Sunscreen is one of the few consumer products with a genuine effect on cancer risk, and one of the most consistently misused. The two commonest errors point in opposite directions: not using it, and using it as permission to stay out longer.
1. The arithmetic nobody is told, and it is the whole chapter
SPF is measured in a laboratory at 2 mg of product per square centimetre of skin. In real use, people apply between a quarter and a half of that.
Halving the amount does not halve the protection — it collapses it
- The relationship is exponential, not linear. As a working approximation, applying half the tested thickness gives roughly the square root of the labelled SPF.
- So an SPF 50 applied at half thickness behaves like about SPF 7. At a quarter thickness it is worse again.
- Which is why "how much" beats "which one". A cheap SPF 30 applied properly outperforms an expensive SPF 50 applied the way most people apply it, by a wide margin.
- The amount for a whole adult body is about 35 ml — six to seven teaspoons, or roughly a shot glass. For the face and neck alone, about two fingers' length of product squeezed onto the index and middle fingers. Almost nobody uses this much, and seeing it measured out once is genuinely surprising.
- Reapplication is not optional, and it is not mainly about the product degrading — it is about replacing what has been rubbed, sweated and washed off. Every two hours, and immediately after swimming or towelling.
2. SPF describes sunburn. It says nothing about the rest.
| UVB | UVA | |
|---|---|---|
| What it does | Burns. Directly damages DNA in the outer skin | Penetrates deeper. Drives photoaging, collagen breakdown and pigmentation, and contributes to skin cancer |
| Measured by | SPF | Not by SPF at all. In the UK and EU, by the UVA-in-a-circle logo, and by the star rating on some brands |
| Gets through glass? | Largely no | Yes — car and office windows. Relevant to drivers and to anyone sitting by a window |
| Varies with weather? | Falls in cloud | Much less. Cloud lets a large proportion through |
- The UVA circle is a floor, not a grade. It means UVA protection is at least one-third of the SPF — the EU and UK minimum. A product can carry it and still be mediocre.
- The star rating is the more useful signal where it appears. It measures UVA protection relative to SPF, so five stars on an SPF 50 is a genuinely different product from five stars on an SPF 15.
- PPD is the underlying number if a brand publishes it — higher is better, and anything above about 30 on an SPF 50 is strong.
- Above SPF 30 the returns are small but not nothing. SPF 30 blocks about 97% of UVB and SPF 50 about 98%. That sounds trivial and is not, because what matters is the fraction getting through: 3% against 2% is a third less. Given that everyone under-applies, buying SPF 50 is a sensible hedge rather than vanity.
3. The filters — and a full account of the safety question
This is where most of the online anxiety sits, and where the picture is more boring than it looks.
| Filter | Status |
|---|---|
| Oxybenzone (benzophenone-3) | The one with a real case against it. FDA maximal-usage studies found blood levels far above the threshold at which the agency asks for further toxicology; there are endocrine-activity signals in laboratory work and coral concerns. The EU cut it to 2.2% in body products in 2022 and the UK followed only in January 2026. Products must carry the words “Contains Benzophenone-3” unless it is present at 0.5% or less purely to protect the formulation rather than as a UV filter — so the warning on a label tells you it is doing sunscreen work. Easy to avoid, and worth avoiding |
| Octocrylene | Widely used, mainly to stabilise avobenzone. Degrades slowly to benzophenone, which is the basis of the concern. Contact allergy is more common than with most filters. Avoidable, and there is little reason to seek it out now that better stabilisers exist |
| Homosalate | A weak UVB filter used in quantity. The EU restricted it to 7.34% in face products; the UK reviewed it separately and deliberately kept 10% across all products, having judged it safe at that level. A real regulatory divergence, and a reasonable one to be relaxed about |
| Octinoxate | UVB only, and photounstable — it degrades in sunlight, which is a performance problem before it is a safety one |
| Avobenzone | The best broad UVA absorber among the older filters, and unstable on its own. Fine in a modern formulation that stabilises it; poor in an old one that does not |
| Zinc oxide and titanium dioxide | Mineral filters. Very well tolerated, no meaningful absorption, and the usual choice for babies and reactive skin. The trade-off is cosmetic — a white cast, worse on deeper skin tones — and zinc's UVA coverage is good but its SPF-per-gram is low, so high-SPF mineral products are thick |
4. The modern filters, and a piece of news that is not UK news
A newer generation of filters is photostable, covers UVA properly, stabilises the older filters alongside them, and is poorly absorbed because the molecules are large.
