Sunscreen is one of the most studied skincare products in existence. Decades of randomised controlled trials, population studies, regulatory research.
And yet most people using it daily are working from assumptions — about what it does, how much to use, when it's actually needed — that the evidence doesn't support.
This isn't a warning. It's an invitation to understand what you're putting on your skin, why the number on the bottle is a best-case scenario you'll never reach, and why the cloudy October morning is not the free pass it feels like.
The sun you don't feel
UVB: the one you notice
UVB occupies the 290–320 nm range of the ultraviolet spectrum. It's responsible for the obvious, immediate damage: sunburn, redness, that specific sensation of having stayed out too long.
UVB intensity varies with season, time of day, and weather. It is largely blocked by clouds and almost entirely blocked by ordinary glass. On an overcast winter day, your UVB exposure is genuinely low.
This is where most people stop thinking about sun protection — and exactly where the science asks them not to.
UVA: the one you don't
UVA occupies the 320–400 nm range. It penetrates deeper into the dermis, works without any visible or physical signal, and is present at significant levels year-round — on cloudy days, in winter, through the window next to your desk.
The shorthand most people know: UVB burns, UVA ages. Not wrong. But dangerously incomplete.
UVA is also carcinogenic. Tanning beds emit up to 10–15 times more UVA than natural sunlight. The World Health Organisation classified them as a known human carcinogen, specifically because of a dose-response relationship with melanoma — particularly in young women.
UVA causes DNA lesions, collagen degradation, and photoimmunosuppression. No sunny day required.
What actually stops UV — and what doesn't
Clouds are deceptive.
They block heat and visible light effectively — which is why an overcast day feels cool and safe. UV behaves differently. A fully overcast sky still transmits anywhere from 30% to 70% of UV radiation.
Broken or partial cloud cover can actually increase UV levels at the surface by up to 20–30%, as rays scatter off cloud edges. ¹
The sky that looks protective often isn't.
Glass tells a different story.
Standard window glass — homes, offices, car side windows — blocks UVB effectively. It does not block UVA. Smooth annealed glass transmits up to 74% of incident UVA. ² Car windshields (laminated glass) block around 98% of UVA; side and rear windows (tempered glass) transmit roughly half.
The clinical evidence is striking. Researchers enrolled ten patients with asymmetrical facial photoaging — measurably more wrinkles, roughness, and skin laxity on one side — on the side chronically exposed through a window over years of daily life. The subjects had varied occupations. The pattern held regardless. ³
Skipping SPF on a grey November morning because you're not going outside still means accumulating UVA through every window you sit near.
What SPF actually measures — and what it doesn't
SPF measures one thing only: UVB.
Sun Protection Factor measures how much UVB is required to cause a just-visible redness on protected versus unprotected skin. The scale is not linear:
- SPF 15 blocks around 93% of UVB
- SPF 30 blocks 97%
- SPF 50 blocks 98%
- The gap between SPF 30 and SPF 100: roughly 2 percentage points
The number on the bottle says nothing about UVA — the radiation that passes through your window all day.
The dose problem.
SPF is tested at 2 mg per cm² of skin — roughly a quarter teaspoon for the face alone. In practice, people apply between 0.39 and 1.0 mg/cm². The lower end of that range is less than a quarter of the test dose. ⁴
The relationship between dose and protection follows an exponential curve. A clinical study tested SPF 70 and SPF 15 at full lab dose and real-world half dose, after exercise and water exposure: ⁵
- SPF 70 at full dose → in vivo protection: SPF 64
- SPF 70 at half dose → in vivo protection: SPF 26
- SPF 15 at full dose → in vivo protection: SPF 13
- SPF 15 at half dose → in vivo protection: SPF 7
Using half the product doesn't give half the protection — it destroys it.
SPF 30 and SPF 50 may not deliver adequate protection at typical consumer doses. SPF 70 and above can compensate for under-application. ⁶
If you want real-world protection equivalent to SPF 30, apply SPF 60.
Reading the label: SPF, PA+++, and the UVA gap
The broad-spectrum problem.
In Europe, a product's UVA protection factor must be at least one third of its SPF. An SPF 50 product needs a UVA-PF of at least 16.7 to carry a broad-spectrum label. That's a floor, not a ceiling.
A PMC study testing 32 commercially available sunscreens found that real-world UVA protection was lower than labels indicated in many products — and that the EU 1/3 ratio allows certification with considerably weaker UVA coverage than ideal.
The PA+++ rating: a direction, not a measurement.
PA+++ is a Japanese regulatory standard based on Persistent Pigment Darkening measurements. PA+ through PA++++ indicates increasing UVA protection, with each plus corresponding to a higher PPD value.
More granular than the EU broad-spectrum label — but still a regulatory classification, not a clinical measurement. Two products with identical PA+++ ratings can have meaningfully different actual UVA protection.
