How much sun does the body really need?

Holick and colleagues exposed human skin to simulated sunlight and observed what happens to the precursor molecule. With longer irradiation, previtamin D3 production reached a plateau at around 10 to 15 percent of the original 7-dehydrocholesterol content. The excess was converted into two biologically inactive isomers, lumisterol and tachysterol.

For you this means something very practical: you cannot sunbathe your way into a vitamin D overdose. Beyond a certain point, no additional vitamin D is made, no matter how long you stay lying there. What keeps increasing from then on is only the damage.

The authors rank the limiting factors in this order: first the photochemical self-regulation, second pigmentation, third latitude.

Webb, Kline and Holick quantified exactly this. They exposed human skin and the precursor substance to sunlight on cloudless days in different locations and checked whether previtamin D3 was formed.

The result: in Boston at 42.2 degrees north, no previtamin D3 was formed from November to February. In Edmonton at 52 degrees north, this ineffective period stretched from October to March. Further south, at 34 and at 18 degrees north, the conversion worked even in midwinter.

Berlin lies at around 52.5 degrees north. That puts us in the same category as Edmonton. For roughly six months of the year, your skin cannot make vitamin D here, no matter how much time you spend outdoors.

In the nationwide German Health Interview and Examination Survey for Adults, 25-hydroxy vitamin D levels were measured in 6,995 people. The mean was 45.6 nanomoles per litre.

61.6 percent were below the frequently used threshold of 50 nanomoles per litre, and 30.2 percent were even below 30 nanomoles per litre. In winter, a quarter of participants had values below 30. A gradient by latitude was also detectable.

As independent factors for lower levels, the authors found: examination in winter, living further north, no supplementation, little sport, a high body mass index and high media consumption. That last point is essentially a measure of how much time someone spends indoors.

A research group in Aachen irradiated healthy volunteers over the whole body with UVA. Systolic and diastolic blood pressure fell by 11 percent, measured 30 minutes after irradiation, and the effect lasted for up to an hour.

In parallel, forearm blood flow rose by 26 percent and flow mediated dilation of the arm artery increased markedly. The drop in blood pressure correlated closely with the increase in nitrosated compounds in blood plasma.

Important for context: this is a small mechanistic study in healthy people, not a treatment study in high blood pressure. It shows that the route exists, not that it works as a therapy.

A group in Edinburgh around Richard Weller and Martin Feelisch took the mechanism further apart. 24 healthy volunteers received two standard erythema doses of UVA on the skin. Blood pressure fell, circulating nitrate fell, nitrite rose.

The decisive control experiment: a targeted dietary intervention designed to change blood nitrate levels had no influence on the circulatory response. The UVA related increase in forearm blood flow also occurred when nitric oxide synthase was blocked. Under the fluorescence microscope, the light sensitive nitric oxide store turned out to sit mainly in the upper epidermis.

In short: the effect comes from a store in the skin, not from an enzyme and not from vitamin D.

Weller and colleagues analysed the blood pressure of 342,457 dialysis patients in 2,178 US centres over three years. In these people, blood pressure is measured regularly and in a standardised way anyway. The radiation and temperature data came from the US weather service.

Temperature, UVA and UVB were each linearly and inversely associated with systolic blood pressure. Crucially, the association with UV radiation remained statistically significant even after ambient temperature was accounted for.

The authors phrase it as a question, not as a statement: insufficient sunlight may be an overlooked risk factor for high blood pressure. An observational study cannot prove that. But it can show that the trail stays consistent when the most obvious confounder is removed.

A review from Basel and Oxford sums up the state of knowledge. The master clock, the suprachiasmatic nuclei in the hypothalamus, is synchronised to the 24-hour day through a direct pathway from the retina. Artificial light in the evening can shift this synchronisation.

Conversely, light can be used as an intervention with few side effects to influence sleep, mood and general wellbeing. The practically most important point: it is not only about too much light at night, but at least as much about too little light during the day.

An analysis of the UK Biobank examined how much time people spend in daylight and how that relates to mood, sleep and the body clock. The median was 2.5 hours of daylight per day.

