How vaccines work

A vaccine shows your immune system a harmless piece or weakened form of a pathogen — a surface protein, an inactivated virus, a live virus that has been attenuated so it can't cause disease, or an mRNA instruction to make one protein and nothing else. Your immune system responds by producing antibodies and memory cells. If you later meet the real pathogen, that memory lets your body respond in hours rather than the week or more it would otherwise take, which is often the difference between a mild illness and a severe one.

This is the same mechanism as immunity from infection, reached without having to survive the disease first. For several pathogens — tetanus, HPV, pertussis — vaccine-induced immunity is actually more reliable or longer-lasting than what infection provides.

Routine childhood schedule, birth to 18

This is the schedule endorsed by the American Academy of Pediatrics, the American Academy of Family Physicians, and the American College of Obstetricians and Gynecologists, and reflected in the immunization requirements of most states. The timing is not arbitrary: each dose is placed at the earliest age where the immune response is reliable and before the age of peak risk for that disease.

Adult schedule

Adult vaccination is where the largest coverage gaps are, and where a large share of preventable hospitalizations in people over 65 come from.

Vaccines in pregnancy

Two vaccines are recommended during every pregnancy, and one seasonally, because antibodies cross the placenta and protect the newborn during the months before the baby can be vaccinated directly.

• Tdap, every pregnancy, at 27–36 weeks. Timed so that pertussis antibody transfer peaks before delivery. This is the single most effective protection available for a newborn against whooping cough, which is most lethal in the first two months of life.
• Influenza, any trimester during flu season. Pregnancy substantially raises the risk of severe flu; vaccination also protects the infant for the first several months.
• RSV vaccine at 32–36 weeks during RSV season, as an alternative to giving the infant nirsevimab after birth.
• COVID-19 vaccination is recommended in pregnancy; the accumulated safety data across large pregnancy registries has not shown increased risk of miscarriage, preterm birth, or birth defects.

Avoid live vaccines in pregnancy — MMR and varicella in particular. These are given before pregnancy or immediately after delivery. This is a theoretical precaution rather than a demonstrated harm; inadvertent MMR in early pregnancy has not been shown to cause congenital rubella syndrome, and is not a reason to end a pregnancy.

Missed doses and catch-up

A late dose is not a wasted dose, and a series almost never has to be restarted. The immune system does not reset; it picks up where it left off. If your child is behind — for any reason, including a pandemic gap or a change in guidance — a clinician can build a catch-up plan that closes the gap with the minimum number of visits.

Two practical points that catch people out: minimum intervals between doses matter more than the exact ages, and doses given slightly early (more than 4 days before the minimum interval) generally don't count and must be repeated.

Safety: the real numbers

Vaccines have side effects. Presenting them honestly is the point — a claim that anything in medicine is risk-free is a reason to distrust the source. Here is what the surveillance data actually show.

Common and expected

Sore arm, redness at the injection site, fatigue, headache, low-grade fever, muscle aches. These usually appear within a day and resolve in one to three days. They reflect the immune system responding, which is the intended effect.

Rare, real, and quantified

Ingredients, plainly

• Aluminum salts (adjuvants) help inactivated vaccines produce a durable response, allowing lower antigen doses. An infant receives roughly 4 mg of aluminum total from the entire first-six-months vaccine schedule, compared with roughly 7 mg from breast milk and up to ~40 mg from formula over the same period. Aluminum is also ubiquitous in food and water. It has been used in vaccines since the 1930s.
• Thimerosal, an ethylmercury preservative, was removed from routine U.S. childhood vaccines by 2001 as a precaution — not because harm was found. Ethylmercury clears from the body far faster than the methylmercury in fish, which is the form associated with toxicity. Autism diagnoses continued rising after its removal, which is a direct falsification of the hypothesis that it was driving them.
• Formaldehyde is used to inactivate viruses and remains only in trace amounts. Your own body produces and metabolizes far more of it than any vaccine contains — a 2-month-old has roughly ten times more circulating naturally than the residual amount in a dose.
• Fetal cell lines. A few vaccines are grown in cell lines originally derived from two elective abortions in the 1960s. No fetal tissue is used in production today, and the finished vaccines contain no fetal cells. Several major religious authorities, including the Vatican, have explicitly stated that this is not a barrier to vaccination.

Vaccines and autism

This question deserves a direct answer rather than a dismissal, because it was asked in good faith by a great many parents and was given a bad answer for years.

Vaccines do not cause autism. This is one of the most thoroughly investigated questions in modern medicine, and the answer is not close.

Where the claim came from

A 1998 paper in The Lancet by Andrew Wakefield proposed a link between MMR and autism based on 12 children. It was fully retracted in 2010. An investigation by the UK General Medical Council found that patient data had been altered, that the children had been subjected to invasive procedures without ethical approval, and that Wakefield had been paid by lawyers preparing litigation against vaccine manufacturers while also filing a patent on a competing single-antigen measles vaccine. He was struck off the medical register for dishonesty and abuse of children in his care.

The replication attempts that followed involved millions of children across multiple countries and health systems, with different funding sources and research groups, and consistently found nothing.

What the evidence does support about autism

Autism is substantially heritable, with twin and family studies putting genetic contribution high, and hundreds of associated genetic variants identified. Prospective imaging and behavioral studies show that differences in brain development are detectable well before the age at which the MMR is given — meaning the developmental trajectory is already underway before the vaccine that was blamed for it. Rising diagnosis rates track closely with expanded diagnostic criteria, better recognition of autism in girls and adults, diagnostic substitution away from other labels, and increased screening.

