E. coli is moderately contagious and usually requires a substantial oral dose to cause illness, so most exposure events do not end in infection. The infectious dose for the most outbreak-prone strain, Shiga toxin-producing E. coli O157:H7, can be extremely low, on the order of tens to a few hundred cells. Ordinary diarrheagenic strains typically need far more, often 100,000 cells or more. Person-to-person spread is real but is a minor route compared with food and water.
Contagiousness is the wrong single question. A better frame asks three questions at once. How many bacteria must be swallowed? How well does the strain survive stomach acid? How readily does an infected person shed the organism onto hands, food, or surfaces? Those three variables explain why one family member can infect a household while another eats the same food and stays well.
Dose and stomach defenses
Infection begins when live bacteria reach the intestine and attach to its lining. The stomach is the first barrier. Gastric acid at pH 1.5 to 3.5 kills most swallowed bacteria within minutes, so the effective dose is whatever survives that transit. This is why the infectious dose is not a fixed number and why antacid use, fasting, and age shift susceptibility.
The O157:H7 strain is unusually good at surviving this gauntlet. It tolerates acid better than many commensal E. coli and needs only a small number of cells to establish infection. Estimates from outbreak investigations place the O157 infectious dose in the range of 10 to 100 cells, though exact figures are difficult to confirm in humans. Other pathotypes behave differently. Enterotoxigenic E. coli, a common cause of traveler's diarrhea, typically requires a much larger dose, often cited in the range of 100,000 to 10 million cells.
Attachment matters as much as survival. Pathogenic strains carry adhesion factors that let them cling to intestinal cells and resist being washed away. Once attached, they multiply and, for Shiga toxin-producing strains, begin releasing toxin that damages blood vessels in the gut and, in severe cases, the kidneys. The toxin, not the bacteria themselves, drives the most dangerous complications.
A plain-language summary: the bug has to survive the acid bath, grab hold of the gut wall, and multiply faster than the body can clear it. Small doses favor strains that do all three well.
Quantified context: outbreak size and secondary spread
E. coli infections are common. The CDC estimates that Shiga toxin-producing E. coli cause roughly 265,000 illnesses in the United States each year, with about 3,600 hospitalizations and 30 deaths. Most of those cases are foodborne, but a meaningful share come from person-to-person contact and animal exposure.
Secondary attack rates show how readily it spreads in close settings. In household outbreaks of O157, studies have reported secondary infection rates ranging from roughly 4 percent to 15 percent of contacts, with higher rates among young children and in settings with poor hand hygiene. Daycare outbreaks can spread quickly because toddlers are prone to hand-mouth behavior and diaper changes concentrate the organism.
Shedding drives transmission. Infected people can shed O157 in stool for one to three weeks, and children can shed longer. The organism is present in high numbers during the acute diarrheal phase, which is when contamination of hands, toys, and food is most likely. Asymptomatic or mild infections can still shed bacteria, which helps explain household spread that seems to have no clear source.
The dose and the setting interact. A small inoculum that would be harmless in a healthy adult may cause illness in a toddler who shares toys and puts hands in the mouth. Contagiousness is therefore not a property of the organism alone. It is a property of the organism, the host, and the environment together.
Edge cases that change the risk
Several conditions raise the odds that a small exposure becomes an infection. Infants and young children have immature immune systems and lower stomach acid production. Older adults have weaker immune defenses and more frequent acid-suppressing medication use. People taking proton pump inhibitors or antacids have less gastric acid, which lowers the dose needed to establish infection.
Pregnancy adds a specific concern. O157 infection during pregnancy carries a higher risk of severe complications, including hemolytic uremic syndrome, which can threaten both mother and fetus. Pregnant people should treat bloody diarrhea as a reason to seek care promptly rather than wait.
Immunocompromised people, including those on chemotherapy or transplant immunosuppression, face higher risk of severe disease and prolonged shedding. They are also more likely to spread the organism for longer periods.
Animal contact is an underappreciated route. Petting zoos, farm visits, and live animal exhibits have produced outbreaks, particularly among children. The organism lives in the intestines of cattle and other ruminants and is shed in manure. Handwashing after animal contact is not a formality; it is the main control measure.
A professional caveat worth stating plainly: the infectious dose numbers are estimates, not constants. They come from outbreak reconstructions and experimental challenge studies, and they vary with strain, host, and food matrix. Treat them as order-of-magnitude guides, not thresholds that guarantee safety or illness.
Misconceptions that distort the risk
The most common misconception is that E. coli is uniformly dangerous. The bacteria are a normal part of the healthy human gut. Most strains are harmless or beneficial. Only certain pathotypes, chiefly Shiga toxin-producing strains such as O157:H7 and a handful of others, cause serious illness. The name on the label is not the whole story.
A second misconception is that food poisoning is always foodborne. Person-to-person spread is well documented in households and daycares. So is spread from animals. Assuming a food source can lead people to miss the hand-hygiene steps that stop secondary cases.
A third misconception is that freezing or refrigeration kills the organism. It does not. Cold storage slows growth but does not eliminate E. coli. Thorough cooking is what kills it, and the temperature that matters is the internal temperature of the food, not the surface.
