Norovirus Immunity: Why You Can Get It Again
Most viral infections leave you protected for years. Norovirus is the exception people discover the hard way — a second bout the following winter, or the same month. The reasons are known, the numbers behind them disagree with each other in an interesting way, and about one person in five turns out to be largely immune from birth.
The short version
Yes, you can get norovirus twice — CDC says infection with one type "may not protect you against other types" and that "it is not known exactly how long protection lasts". Two things are going on: protection against the strain you had is temporary, and there are many strains, so protection against one does little against the next. The widely quoted "6 months to 2 years" is a real figure from real experiments, but a model built on real-world data puts natural immunity at 4 to 9 years. Both are explained below.
Where "6 months to 2 years" comes from
Almost every source repeats that norovirus immunity lasts six months to two years. The figure traces to a handful of human challenge studies — experiments in which healthy adult volunteers swallowed the prototype Norwalk virus on purpose and were watched. They are the only direct measurements that exist, and it is worth seeing what they actually found.
| Study | What was done | What happened |
|---|---|---|
| Parrino et al., NEJM 1977 | 12 volunteers challenged; 6 fell ill. All 12 re-challenged 27 to 42 months later. Four of the ill volunteers challenged a third time 4 to 8 weeks after their second illness. | At 27–42 months, the same six became ill again and the same six stayed well. At 4–8 weeks, three of four were protected. The authors' reading: "two forms of immunity … one of short and the other of long duration" — and serum antibody was not what predicted protection. |
| Johnson et al., J Infect Dis 1990 | 42 volunteers challenged; 22 challenged again 6 months later; 19 a third time after 6 more months. | Only 4 of those challenged twice became ill, and none after a third challenge. "Short-term resistance lasts ≥ 6 months after challenge." Pre-existing antibody was not protective at first but "became associated with protection after repetitive exposure". |
So the experimental picture is: protection against the same strain is solid for weeks to months, and gone by two to three years. That is the whole basis of "6 months to 2 years", and within its limits it is sound.
Why a model says 4 to 9 years instead
In 2013 a group at Emory and CDC pointed out that several real-world observations do not fit a six-month-to-two-year immunity — if protection were that short, given how infectious norovirus is, adults would be ill far more often than they are. They built a transmission model, fitted it to age-specific incidence data from England and Wales, and estimated the duration of immunity to norovirus gastroenteritis at 4.1 years (95% CI 3.2–5.1) to 8.7 years (95% CI 6.8–11.3), depending on the scenario (Simmons et al., Emerg Infect Dis 2013).
The reconciliation is the dose. The challenge studies gave volunteers doses vastly larger than anything encountered in a kitchen or a cruise-ship buffet — a separate analysis of the same challenge data estimated that a single Norwalk virus particle has close to a 50% chance of establishing infection, and that the probability of becoming ill rose with dose from about 0.1 to 0.7 across the range tested (Teunis et al., J Med Virol 2008). Immunity that fails against a deliberate high-dose challenge may hold up well against ordinary exposure. The model measures the second thing; the experiments measured the first.
How to hold both numbers
Against the strain you had: expect months of solid protection, probably some years of partial protection against ordinary exposure, and none by three years in the face of a heavy dose. CDC's official position — "it is not known exactly how long protection lasts" — is the accurate summary. The bigger reason you get it again is not that immunity fades; it is the next section.
Immunity is strain-specific, and the strains keep changing
Noroviruses are classified into ten genogroups and dozens of genotypes (Chhabra et al., J Gen Virol 2019); two genogroups, GI and GII, cause nearly all human illness, and one genotype, GII.4, has caused the majority of outbreaks worldwide since 2002 (CDC). Immunity from one genotype does little against another — which is why CDC's warning is phrased around "types".
GII.4 adds a second problem. It evolves. Its capsid — the protein shell that antibodies recognise and that binds to gut cells — changes in and around the receptor-binding region in a way that lets new variants both escape existing antibodies and keep binding (Donaldson, Lindesmith et al., Immunol Rev 2008). CDC records that new GII.4 variants emerged roughly every 2 to 4 years between 2002 and 2012 — Minerva 2006, New Orleans 2009, Sydney 2012 — each driving a global rise in outbreaks until population immunity caught up. Then a different genotype altogether, GII.17, caused three-quarters of typed US outbreaks in 2024–2025 and nearly doubled the outbreak count, before GII.4 Sydney returned to the top in 2025–2026. Our outbreaks page carries the CaliciNet season table.