| On a UK label | Also known as | What it does |
|---|---|---|
| Bis-ethylhexyloxyphenol methoxyphenyl triazine | Bemotrizinol, Tinosorb S, Parsol Shield | Broad-spectrum, absorbing both UVB and UVA in one molecule. Highly photostable, and stabilises avobenzone alongside it |
| Methylene bis-benzotriazolyl tetramethylbutylphenol | Bisoctrizole, Tinosorb M | A different compound — particulate, water-dispersible, absorbing and scattering. Frequently confused with the one above |
| Diethylamino hydroxybenzoyl hexyl benzoate | Uvinul A Plus, DHHB | A stable, strong, long-UVA filter |
| Methoxypropylamino cyclohexenylidene ethoxyethylcyanoacetate | Mexoryl 400 | Covers the far end of UVA, to about 400 nm, which most products miss |
The headline everyone is reading is an American one
- On 9 June 2026 the US FDA approved bemotrizinol — its first new sunscreen active ingredient in about 25 years — with products expected on American shelves from August 2026.
- It has been in European and UK sunscreens since around 1999. The news is that the United States has caught up on one filter, not that something new has been invented. Bisoctrizole, Uvinul A Plus and the Mexoryls are still not available there.
- So for a UK reader the practical implication is small and reassuring: you already have access to all of it, and have done for a quarter of a century. The advice is simply to buy a product that contains these filters, which is easy and does not require importing anything.
- And a correction worth making, because the two get swapped constantly: bemotrizinol and bisoctrizole are different molecules — Tinosorb S and Tinosorb M respectively. Only the first has FDA approval.
5. What to actually buy in the UK
Reading a label is more reliable than any recommendation, so the criteria first: SPF 30 or 50, a UVA circle and ideally five stars, at least one modern filter from the table above, and a texture and price you will genuinely use in the quantity described in section 1. That last clause does more work than the others.
| Product | Why it is here |
|---|---|
| Altruist Dermatologist Sunscreen SPF 50 | The one to buy if you buy one thing. Developed by a UK consultant dermatologist and skin cancer specialist, it uses modern filters, carries five-star UVA with a published PPD in the fifties, is fragrance-free, and costs a few pounds for 100 ml. The low price is the clinical argument, not a caveat — the failure mode of sunscreen is applying too little, and cheap product bought in quantity is the most direct fix for that |
| La Roche-Posay Anthelios UVMune 400 | The most complete UVA coverage widely sold, being one of the few containing Mexoryl 400 alongside bemotrizinol and others, with a published PPD in the mid-forties. Cosmetically excellent and considerably more expensive. Worth it for daily facial use, particularly with melasma or pigmentation, and hard to justify for coating a family at the beach |
| Boots Soltan five-star range | Own-brand, cheap, everywhere, and formulated with modern filters including bemotrizinol. The five-star UVA rating originated with Boots. A perfectly good default |
| Any mineral SPF 50 for babies and reactive skin | Zinc-based, for under-sixes, eczema, rosacea or a history of reacting to sunscreen. Expect a white cast and accept it |
Checked August 2026, and formulations change without the packaging changing. Filters get reformulated, ranges get renamed, and a product that was excellent two years ago may not be the same product today. The criteria in the paragraph above will outlast this list — read the label rather than trusting the name, and if what you find no longer matches, the criteria win.
This site sells nothing and takes nothing from anyone named above. The list is short deliberately: the differences between competent modern sunscreens are small next to the difference between using enough and not. If a product you already own has SPF 30+, a UVA circle and no oxybenzone, the best thing you can do is use more of it rather than replace it.
6. The false security problem, which is real and measurable
- People given higher-SPF sunscreen stay out longer. This has been shown experimentally, and it is the mechanism by which sunscreen can fail to reduce — or in some studies appears to increase — total UV dose. The product works; the behaviour it licenses undoes it.