A fourth misconception is that symptoms appear immediately. The incubation period for O157 is typically three to four days, with a range of roughly two to eight days. That delay makes source tracing harder and makes it easy to blame the wrong meal.
[RED FLAGS] Bloody diarrhea, especially in a child or older adult, warrants prompt medical evaluation. Diarrhea with decreased urination, extreme fatigue, or pallor can signal hemolytic uremic syndrome and is an emergency. Do not use antidiarrheal medications for suspected E. coli O157 without medical advice; slowing transit may worsen toxin exposure. Antibiotics for suspected O157 are controversial and should be decided by a clinician, not self-treated.
Side-by-side routes and their relative risk
| Route | Typical source | Relative contribution | Key control |
|---|---|---|---|
| Food | Undercooked ground beef, raw produce, unpasteurized juice | Largest single route | Cook to safe internal temperature; wash produce |
| Water | Contaminated wells, recreational water | Moderate, outbreak-linked | Test private wells; avoid swallowing untreated water |
| Person-to-person | Households, daycares, care settings | Minor but important | Handwashing; exclude symptomatic food handlers and children |
| Animal contact | Petting zoos, farms, livestock | Small but preventable | Handwashing after contact; supervise children |
| Raw milk | Unpasteurized dairy | Small but high-risk per exposure | Pasteurization |
The table is a ranking of contribution, not of severity per exposure. Raw milk and animal contact involve fewer total cases but carry high risk per event because the dose can be substantial.
Timeline: what happens after exposure
| Time after exposure | What is happening | Practical implication |
|---|---|---|
| 0 to 24 hours | Bacteria survive stomach acid and reach the intestine | Exposure is silent; nothing detectable yet |
| 1 to 3 days | Attachment and multiplication begin | No symptoms; person may already shed |
| 3 to 4 days | Diarrhea typically starts, often with cramps | Peak shedding; highest transmission risk |
| 4 to 7 days | Diarrhea peaks; bloody stools may appear | Seek care for blood or severe symptoms |
| 7 to 10 days | Most cases resolve | Shedding continues; hygiene still matters |
| 1 to 3 weeks | Shedding tapers; HUS risk window passes | Children may shed longer |
| Up to 6 weeks | Rare prolonged shedding | Food handlers and childcare workers need clearance guidance |
This timeline explains why exclusion policies for food handlers and daycare attendees are not arbitrary. A person can feel well and still shed bacteria. Public health guidance varies by jurisdiction, but the principle is consistent: keep shedding people out of high-risk settings until clearance criteria are met.
Practical actions with concrete thresholds
The following steps reduce both primary exposure and secondary spread. They are deliberately concrete so they can be acted on, not just understood.
Cook ground beef to an internal temperature of 160 degrees Fahrenheit, and poultry to 165 degrees Fahrenheit. Use a digital thermometer; color and juice clarity are unreliable. Wash hands with soap and water for at least 20 seconds after using the bathroom, changing diapers, handling raw meat, and touching animals. Alcohol-based sanitizer is a fallback, not a substitute, when hands are visibly soiled. Keep raw meat separate from ready-to-eat foods. Use separate cutting boards, and wash utensils and surfaces that touched raw meat with hot soapy water. Avoid unpasteurized milk, juice, and cider. Pasteurization is the control step; refrigeration is not. Test private wells annually for coliform bacteria, and treat any positive result before drinking. Boiling water for one minute at a rolling boil kills E. coli and other bacteria. Exclude anyone with diarrhea from food preparation and from childcare until symptoms resolve and, where local guidance requires it, until stool testing clears them.
Two thresholds deserve explicit reconciliation because they are often confused. The cooking threshold is an internal temperature: 160 degrees Fahrenheit for ground beef kills E. coli. The handwashing threshold is a duration: 20 seconds of soap and water. The first addresses the food; the second addresses the transmission route that food cannot cover. Both are needed, and neither substitutes for the other.
A final practical note on testing. Clinical stool testing for Shiga toxin is what identifies dangerous strains. If a clinician suspects E. coli O157, the sample should be sent for Shiga toxin testing specifically, because routine culture may miss the strain or fail to distinguish it from harmless E. coli. This is a laboratory decision, not a patient decision, but it is worth knowing that a generic "stool culture" may not answer the question.
Limits of what the evidence can say
Several parts of this picture remain uncertain. Infectious dose estimates are derived from outbreaks and challenge studies, not from controlled human trials with O157, which would be unethical. Secondary attack rates vary widely across studies because of differences in household size, hand hygiene, and case definitions. Shedding duration is measured in days to weeks but varies by age and immune status, and the exact relationship between shedding and transmission is not fully quantified.
The strain picture is also moving. Non-O157 Shiga toxin-producing strains, including O26, O111, and O103, account for a growing share of illnesses in some surveillance systems. Their infectious doses and transmission dynamics are less well characterized than O157's. That gap is a reason to treat the O157 numbers as the best-studied case rather than the universal rule.
What can be said with confidence is that contagiousness is not a fixed property. It depends on the dose, the strain, the host, and the setting. Small doses of the right strain in the wrong host can cause illness; large doses of ordinary strains often do not. Interventions that interrupt the fecal-oral route, especially handwashing and safe food handling, work regardless of which strain is involved. That is the practical takeaway, and it is more useful than a single contagiousness number.