paper's two figures). A single vomiting episode produces far more than one millilitre. This is why the CDC's cleanup guidance treats a vomiting incident as contaminating "the entire area," not just the visible spot — see Prevention.

How Long an Infected Person Sheds Virus

The CDC's guidance is that you are most contagious while symptomatic and "during the first few days after you feel better," but that "you can still spread norovirus for 2 weeks or more after you feel better." The study behind the "2 weeks or more" is Atmar et al., 2008 (Emerging Infectious Diseases, DOI 10.3201/eid1410.080117), which followed 16 experimentally infected volunteers and found:

• Shedding began before symptoms — virus was first detectable in stool 18 hours after inoculation, while symptoms typically begin 12–48 hours after exposure (CDC).
• Shedding lasted a median of 28 days after inoculation, with a range of 13 to 56 days — roughly four weeks, and in some volunteers eight.
• Peak shedding was enormous — a median of 95 billion genome copies per gram of feces (range 0.5 to 1,640 billion).
• Five of the 16 infected volunteers never developed gastroenteritis at all, yet were infected and shed virus — the experimental basis for "asymptomatic shedding."
• Symptomatic illness itself "lasted 1–2 days."

Put the peak figure against the infectious-dose figures above: 95 billion copies per gram divided by an infectious dose of ~1,320 is about 72 million 50%-infectious doses in a single gram (our arithmetic from the two papers' figures; using the Teunis estimate of ~18 the number is larger still). The authors were careful to note that detecting viral RNA is not the same as proving the virus is still infectious — "understanding of the relevance of prolonged fecal norovirus excretion must await the development of sensitive methods to measure virus infectivity." That caveat is why the CDC's advice is framed around 48 hours after symptoms stop for food handling and caregiving, not 28 days: the highest-risk window is the acute illness and the few days after it. Caring for someone during that window (fluids, rest, and when to seek care) is covered on the Treatment page.

Why Norovirus Spreads So Easily

Pulling the evidence together, five properties make norovirus unusually hard to contain:

• Extremely low infectious dose — somewhere between ~18 and ~1,300 particles depending on the study, either of which is invisible (Teunis 2008; Atmar 2014).
• Massive shedding, starting before symptoms — billions of copies per gram, detectable within 18 hours of infection (Atmar 2008).
• Environmental persistence — the CDC states that norovirus "can persist on surfaces for days or weeks and is resistant to many common disinfectants."
• Resistance to alcohol — per CDC, "hand sanitizer does not work well against norovirus" and "is not a substitute for handwashing." Norovirus has no lipid envelope for alcohol to disrupt. Washing with soap and water physically removes the virus; sanitizer may be used in addition to — not instead of — handwashing.
• Aerosolization during vomiting — vomit droplets carry tens of infectious doses per millilitre and can reach food, surfaces, and bystanders' mouths (CDC; Marks 2000; Atmar 2014).

High-Risk Settings

The CDC's Norovirus Outbreaks page describes the settings where reported US outbreaks concentrate. The table below records only what the CDC publishes — several widely repeated per-setting percentages, including ones this page used to carry, do not appear on any CDC page.

Foodborne Transmission in Detail

Norovirus "causes 58% of foodborne illnesses acquired in the United States," per the CDC's Facts and Stats page. Food becomes contaminated in two main ways:

1. Infected food handler. The CDC says infected workers "are frequently the source of outbreaks in food-service settings, often by touching ready-to-eat foods (such as raw fruits and vegetables) with their bare hands before serving them" — and that "any food served raw or handled after being cooked can get contaminated." The FDA Food Code's exclusion rules for diagnosed workers are set out on the Food Safety page.
2. Contamination at the source. Oysters and other filter-feeding shellfish concentrate virus from sewage-contaminated growing water, and produce can be contaminated by water sprayed in the field (CDC). The CDC's outbreak page lists current FDA shellfish advisories; in 2024–2026 they have involved oysters and clams from Washington State, British Columbia, Louisiana, and the Republic of Korea.

The foods the CDC names as "commonly involved in norovirus outbreaks" are leafy greens (such as lettuce), fresh fruits, and shellfish (such as oysters).

How Long Does Norovirus Survive on Surfaces?

The CDC's statement is that norovirus "can persist on surfaces for days or weeks." It does not publish separate survival times by surface type on its consumer pages, and neither do we — the table below records what the CDC says and what it does not.

For what this persistence means when a space has to be handed to someone else — a rental unit, a shared building — see our turnover guide for hosts and landlords.

What Is Not Known About Norovirus Transmission

Being clear about the limits of the evidence is part of reporting it accurately:

• The relative contribution of each route is not measured. The CDC says "most norovirus outbreaks happen when infected people spread the virus to others through direct contact," but no US surveillance system attributes a percentage of illnesses to airborne, surface, food, or water transmission.
• Individual cases are not reported. Per the CDC's reporting page, health departments "are not required to report individual cases of norovirus illness to CDC," most clinics do not test for it, and the 19–21 million annual US illnesses are therefore a modeled estimate. There is no case count to divide into, which is why the case-fatality rate on the Norovirus Risk Perspective page is shown as unavailable rather than computed.
• Infectious dose has been measured in one genotype, in adults. Both challenge studies used Norwalk virus (GI.1) in healthy adult volunteers. The GII.4 and GII.17 strains that cause most current outbreaks (see Outbreaks) have not been dosed into humans under controlled conditions, and children and older adults have not been studied this way. The same challenge studies are the source of the widely quoted "6 months to 2 years" immunity figure — and of the reason it is disputed — on our immunity and reinfection page.
• Detectable RNA is not proven infectious virus. The 28-day shedding figure measures genetic material, and the study's own authors flagged that limitation.

Compared to What?

Norovirus's transmissibility is exceptional, but its severity is not: with an estimated 19–21 million US illnesses and about 900 deaths a year (CDC), it sits at the high-incidence, low-lethality corner of the risk map. The Norovirus Risk Perspective page carries this site's sourced figures with their provenance, and the hub's Virus Risk Perspective tool places them beside 14 other viruses on the same five measures.

This content is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare provider for medical decisions.