This is what "every year" means in practice. A person who had GII.4 Sydney in 2023, GII.17 in the 2024–25 wave and GII.4 again in 2025–26 did not have failing immunity — they met three different targets. Strain-specific immunity against a virus that rotates strains is, from the patient's point of view, almost no immunity at all.
The one in five who are resistant from birth
The most striking finding in the challenge studies was that some volunteers could not be infected at any dose. In 2003 the reason was identified. Norovirus attaches to gut cells using histo-blood group antigens — the same sugar structures that determine ABO blood type — and their display on the gut lining depends on a working FUT2 gene. People with two non-functional copies, called non-secretors, do not present the sugar the virus needs.
- In the 2003 challenge study, 29% of volunteers were non-secretors, and none of them became infected regardless of dose — the resistance was "fully penetrant" (Lindesmith et al., Nat Med 2003).
- Non-secretors are resistant to several genotypes, "including the predominant GII.4" — but not to all of them; some genotypes bind different sugars. Non-secretor frequency varies widely between populations, which is one reason norovirus epidemiology differs by region (Nordgren & Svensson, Viruses 2019).
- Among secretors, ABO type matters too. In a 51-volunteer challenge study, people with blood group O were more likely to be infected (odds ratio 11.8, 95% CI 1.3–103) and people with a B antigen less likely (Hutson et al., J Infect Dis 2002). The intervals are wide — this is a small study — and the effect is for the GI.1 strain; the pattern differs by genotype.
- Even among secretors with a functional gene, a portion resisted infection, "suggesting that a memory immune response or some other unidentified factor also affords protection" (Lindesmith 2003). Resistance is multifactorial.
The practical reading: if you are the person in the household who never seems to get it, there is a decent chance you are a non-secretor — protected against the dominant strains, not against every strain, and not detectably different in any other way. There is no routine test for it and no reason to seek one.
Why repeated exposure still matters — and why it does not add up to a vaccine yet
Immunity does accumulate against a given strain. In the multiple-challenge study nobody fell ill after the third exposure, and antibody levels became associated with protection only after repeated challenge (Johnson 1990). This is why the disease is hardest on the youngest: CDC reports that young children account for most norovirus emergency-department visits, and the 2013 model calculated that children under five are much more infectious than older children and adults.
It is also why a vaccine is possible in principle. The first efficacy trial (Atmar et al., NEJM 2011) gave 98 adults two doses of a virus-like-particle vaccine or placebo, then challenged them with the matching Norwalk strain: gastroenteritis occurred in 69% of placebo recipients against 37% of vaccine recipients. Protection can be induced. The difficulty is everything above — breadth across genotypes, durability, and a dominant lineage that changes its coat every few years. As of September 2026 there is no approved norovirus vaccine; the status of the leading Phase 3 candidate is kept current on our FAQ.
What this changes about prevention
Because you cannot rely on having had it, every defence is hygiene, every time: soap-and-water handwashing (alcohol sanitiser underperforms against a virus with no lipid envelope), chlorine bleach or an EPA-registered product with a norovirus claim for surfaces, and staying away from food preparation for 48 hours after symptoms stop — the guidance on our prevention page and the reason it matters is on how norovirus spreads. A previous infection is not a reason to relax any of it.
This page explains published research and CDC guidance and is not medical advice. It cannot tell you whether you are immune, whether a current illness is norovirus, or whether you or a family member needs care — for those questions, a clinician can help. Signs of dehydration in a child or older adult warrant prompt medical attention.
Frequently Asked Questions
Can you get norovirus twice?
Yes. CDC says one type may not protect against others and the duration of protection is not known exactly. Immunity to a strain is temporary — volunteers re-exposed 27–42 months later became ill again — and immunity is strain-specific, so last winter's strain does little against this winter's. About a fifth of people of European descent are non-secretors and resistant to the dominant strains.
How long does immunity last?
The experiments say months to under three years against the same strain at high dose; a model fitted to real-world data says 4.1 to 8.7 years after natural infection. The gap is dose — challenge studies use far more virus than ordinary exposure. CDC's summary, "not known exactly", is the fair one. Either way, protection is against the strain you had.
Why do some people get it every year and others never?
Genetics and exposure. Non-secretors (two non-working FUT2 copies; 20–30% of Europeans) do not display the sugar the dominant strains bind and were completely resistant in the 2003 challenge study. Blood group O was associated with more infection and a B antigen with less in one small study. Everyone else depends on which strain they meet and when.
Can you catch it again straight away?
From the same strain, probably not for months — re-challenge at 4–8 weeks left three of four protected and short-term resistance lasted at least 6 months. From a different strain circulating in the same house or ship, yes; back-to-back illnesses in one season usually mean two strains.
Is the second infection milder?
