The experts who sounded worried mid-season turned out to be right. Now the season is over and the final numbers are in, they’re worse than the warnings suggested.
The 2024–25 season was classified by the CDC as a high severity season — overall and for every age group: children, adults and older adults. It was the first high severity season since 2017–18.
The hospitalisation rate was the highest in the surveillance record. FluSurv-NET recorded 38,960 influenza-associated hospitalisations between 1 October 2024 and 30 April 2025, at a cumulative rate of 127.1 — against a median of 62.0 for the 2010–11 through 2023–24 seasons, and above even 2017–18’s 102.9.
And 289 children died. That is the highest paediatric flu death toll in the United States since the 2009–10 H1N1 pandemic, and the highest for a non-pandemic season since child deaths became nationally notifiable in 2004. 🌡️
- What “high severity” means
- The hospitalisation numbers
- 289 children
- The season in totals
- How the season ran
- Why it was so bad
- What to take from it
- The vaccination conversation
- Antivirals and the 48-hour window
- Who’s at higher risk
- When flu becomes an emergency
- Frequently asked questions
- Your checklist
What “high severity” means
This isn’t a journalistic adjective. The CDC assigns a formal severity classification to each season using hospitalisation rates, outpatient visits for influenza-like illness and mortality, compared against historical thresholds.
2024–25 was classified as high severity overall, and high severity for children, adults and older adults separately. That combination is what makes it notable — a season can be hard on one age group and unremarkable for others. This one was severe across the board.
It was also the first high severity season since 2017–18 — seven seasons earlier, and a benchmark year in flu epidemiology precisely because of how bad it was.
The hospitalisation numbers
| Measure | 2024–25 | Comparison |
|---|---|---|
| Hospitalisations recorded | 38,960 (FluSurv-NET) | 1 Oct 2024 – 30 Apr 2025 |
| Cumulative rate | 127.1 | Median 2010–11 to 2023–24: 62.0 |
| Previous highest | — | 102.9 (2017–18) |
| Previous lowest | — | 8.7 (2011–12) |
Read that table slowly, because the comparison is what matters. The rate was roughly double the median of the previous fourteen seasons, and it exceeded the worst of them.
FluSurv-NET is a surveillance network covering a defined catchment population rather than the whole country, so the 38,960 figure is a network count, not a national total — that comes later. The rate is the number designed for comparison across seasons, and it’s the one that stands out.
289 children
The paediatric figure deserves its own section, because it’s the one that shouldn’t be scrolled past.
289 paediatric deaths were reported for the 2024–25 season. Two comparisons put it in context:
- It is the highest number reported in the US since the 2009–10 H1N1 pandemic.
- It is the highest for a non-pandemic flu season since paediatric flu deaths became nationally notifiable in 2004.
That second point is the significant one. Pandemic years are, by definition, exceptional. This was an ordinary seasonal influenza year — no novel virus, no global emergency — and it killed more American children than any other ordinary year in two decades of national reporting.
Flu is widely treated as an adult-inconvenience illness — something that makes you miserable for a week. The paediatric figures from this season are a direct contradiction of that, and they’re the reason paediatric flu vaccination guidance exists.
Historically, a substantial proportion of children who die from influenza were not vaccinated, and many had no underlying medical condition. Previously healthy children do die of flu. That is not a scare tactic; it is what the notifiable-disease reporting shows.
The season in totals
Beyond the surveillance network, the CDC produces national burden estimates. For 2024–25 those reached at least:
| Estimate | Figure |
|---|---|
| Illnesses | 46 million |
| Hospitalisations | 600,000 |
| Deaths | 26,000 |
These are modelled estimates rather than counted cases — most people with flu are never tested, so any national figure has to be estimated from surveillance data. That’s standard practice and it’s worth knowing when comparing them with confirmed-case counts, which are always far lower.
How the season ran
The shape of the season is useful for anyone planning around the next one:
- Mid-November 2024: activity began increasing.
- Early February 2025: peak.
- By May 2025: declined to interseasonal levels.
- Through August 2025: remained low.
That’s a fairly typical timing pattern — the severity was in the height of the curve rather than an unusual schedule. It also explains why vaccination timing matters: a peak in early February means a vaccine given in October or November has had time to take effect before the worst of it, while waiting until January is cutting it fine.
Why it was so bad
Season severity is driven by several things interacting, and it’s worth being honest that attribution after the fact is imprecise:
- Which viruses dominate. Influenza A(H3N2)-dominant seasons have historically tended to be harder, particularly on older adults, than H1N1-dominant ones.
