When Should You Get a Flu Shot? Korea Starts at 75, Japan's Deadline Is Mid-December

Why is timing even a question

The advisory committee behind US federal guidance puts it this way: for most people who need only one dose, September or October is ideal, but vaccination should continue throughout the season as long as flu is circulating and unexpired vaccine is available. Vaccinating in July or August is generally not recommended for most groups, especially people 65 and older, because of the possibility that vaccine-induced immunity wanes over the course of the season. The CDC breaks this into four specific cases: most people needing one dose do well with September or October, ideally finishing by the end of October; most adults, especially those 65 and older and people in their first or second trimester of pregnancy, should generally avoid vaccinating too early in July or August; children needing two doses should get their first dose as soon as vaccine becomes available, since the second dose has to come at least four weeks later; and pregnant women whose third trimester falls in July or August may want to vaccinate then, since it helps protect the baby in its first few months, before it's old enough to be vaccinated itself. "Don't go too early" is not a universal rule. It reverses entirely for two specific groups.

The reasoning connects to when flu activity actually peaks. Across 40 flu seasons from 1982-83 through 2021-22, February was the most common peak month, occurring 17 times. Given that, finishing by the end of October might look too early. Both facts need to sit together: February is the most common peak, and the deadline is still set for October, because the concern is protection fading, not protection arriving too late.

What the calendar cannot settle is what the symptoms actually are. Congestion and sneezing at the turn of a season can be allergic, non-allergic or viral, and the three call for different responses: what sets the nose off when the weather changes, and why a mask stops particles but not a nerve reflex.

How do three countries handle it

Recommendations differ because flu seasons peak at different times in each country. One country's timing doesn't transfer to another.

Table: flu shot timing guidance by country. Since peak season timing differs, guidance differs too.

United States: peak in February, deadline in October

US

US flu activity typically peaks between December and February, and February has been the single most common peak month across four decades of tracked seasons, 17 out of 40. Despite that, the recommended cutoff for most people is the end of October. The entire justification is waning immunity: get vaccinated too early, and protection may fade before the season's later peak. A federal recommendation reaffirmed in August 2025 kept the underlying principle unchanged: annual vaccination for everyone six months and older without a contraindication.

That baseline hasn't moved. What has become more visible is a separate question that has nothing to do with when to get the shot, and everything to do with whose guidance to follow. That's covered on its own below, because it deserves care rather than a quick mention.

Japan: no start date, just a deadline

JP

Japan's health ministry answers the timing question with a single number, not a range. Its official guidance states that flu in Japan typically circulates from December to March, peaking in late January to early March, so "it is considered desirable to finish vaccination by mid-December." There is no recommended start date in that document. Local government and clinic websites often suggest starting in October, but that guidance doesn't come from the health ministry itself. The same document notes that 2023 was an exception, with flu activity peaking as early as December that year, an admission that "typical" doesn't hold every season.

What sets Japan apart is who's covered and at what cost. Routine vaccination is limited to people 65 and older, plus a narrow group aged 60 to 64 with specific heart, kidney, respiratory or HIV-related conditions severe enough to limit daily life. Children are not part of the routine vaccination category at all. Vaccination is generally not covered by national health insurance and is paid out of pocket, though municipalities may subsidize costs for those in the routine category, and some offer their own subsidy programs even for people outside it. Compare that to Korea, where children, pregnant women and seniors are vaccinated for free through a national program. In Japan, the question of when to get vaccinated comes after a different question: whether you're covered at all.

South Korea: vaccination by age tier, and a declaration that changes drug costs

KR

South Korea's national vaccination program runs on a staggered schedule by age and group. In the 2025-26 season, children 6 months to 13 years started September 22, pregnant women September 29, and adults 65 and older started in three separate waves by birth year, October 15, 20 and 22. Splitting seniors into three age bands is a deliberate design to spread out demand; on the very first day for the oldest group, 762,000 people were vaccinated. This staggered rollout has run at least since the 2022-23 season.