Against related strains, often — nobody was ill after a third challenge with the same virus, and this is why the disease is hardest on young children with no immune history. But GII.4 evolves to escape prior immunity every few years, so a prior infection is not a reliable guarantee against what is circulating now.
Why is there no vaccine?
Natural immunity is short-lived and strain-specific, and a vaccine has to beat that against a moving target. The 2011 challenge trial showed protection is inducible (illness 69% on placebo vs 37% vaccinated, same strain). Breadth, durability and GII.4 drift are the obstacles. No approved vaccine as of September 2026; see the FAQ for the Phase 3 candidate.
Sources & References
- Centers for Disease Control and Prevention. About Norovirus — "infection with one type of norovirus may not protect you against other types"; "it is not known exactly how long protection lasts". cdc.gov/norovirus/about
- Parrino TA, Schreiber DS, Trier JS, Kapikian AZ, Blacklow NR. Clinical immunity in acute gastroenteritis caused by Norwalk agent. N Engl J Med 1977;297(2):86–9 — re-challenge at 27–42 months and at 4–8 weeks; "two forms of immunity … one of short and the other of long duration". doi:10.1056/NEJM197707142970204
- Johnson PC, Mathewson JJ, DuPont HL, Greenberg HB. Multiple-challenge study of host susceptibility to Norwalk gastroenteritis in US adults. J Infect Dis 1990;161(1):18–21 — short-term resistance ≥ 6 months; none ill after a third challenge; antibody associated with protection only after repeated exposure. doi:10.1093/infdis/161.1.18
- Simmons K, Gambhir M, Leon J, Lopman B. Duration of immunity to norovirus gastroenteritis. Emerg Infect Dis 2013;19(8):1260–7 — model fitted to England and Wales incidence: 4.1 (3.2–5.1) to 8.7 (6.8–11.3) years; children under 5 much more infectious. doi:10.3201/eid1908.130472
- Teunis PF, Moe CL, Liu P, et al. Norwalk virus: how infectious is it? J Med Virol 2008;80(8):1468–76 — probability of infection per single particle close to 0.5; dose-dependent probability of illness 0.1 to 0.7. doi:10.1002/jmv.21237
- Lindesmith L, Moe C, Marionneau S, et al. Human susceptibility and resistance to Norwalk virus infection. Nat Med 2003;9(5):548–53 — 29% of volunteers FUT2 non-secretors, none infected at any dose; resistance multifactorial. doi:10.1038/nm860
- Hutson AM, Atmar RL, Graham DY, Estes MK. Norwalk virus infection and disease is associated with ABO histo-blood group type. J Infect Dis 2002;185(9):1335–7 — group O OR 11.8 (95% CI 1.3–103); B antigen protective; n = 51. doi:10.1086/339883
- Nordgren J, Svensson L. Genetic susceptibility to human norovirus infection: an update. Viruses 2019;11(3):226 — non-secretors resistant to several genotypes including GII.4; population variation in HBGA expression. doi:10.3390/v11030226
- Donaldson EF, Lindesmith LC, Lobue AD, Baric RS. Norovirus pathogenesis: mechanisms of persistence and immune evasion in human populations. Immunol Rev 2008;225:190–211 — GII.4 capsid evolution around the receptor-binding domain as the mechanism of escape from herd immunity; GII.4 ≈ 80% of infections at the time. doi:10.1111/j.1600-065X.2008.00680.x
- Chhabra P, de Graaf M, Parra GI, et al. Updated classification of norovirus genogroups and genotypes. J Gen Virol 2019;100(10):1393–1406 — ten genogroups (GI–GX). doi:10.1099/jgv.0.001318
- Atmar RL, Bernstein DI, Harro CD, et al. Norovirus vaccine against experimental human Norwalk virus illness. N Engl J Med 2011;365(23):2178–87 — gastroenteritis in 69% of placebo vs 37% of vaccine recipients after homologous challenge. doi:10.1056/NEJMoa1101245
- Centers for Disease Control and Prevention. CaliciNet Data — GII.4 variant emergence every 2–4 years 2002–2012; the 2024–2025 GII.17 season and the 2025–2026 return of GII.4 Sydney, as tabulated on our outbreaks page. cdc.gov/norovirus/php/reporting/calicinet-data
How It Spreads
The infectious dose, the shedding window, and why sanitizer underperforms.
Outbreaks & Strains
The CaliciNet season table — GII.17's year and GII.4 Sydney's return.
Prevention
CDC's handwashing, bleach and 48-hour guidance — the defences that do not depend on immunity.
FAQ
Contagious period, sanitizer, the circulating strain and the vaccine pipeline.