- How well the vaccine matches the circulating strains, which varies year to year because strain selection happens months in advance.
- Population immunity from prior seasons.
- Vaccination coverage, which affects both individual protection and transmission.
- Chance. Genuinely — the specific mix and timing of circulating strains varies in ways nobody fully predicts.
What can be said with confidence is the outcome, and the outcome was a season worse than any in the preceding fourteen by hospitalisation rate.
What to take from it
The slug on this article says 2025 and the season is over — so the honest value now is what it changes going forward.
- “Just flu” is not a defensible position. 289 children and an estimated 26,000 deaths in a non-pandemic year settles that.
- Severity varies enormously between seasons — from a rate of 8.7 to 127.1 across fourteen years. A mild season tells you nothing about the next one.
- You cannot know in advance which kind of season you’re heading into. Decisions have to be made before the answer is available.
- Timing matters. A February peak means autumn vaccination, not January.
- Children are not exempt, including healthy ones.
The vaccination conversation
Worth being straightforward about what the flu vaccine does and doesn’t do, because overselling it has done real damage to trust.
It is not like the measles vaccine. Effectiveness varies by season and by strain match, and in a poorly matched year it can be modest. People who get vaccinated do sometimes get flu.
What it more reliably does is reduce severity — the likelihood of hospitalisation, intensive care and death — even when it doesn’t prevent infection. In a season with a hospitalisation rate double the historical median, that is the relevant benefit.
Practical points: get it in autumn, ahead of a typical winter peak; it’s recommended annually because both the virus and your immunity change; and if you have animal exposure, there’s a separate argument for it covered in our piece on avian influenza surveillance.
Antivirals and the 48-hour window
This is the most actionable and least known part of flu care.
Antiviral medicines for influenza work best when started early — ideally within 48 hours of symptoms beginning. They are prescription medicines, they are not for everyone, and they are most valuable for people at higher risk of complications.
The practical consequence: if you’re in a higher-risk group, “wait and see if it gets better” can close the window that matters. Contact a doctor early rather than after several days, and say clearly when your symptoms started — that time is what determines the decision.
Who’s at higher risk
| Group | Note |
|---|---|
| Adults 65 and over | Highest hospitalisation and death rates |
| Children under 5, especially under 2 | Higher complication risk |
| Pregnancy and up to 2 weeks postpartum | Higher risk of severe illness |
| Chronic lung disease | Asthma, COPD |
| Heart disease | Flu can precipitate cardiac events |
| Diabetes | Higher complication risk |
| Weakened immune system | Including treatment-related |
| Neurological conditions | Particularly those affecting swallowing or breathing |
| Care home residents | Age plus close-contact setting |
Digital Thermometer
Knowing whether there’s a fever, and how it’s tracking, is what a clinician will ask first
Unglamorous and genuinely useful. When you call a doctor about a possible flu, the first questions are when did symptoms start and is there a fever — and guessing is unhelpful, particularly with children. A working thermometer lets you answer both, and lets you track whether things are improving or not. Two points that matter more than the device. First, note the date and time symptoms began — that determines whether the antiviral window is still open. Second, and importantly: the absence of a fever does not rule out flu, and a normal temperature never overrides worrying symptoms. Breathing difficulty, chest pain, confusion, dehydration or a child who is unusually drowsy or hard to rouse need urgent assessment regardless of what the thermometer says. For babies under three months, any fever needs same-day medical advice.
- Write down when symptoms started — it decides the antiviral question
- No fever doesn’t rule out flu, and doesn’t override red-flag symptoms
- Any fever under 3 months needs same-day medical advice
- Contact a doctor early if you’re in a higher-risk group
- A thermometer informs a decision; it doesn’t make one
When flu becomes an emergency
In adults: difficulty breathing or shortness of breath; persistent chest or abdominal pain or pressure; persistent dizziness, confusion or difficulty rousing; seizures; not urinating; severe muscle pain; severe weakness or unsteadiness; symptoms that improve then return with worsening fever or cough.
In children: fast or laboured breathing; blue or grey lips or face; ribs pulling in with each breath; chest pain; severe muscle pain such that a child refuses to walk; dehydration — no urine for 8 hours, dry mouth, no tears when crying; not alert or interacting when awake; seizures; fever above 40°C; any fever in a baby under 12 weeks; worsening of an existing condition; and again, symptoms that improve then return worse.