- Sunscreen is the weakest of the three protections and the one everyone reaches for first. Shade and timing beat clothing, and clothing beats sunscreen. Sunscreen is for the skin you cannot cover and the hours you cannot avoid.
- Not burning is not the same as not being damaged. UVA does not cause the burn, so the absence of redness says nothing about the exposure that drives photoaging and contributes to cancer.
- Sunbeds are in a different category and no amount of protection makes them reasonable.
7. The other half — why total avoidance is not the safe option either
Everything above is about protecting skin. This section is the counterweight, and it is here rather than in the register because it needs the space.
The part that gets left out, and it is the more interesting half. Vitamin D is usually treated as the reason sunlight is good for you. The evidence points the other way: in a pooled analysis of 25-hydroxyvitamin D data from around 849,000 people, low levels were strongly associated with death from cardiovascular disease, cancer and all causes — and the same authors noted that supplementing vitamin D has not been shown to prevent any of it. That combination is the signature of a marker rather than a mechanism. Blood vitamin D looks like a good proxy for how much sun someone gets, and the sun appears to be doing something else as well.The leading candidate is nitric oxide. Skin holds a substantial store of nitrite and nitrate, and UVA mobilises it into the circulation as nitric oxide, which dilates blood vessels and lowers blood pressure. Human exposure studies have shown the fall, and ruled out warmth and dietary nitrate as the explanation. It is independent of vitamin D entirely, which is why no supplement reproduces it. Alongside that: beta-endorphin release in the skin, and the circadian signal from morning light, neither of which comes in a capsule.
In the Swedish Melanoma in Southern Sweden cohort, women with the lowest sun-exposure habits had the highest cardiovascular and all-cause mortality, and the authors estimated that avoiding the sun was a risk factor for death of a magnitude similar to smoking. That sentence gets quoted a great deal and deserves its caveats: it is observational, in a fair-skinned Nordic population where sun-seeking also tracks affluence, holidays, outdoor activity and better health to begin with, and reverse causation is a real possibility because unwell people go outside less. Moderate for the direction, and not the settled equivalence the headline implies.
What none of this licenses. It is not an argument for burning, for sunbeds, or for abandoning shade at midday — the melanoma evidence is unchanged and stronger than any of the above. What it does argue against is total avoidance, which is a real behaviour with real costs, and it explains why the advice is a dose rather than a prohibition: short, regular, non-burning exposure, taken incrementally, eyes protected, and stopping well before any colour change.
Sunscreen has its own chapter now — how much to apply, what SPF does not measure, and a full account of the filter safety question. In short: not all are equal. The chemical filters oxybenzone, octinoxate, octocrylene and homosalate are absorbed into the bloodstream above the level at which the FDA asks for further safety data — oxybenzone by far the most, and it's been detected in blood, urine and breast milk with hormone-like activity. Prefer mineral sunscreens (non-nano zinc oxide / titanium dioxide), which the FDA treats as safe and which aren't meaningfully absorbed. This matters more for children: they absorb more per kilo, so use mineral sunscreen on kids, avoid spray sunscreens (inhalation and patchy cover), keep babies under 6 months out of direct sun rather than relying on cream, and lean on hats, shade and UV clothing. (Keep perspective — absorption alone isn't proven harm, and protecting against UV still wins; the point is to choose the cleaner filter.)
How this sits with the rest of the chapter, since it can read as a contradiction. It is not an argument against sunscreen and it is certainly not an argument for burning — the melanoma evidence is unchanged and is stronger than anything above. It is an argument against total avoidance, which is a real behaviour with real costs. The target is exposure without burning, not maximal blocking, and in practice normal sunscreen use does not prevent it, because people apply too little and miss too much.
8. What still matters more than any of this
- Know your own skin. A new mole, or an existing one that changes in size, shape, colour or outline, itches or bleeds, needs seeing. Melanoma caught early is usually curable; caught late it is not, and that gap is larger than anything a product does.
- Sunburn in childhood matters for decades. Protecting children is a different order of priority from protecting yourself.
- Risk is not uniform. Fair skin that burns easily, many moles, a family history of melanoma, immunosuppression, or a previous skin cancer all move the calculation.
- And it does not disappear on darker skin. Skin cancer is less common but is diagnosed later and does worse, and it appears in places people do not check — palms, soles, under nails.