Common questions

Isn't natural immunity better than vaccine immunity?

Sometimes it's stronger, and it always costs more to acquire. For measles, natural infection does produce robust lifelong immunity — obtained by having a disease that hospitalizes about 1 in 5 U.S. cases and kills roughly 1–3 per 1,000. That's the trade being proposed.

For several pathogens the vaccine is simply better. Tetanus infection produces no useful immunity at all — the toxin dose that makes you sick is too small to train the immune system, so survivors still need vaccinating. HPV infection often fails to generate protective antibodies against reinfection, while the vaccine reliably does. Pertussis immunity wanes after infection much as it does after vaccination.

Hybrid immunity — infection plus vaccination — is generally the most robust of all for COVID-19, but that's an argument for vaccinating people who've been infected, not for skipping vaccination.

Why vaccinate against diseases nobody gets anymore?

Because nobody gets them because of vaccination. This is the paradox at the center of vaccine hesitancy: success removes the visible evidence for itself.

The natural experiment has been run repeatedly. Japan cut pertussis vaccination in the mid-1970s and went from a few hundred cases to over 13,000 with dozens of deaths within a few years. The UK's coverage drop after the 1998 Wakefield paper was followed by measles becoming endemic again in 2008 after 14 years of elimination. Wherever coverage falls, these diseases return — usually within a few years, and reliably.

The pathogens have not gone anywhere. Measles remains endemic in much of the world and travels by airplane. Polio still circulates. Tetanus lives permanently in soil and can never be eradicated at all.

My child had a reaction. Should they get the next dose?

Usually yes, but this is exactly the case to discuss with a clinician rather than decide from a website. Fever, fussiness, and a sore leg are expected reactions, not contraindications. True contraindications are narrow: a confirmed anaphylactic reaction to a previous dose or to a known vaccine component, and for live vaccines, significant immunosuppression or pregnancy.

Egg allergy is no longer a barrier to flu vaccination at any severity, including anaphylaxis — that guidance changed as the evidence came in.

If a reaction was severe, allergy testing can often identify the specific component and let the rest of the schedule proceed safely.

Can I space the shots out to be safe?

You can, but there's no evidence it helps and clear evidence it costs something. Alternative schedules have never demonstrated a benefit on any measured outcome. What they reliably do is extend the period during which a child is susceptible, and add clinic visits — which in practice means more missed doses, not fewer.

The specific worry usually underneath this question — that simultaneous vaccines overload a child's immune system — has been tested directly. Studies comparing children on the recommended schedule with those on delayed schedules found no neuropsychological benefit to delay.

Does the flu shot give you the flu?

No. The injected vaccine contains no live virus and cannot replicate. What happens is that it takes about two weeks to develop protection, and it's given during the season when respiratory viruses are everywhere — so people who catch something in that window naturally connect the two. Randomized trials comparing flu vaccine with saline placebo find similar rates of subsequent illness, but more arm soreness in the vaccine group.

Effectiveness varies year to year — typically 40–60% against symptomatic illness when the strain match is good, less in a mismatched year. That's genuinely lower than most vaccines. It still translates into a substantial reduction in hospitalizations and deaths, and vaccinated people who do get sick tend to get less sick.

Do vaccine manufacturers just profit from this?

They do make money, and that's a fair thing to keep an eye on. But it's a weak explanation for the evidence base. Vaccines are among the lower-margin pharmaceutical products — they're given a handful of times over a lifetime, unlike a daily medication for a chronic condition. Preventing disease is, commercially speaking, worse business than treating it.

More to the point, the safety and effectiveness data come from many independent sources: national health systems in Denmark, Finland, the UK, and Israel with no financial stake in U.S. vaccine sales, academic groups, and publicly funded surveillance networks. A conflict of interest is a reason to scrutinize a claim, not a substitute for examining whether the claim is true. Here, independent replication is exactly what has happened.

What if I've lost my vaccination records?

Most U.S. states maintain an immunization information system that your clinician can query. If nothing turns up, options are to repeat doses — safe, if sometimes producing a stronger sore arm — or to check antibody titers for a few diseases where a blood test is informative, such as measles, rubella, hepatitis B, and varicella. Titers are not useful for every vaccine, so ask which applies.

Community immunity, and why the threshold matters

Vaccination protects the person vaccinated and, past a certain coverage level, protects people who can't be vaccinated at all — infants too young for their first dose, people on chemotherapy or immunosuppressants, transplant recipients, and the small percentage in whom a vaccine simply didn't take.

The threshold isn't a matter of opinion; it follows from how contagious the pathogen is. Measles is one of the most transmissible diseases known — one case in a fully susceptible population produces roughly 12 to 18 more — which is why it needs about 95% coverage to stop chains of transmission, a higher bar than almost anything else.

This is also why coverage statistics that look fine nationally can be misleading. Immunity isn't evenly distributed — it clusters by school, neighborhood, and community. A county at 93% can contain a school at 70%, and that school is where an outbreak starts.

Last reviewed: August 2026. Schedules summarized here reflect the recommendations of the American Academy of Pediatrics, American Academy of Family Physicians, American College of Obstetricians and Gynecologists, and Infectious Diseases Society of America. Recommendations for COVID-19 and RSV vaccines have changed more than once in recent years — confirm current guidance with a clinician. See Evidence & Sources.