Korea also runs something the other two countries don't have a clear equivalent for: a formal "epidemic alert" system. The threshold is calculated from the average and standard deviation of flu-like illness rates over the previous three seasons, and once actual case rates cross that line, the government issues the alert. For 2025-26, it was declared on October 17. Once declared, high-risk groups, children, pregnant women, seniors, people with chronic conditions, become eligible for insurance coverage on antiviral prescriptions for suspected flu. Whether flu season has officially "started" isn't a personal judgment call in Korea. It's an administrative declaration, and that declaration changes what a prescription costs.

Korea's own health agency adds a caveat worth repeating: flu typically circulates from November to April, but "the timing of the flu season varies every year, and its start and end are difficult to predict." That sentence is the guardrail against treating any single recommended month as a fixed rule.

What's actually shifting in the US isn't the vaccine

This section does not evaluate the scientific safety or effectiveness of any vaccine. That's not something this article decides. What follows covers only observable, reported institutional changes and published survey figures.

In June 2025, the US Department of Health and Human Services removed the entire sitting membership of the CDC's vaccine advisory committee (ACIP), citing conflict-of-interest concerns. A newly formed committee then voted on flu vaccine recommendations, and that vote was adopted by the HHS Secretary the following month, becoming official guidance. Starting that September, state-level guidance began to diverge: some states formed regional public health alliances issuing their own recommendations separate from federal guidance, while at least one state moved to eliminate school vaccination requirements altogether. In January 2026, a federal decision memo reportedly reorganized how childhood vaccines, including flu, are recommended, shifting some into a "shared clinical decision-making" category rather than a universal recommendation. That decision and much of the prior committee's voting record were then paused by a federal court's preliminary injunction in March 2026, a ruling that was under appeal as of 12 August 2026. In August 2026, the American Academy of Pediatrics issued its own policy statement recommending annual flu vaccination for all children six months and older, before the federal government had issued its own recommendation for that season. What US readers are actually navigating isn't a claim that vaccines are unsafe. It's a structural question: a medical society issued guidance ahead of the federal agency, states diverged from federal guidance in both directions, and a court paused part of the federal process.

Polling reflects the same shift. In a KFF tracking survey, the share of adults saying they see the CDC as a trustworthy source of vaccine information fell from 59% in April 2025 to 47% in January 2026, with the decline differing by political affiliation. A separate Pew Research Center survey (October 2025, 5,111 adults) found support for school MMR vaccination requirements dropped from 82% in 2019 to 69% in 2025. That Pew survey did not ask about flu vaccines specifically, and this article doesn't apply its childhood-vaccine figures to influenza.

Here's the part that complicates the more alarming version of this story: vaccination rates themselves didn't collapse the way that narrative would predict. US pediatric flu vaccination coverage had already dropped between the 2023-24 and 2024-25 seasons, before the June 2025 committee overhaul. After the overhaul, the 2025-26 season's coverage held roughly flat compared to the prior season. The available data doesn't support a causal story where institutional turmoil caused a vaccination collapse. Trust, measured by survey, fell. Behavior, measured by actual vaccination rates, didn't move much in the same window. Those two facts sit next to each other without one explaining the other.

The institutional picture in this section is current as of 12 August 2026 and is still moving. When you settle on timing, check the latest guidance from the CDC and your state health department alongside it.

The US isn't the only country to have gone through a period of shaken vaccine trust. South Korea had its own in October 2020, when a cluster of deaths following flu vaccination made headlines. According to US CDC records from that period, 59 deaths were reported in the media, mostly among people in their 70s and 80s. Korea's disease control agency investigated 46 of those cases, performed autopsies, and found that every one of the deceased had a serious underlying condition sufficient to explain the death, with no cases linked to anaphylaxis following vaccination. The agency continued the vaccination program rather than pausing it. This isn't the same kind of event as the US advisory-structure changes; one was an epidemiological investigation into specific deaths, the other is a shift in advisory process and policy announcement. But the fact that vaccine trust wobbled in more than one country turns this into less of a spectacle about one nation and more of a mirror.

What to actually do, before and after

The recommended month differs by country, but the order of things to check is similar. Everything below comes from the health authorities' own published guidance, and the timing that fits your own situation is a decision for your clinician.