In short: SPF is your most reliable data point — UVB only, lab conditions. Look for the highest UVA-PF disclosed. Treat PA as a direction, not a specification.
Mineral vs chemical: what the evidence actually says
How they work.
Mineral filters — zinc oxide and titanium dioxide — sit on the skin's surface and physically scatter and reflect UV radiation.
Chemical filters — oxybenzone, avobenzone, octocrylene, homosalate, octisalate, octinoxate — absorb into the upper skin layers and convert UV energy into heat.
Zinc oxide covers both UVA and UVB across a broad spectrum. Titanium dioxide offers stronger UVB but weaker UVA coverage. Together: comprehensive broad-spectrum protection.
Human health risks from mineral filters are extremely low. Less than 0.03% of zinc nanoparticles penetrate beyond the outermost stratum corneum layer in vitro. ¹⁰ One legitimate concern: spray and powder formats. Inhaled titanium dioxide nanoparticles carry a distinct risk profile from topically applied ones. ¹¹
What the FDA found.
In 2019 and 2020, the FDA published two randomised clinical trials in JAMA on the systemic absorption of chemical UV filters under maximal use conditions. ⁷˒⁸ All six tested ingredients were absorbed into the bloodstream and exceeded the FDA threshold requiring additional safety studies.
Oxybenzone was absorbed at 50–100 times the concentration of the others. Homosalate and oxybenzone remained above safety thresholds 21 days after the last application.
This doesn't mean chemical sunscreens are unsafe. It means the safety data needed to confirm they are safe — under the revised framework accounting for systemic absorption — doesn't exist yet. Only zinc oxide and titanium dioxide hold a positive GRASE designation from the FDA.
The European Commission has separately concluded that homosalate and oxybenzone are not safe at US-permitted concentrations, and has set lower limits for EU products.
For most adults, the risk-benefit calculation still favours use. Long-term RCTs show sunscreen reduces squamous cell carcinoma incidence by 40% and measurably reduces melanoma risk. ⁹ But for pregnant women, parents of young children, or daily full-body users, the preference for mineral formulations is scientifically grounded — not irrational.
The oil mythology
Where the SPF 28–50 claim comes from.
The raspberry seed oil figure originates from a 2000 lab study that measured optical UV transmission and noted a passing resemblance to titanium dioxide preparations. No SPF testing on human volunteers. The oil was hexane-extracted — not the cold-pressed version available to consumers.
One uncontrolled lab comparison became "SPF 28–50" across a thousand blogs.
The carrot seed oil figure traces back to an Indian study that tested a commercial sunscreen containing carrot seed oil alongside zinc oxide and other ingredients. The combined product had an SPF of 38–40. That SPF was attributed to the carrot seed oil alone — which is not what the study showed.
What the actual numbers are.
A 2021 PubMed study tested the five most promoted oils in vivo, on human subjects, using ISO-standard methodology: ¹³
- Carrot seed oil: SPF 2.5
- Coconut oil: SPF 1.2
- Raspberry seed oil: SPF 2.6
- Rosehip seed oil: SPF 2.6
- Wheat germ oil: SPF 2.8
The remaining eight oils tested showed no measurable SPF at all.
An SPF of 2.6 blocks 62% of UVB. SPF 30 blocks 97%. That's not a minor shortfall — it's the difference between meaningful protection and effectively unprotected skin.
A note on DIY zinc oxide.
Zinc oxide is a legitimate UV filter. The problem is formulation. Commercial sunscreens undergo stability testing, homogeneity validation, particle size verification, and photostability assessment.
A zinc oxide powder stirred into coconut oil at home can be tested for none of these. The resulting SPF could be 5 or 40 from the same batch. There's no way to know.
What the protocol actually looks like
The 2-hour reapplication rule is so embedded in sun safety messaging that questioning it feels irresponsible. The evidence is more nuanced.
A 2018 clinical trial tested real sunscreens on real subjects after 30 minutes of exercise and 80 minutes of water immersion. SPF decreased by only 15–40% over 8 hours. ⁵ Reapplication intervals may reasonably be longer than currently recommended — when sunscreen hasn't been physically removed.
Researchers at the University of Copenhagen proposed a simpler protocol for daily non-beach use: apply before sun exposure, reapply once within the first hour. ⁴ Once — not every two hours — to correct the inevitable uneven first application.
The essentials
- SPF 50 or higher — not because SPF 30 vs SPF 50 is dramatic in labs, but because under-application is universal and higher SPF compensates. SPF 70+ provides the most meaningful real-world buffer
- Apply more than you think — a quarter teaspoon for the face alone. Most people use a quarter of that
- Reapply after sweat, swimming, or towel-drying — not mechanically every two hours on a normal day
- Daily use: mineral filters — no open questions on systemic absorption
- Beach or sport: water-resistant formulas — they outperform anything homemade, regardless of filter type
- Grey day, near a window: SPF on the face — UVA doesn't need sunshine to accumulate