The “improved then got worse” pattern is important — it can indicate a secondary bacterial infection such as pneumonia, and it’s frequently dismissed because the person appeared to be recovering.
Frequently asked questions
How bad was the 2024–25 flu season?
The CDC classified it as a high severity season overall and for every age group — the first since 2017–18. FluSurv-NET recorded 38,960 influenza-associated hospitalisations between 1 October 2024 and 30 April 2025 at a cumulative rate of 127.1, against a median of 62.0 across the 2010–11 to 2023–24 seasons and above the previous high of 102.9 in 2017–18. National estimates reached at least 46 million illnesses, 600,000 hospitalisations and 26,000 deaths.
How many children died?
289 — the highest US paediatric flu death toll since the 2009–10 H1N1 pandemic, and the highest for a non-pandemic season since paediatric flu deaths became nationally notifiable in 2004. That last comparison is the one that matters: this wasn’t a pandemic year with a novel virus, it was an ordinary seasonal influenza year, and it killed more American children than any other ordinary year in two decades of national reporting.
Why was it so severe?
Season severity comes from several factors interacting, and honest attribution after the fact is imprecise. The main drivers are which influenza viruses dominate — H3N2-dominant seasons have historically tended to be harder, especially on older adults — how well the vaccine matches circulating strains, existing population immunity from prior seasons, vaccination coverage, and a genuine element of chance in the mix and timing of strains. What’s certain is the outcome: the worst hospitalisation rate in fourteen seasons.
Does the flu vaccine actually work?
It works differently from how people often expect, and overselling it has damaged trust. Effectiveness varies by season and strain match, and it can be modest in a poorly matched year — vaccinated people do sometimes get flu. What it more reliably does is reduce severity: the likelihood of hospitalisation, intensive care and death, even when it doesn’t prevent infection. In a season with a hospitalisation rate double the historical median, that’s the benefit that counts. Get it in autumn, ahead of a typical February peak.
What are antivirals and when do they help?
Prescription medicines for influenza that work best when started early — ideally within 48 hours of symptoms beginning. They aren’t for everyone and are most valuable for people at higher risk of complications: older adults, young children, pregnancy, and people with chronic lung, heart, immune or neurological conditions. The practical implication is that “wait and see” can close the window. If you’re higher risk, contact a doctor early and say clearly when symptoms started.
When should I go to hospital?
Adults: difficulty breathing, persistent chest or abdominal pain, persistent dizziness or confusion, seizures, not urinating, severe weakness. Children: fast or laboured breathing, blue or grey lips, ribs pulling in with each breath, dehydration, not alert when awake, seizures, fever above 40°C, and any fever in a baby under 12 weeks. In both: symptoms that improve and then return with worsening fever or cough — that pattern can mean a secondary bacterial infection like pneumonia, and it gets dismissed precisely because the person seemed to be recovering.
Your checklist
- Know 2024–25 was high severity for every age group — first since 2017–18
- Hospitalisation rate 127.1 vs a 62.0 median, above 2017–18’s 102.9
- 289 paediatric deaths — highest non-pandemic total since 2004
- National estimates: 46m illnesses, 600,000 hospitalisations, 26,000 deaths
- Understand severity varies hugely between seasons (8.7 to 127.1)
- A mild season predicts nothing about the next one
- Vaccinate in autumn — peaks are typically early February
- Expect the vaccine to reduce severity, not guarantee prevention
- Note when symptoms start — antivirals work best within 48 hours
- If higher risk, contact a doctor early rather than waiting
- Healthy children are not exempt
- No fever doesn’t rule out flu
- Treat “better then worse” as a red flag
Medical disclaimer. This article is general information, not medical advice. Seek emergency care immediately for difficulty breathing, persistent chest or abdominal pain, persistent confusion or difficulty rousing, seizures, or — in children — fast or laboured breathing, blue or grey lips, ribs pulling in with each breath, dehydration, or not being alert when awake. Any fever in a baby under 12 weeks requires same-day medical advice. Symptoms that improve and then worsen may indicate a secondary bacterial infection and need assessment. Antiviral medicines are prescription-only and work best within 48 hours of symptom onset — contact a doctor early if you are in a higher-risk group. Discuss vaccination with your doctor or pharmacist.
Affiliate disclosure. Some links on this page are affiliate links. As an Amazon Associate we earn from qualifying purchases. The most effective measures described here — vaccination and early medical contact — are not products we sell.