First, work out which of the four cases you're in. CDC guidance splits into four:

• Most adults needing only one dose: September or October, ideally finished by the end of October.
• Adults 65 and older, and women in their first or second trimester: generally avoid vaccinating early, in July or August.
• Children who need two doses: get the first dose as soon as vaccine is available, because the second has to come at least four weeks later.
• Women whose third trimester falls in July or August: vaccinating then may be worth considering.

Second, check who pays. In South Korea, free vaccination under the national program covers children 6 months to 13 years, pregnant women and adults 65 and older, and it can be received at any nearby designated clinic or public health center regardless of where you're registered as living. The eligibility window is announced separately each season, so check the schedule for the season in question (the 2025-26 dates are in the table above). In Japan, if you're not in the routine-vaccination category, vaccination isn't covered by national health insurance and is generally paid out of pocket; municipalities may cover costs for those in the routine category, and some run their own subsidy programs even for people outside it, so the local government's own notice is the thing to check. The cost structure in the United States is something this article was not able to confirm.

Third, if you missed the window. The advisory committee's own wording is that vaccination should continue throughout the season as long as flu is circulating and unexpired vaccine is available. "By the end of October" is an ideal completion date, not a point after which there's no reason to vaccinate.

Fourth, watch for the epidemic alert if you're in Korea. Once it's declared, high-risk groups, children, pregnant women, adults 65 and older, people with chronic conditions, become eligible for insurance coverage on antiviral prescriptions for suspected flu. Whether the season has started isn't a matter of personal impression there; it's an administrative declaration, and it changes what a prescription costs.

Fifth, if you catch it, how long to stay home. Korea's disease control agency advises staying home from school or work until 24 hours after the fever breaks and contagiousness has passed. If fever medication was used, it recommends continued observation for 48 hours after the last dose. During that window, avoiding contact with high-risk household members, including anyone 65 or older, is advised. If any of the emergency warning signs below appear, seek care without delay.

Emergency warning signs

The CDC states that anyone experiencing the following should seek care right away. For adults: difficulty breathing or shortness of breath; persistent pain or pressure in the chest or abdomen; persistent dizziness, confusion, or difficulty waking; seizures; not urinating; severe muscle pain; severe weakness or unsteadiness; fever or cough that improves and then returns worse; and worsening of chronic medical conditions. For children: fast breathing or trouble breathing; bluish lips or face; ribs pulling in with each breath; chest pain; severe muscle pain (a child refusing to walk); dehydration (no urine for 8 hours, dry mouth, no tears when crying); not being alert or interacting when awake; seizures; fever above 104°F (40°C) not controlled by fever-reducing medicine; any fever in a child under 12 weeks; and fever or cough that improves and then returns worse. This list isn't complete, and any other symptom that's severe or concerning is worth discussing with a clinician.

Frequently asked questions

Frequently asked

If I get vaccinated in September, will it lose effectiveness by winter?

That concern is exactly why US guidance discourages vaccinating too early, in July or August. September and October vaccination is still within the recommended window aimed at finishing by the end of October, so getting the shot in September isn't considered too early.

Is December too late?

It depends on the country. Japan's guidance says finish by mid-December; US guidance targets the end of October. The difference comes down to when each country's flu season actually peaks. And CDC guidance also notes that as long as flu is circulating and unexpired vaccine is available, vaccination should continue through the season, not stop after a single deadline.

Why do children need two doses?

Children who need two doses should get the first as soon as vaccine is available, because the second dose has to come at least four weeks after the first. The "don't go too early" advice that applies to most adults is reversed for this group.

Is the flu vaccine safe right now, given everything in the news?

This article doesn't make that determination; that's a scientific question outside its scope. What can be confirmed is that the US has seen real changes to its vaccine advisory structure and policy announcement process recently, and that public trust in health agencies, as measured by survey, has declined. What the available data does not show is that these changes caused an actual drop in vaccination rates.

Can I get vaccinated while pregnant?

CDC guidance generally advises against vaccinating too early (July or August) for women in their first or second trimester, while noting that women whose third trimester falls in that window may want to vaccinate then, since it helps protect the newborn early on. Beyond timing, safety decisions belong with your own clinician.