Coffee gets the headlines, the wellness influencers and the morning ritual status. But when researchers looked at the back of the eye in more than 35,000 people, coffee showed nothing. Tea did.
A new analysis of the UK Biobank, published in Ophthalmic Epidemiology in 2026, found that tea drinkers had a smaller "retinal age gap", a biological ageing marker. The signal was strongest in people who drank more than four cups a day. Coffee drinkers showed no such link.
Before you swap your espresso for a teapot, the details matter. The finding is real, but the effect is small and the study design limits what it can tell us.
What the researchers actually did
The team, led by researchers at the Centre for Eye Research Australia at the University of Melbourne, analysed 35,740 UK Biobank participants. Their mean age was 56.8 years, and 55.7% were men. About 21.5% said they never drank coffee, and 14.4% never drank tea.
Participants reported their average daily coffee and tea intake over the past year on a touchscreen questionnaire. The researchers then compared those habits with each person's retinal age gap, using linear regression models adjusted for demographic factors. They also used restricted cubic spline models to check whether the relationships were curved rather than straight.
Wait, what is a retinal age gap?
The retina is the light-sensing tissue at the back of the eye. It is one of the few places where doctors can see small blood vessels and nerve tissue directly and non-invasively. Deep‑learning models can be trained to estimate a person's age from a retinal photograph.
The retinal age gap is the predicted retinal age minus actual chronological age:
- A positive gap means your retina looks older than your birth certificate says, which is read as accelerated biological ageing.
- A negative or smaller gap means your retina looks relatively younger.
It is a research marker of biological ageing, not a diagnosis. It is not a test you can currently use to judge your own health.
The headline result: tea yes, coffee no
After adjusting for demographic covariates, the results split cleanly:
- Coffee: no association with retinal age gap.
- Tea: an association with a smaller retinal age gap (β = -0.03, 95% CI -0.04 to -0.01, p = 0.006).
- Heavy tea drinking: people drinking more than 4 cups a day showed a larger difference (β = -0.22, 95% CI -0.36 to -0.08, p = 0.003).
The spline models found no significant non‑linear patterns for either drink (P non‑linear = 0.329 for coffee, 0.273 for tea). The tea relationship is better described as a steady, roughly linear trend than as a sweet spot or a threshold.
How big is the effect, really?
Honest answer: small. The retinal age gap is measured in years, so a coefficient of -0.03 amounts to a few days of retinal age per unit of tea. The heavy‑drinker figure of -0.22 is roughly 2.6 months. The abstract does not spell out the exact scaling of the lower coefficient, so treat the arithmetic as a rough guide.
That is a statistically detectable difference in a large sample. It is not a dramatic reversal of ageing. Nobody should expect a few cups of tea to knock years off their eyes.
Why this is association, not proof
This was a cross‑sectional analysis, a snapshot at one point in time. That brings several caveats:
- Direction is unclear. The data cannot show that tea slowed ageing. Healthier or younger‑ageing people might simply drink more tea.
- Confounding is possible. Only demographic covariates are noted in the abstract. Lifestyle, diet and socioeconomic patterns tied to tea habits could play a role.
- Self‑reported intake. A questionnaire about the past year is a rough measure.
- Unspecified tea type. The abstract does not report results by black, green or herbal tea, so no claim about a "best" tea can be drawn from it.
- A single cohort. UK Biobank participants are not necessarily representative of everyone, and tea culture varies widely around the world.
The authors conclude that tea is "a dietary factor linked to" biological ageing as indexed by retinal age gaps. The wording is linked, not causal.
Practical takeaways
Don't ditch coffee over this. The study did not find coffee harmful for retinal ageing. It found no association either way after adjustment.
Tea type and caffeine. The paper's abstract does not compare tea varieties or caffeine content. Tea generally contains less caffeine per cup than coffee, though the amount varies with the leaf, brew time and strength. If you want to add tea without adding much caffeine, decaffeinated or herbal options exist, but this study gives no evidence on whether they carry the same association.
How to drink more tea without sugar. Habit matters more than hype, and sugar can undercut any benefit. A few simple approaches:
- Swap the second or third coffee of the day for tea, rather than adding to your total intake.
- Try a lighter brew or a shorter steep time if bitterness makes you reach for sweetener.
- Add flavor without sugar: a slice of lemon, a cinnamon stick, fresh mint or a splash of milk.
- Keep unsweetened iced tea in the fridge for warm days.
Who should watch caffeine. Drinking more than four cups a day is a lot of tea, and the largest effect appeared at that level. People who are pregnant, sensitive to caffeine, prone to palpitations or anxiety, have sleep problems, or are on medications that interact with caffeine should check with a clinician before raising their intake. This is general guidance rather than a finding of the paper.
The bottom line
In 35,740 adults, tea drinking was linked to a slightly younger‑looking retina, most clearly at more than four cups a day. Coffee showed no link. It is an intriguing signal for ageing research, and a reason to take tea seriously as a variable. It is not a reason to believe tea reverses ageing. Randomized or longitudinal studies would be needed to find out whether tea changes how the eye ages, or whether it just travels with people who age well.