Every additional hour outdoors was linked with a lower probability of a depressive episode over the lifetime, less frequent antidepressant use, less frequent low mood, more reported happiness, easier waking, less tiredness, fewer problems falling asleep and an earlier chronotype.

The longitudinal analyses pointed in the same direction, with smaller effects. This is an observational study. People who spend a lot of time outdoors also move more and are socially more active. The authors therefore call their finding an indication of an environmental risk factor, not a proof.

The same German population study examined the association between vitamin D level and depressive symptoms in 6,331 people, separately for summer and winter.

An association only appeared in summer: in the highest vitamin D quartile, the score on the depression questionnaire was 0.73 points lower than in the lowest. In winter, no association was found.

The authors draw a remarkable conclusion from this: precisely because the association is stronger in summer, a low vitamin D level might be a consequence rather than a cause of depression. People who are depressed go outside less. That is an honest reading of their own data, and it runs counter to many marketing promises.

A systematic review with meta-analysis evaluated seven randomised studies with a total of 397 people that compared light therapy directly with antidepressants, in moderate to severe depressive episodes.

Light therapy alone did not differ statistically from drug treatment. The combination of both was superior to medication alone, including in the subgroup with non-seasonal depression.

That is a strong argument for taking light seriously as a biological quantity in its own right. And at the same time it shows that we are talking about light here, not about UV and not about vitamin D.

A systematic meta-analysis pooled the relative risks for malignant melanoma from 57 observational studies.

Two factors stood out clearly: intermittent sun exposure and a history of sunburn. High occupational sun exposure, meaning chronic daily exposure, was by contrast even inversely associated with melanoma risk.

The authors stress that it was precisely the methodologically better studies that supported the intermittent exposure hypothesis. The risk of sunburn also rose in studies from higher latitudes. In other words, where people suddenly get a lot of sun in summer, on skin that is not used to it.

Armstrong and Kricker summarised the epidemiological evidence for all three types of skin cancer. Their conclusion: incidence is higher in fair skinned, sun sensitive people, it rises with local solar radiation, and the highest densities are found on the most sun exposed parts of the body.

The assignment: squamous cell carcinoma mainly to total and occupational exposure, melanoma and basal cell carcinoma mainly to non-occupational leisure exposure.

Their practical conclusion is remarkable, because it is not: less sun in general. It is: sun protection should avoid an increase in intermittency, and protection should start as early in life as possible.

The Norwegian Women and Cancer study asked 169,768 participants about their frequency of sunburn in childhood, adolescence and adulthood, and followed them for up to 19.5 years.

The group with consistently high sunburn frequency had a clearly increased risk compared with the group with consistently low frequency, both for melanoma and for squamous cell carcinoma. The group that had many sunburns earlier and few later also remained elevated.

That is exactly the decisive message: sunburns in childhood count in a particular way, and they cannot be optimised away later.

In the Swedish Melanoma in Southern Sweden cohort, 29,518 women between 25 and 64 years of age were followed for 20 years. Sun habits and numerous possible confounders were recorded.

Women with active sun habits had lower mortality from cardiovascular disease and from non-cancer, non-cardiovascular causes. Because they lived longer, the relative share of cancer mortality rose among them. The life expectancy of the sun avoiders was 0.6 to 2.1 years below the group with the highest sun exposure.

The frequently quoted sentence from the authors: non-smokers who avoided the sun had a life expectancy similar to smokers in the group with the highest sun exposure. That is pointedly phrased and should be read with caution.

The classic work on this comes from Clemens and colleagues in 1982. They irradiated two lightly pigmented and three strongly pigmented volunteers with one minimal erythema dose of UV.

In the fair skinned participants, the vitamin D level in the blood rose up to sixtyfold within 24 to 48 hours. In the strongly pigmented participants, the same dose did not change the level significantly. Only six times the dose led to a comparable rise there.

The sample is tiny, that has to be said. The direction of the finding has, however, been reproduced many times since.

A London group re-measured the effect in 102 healthy volunteers with skin types II to VI. All received the same suberythemal UV dose on 85 percent of the body surface, five times at intervals of three to four days.

Result: the rise in the 25-hydroxy vitamin D level was linear and highly significant in all groups. Only skin type II had a significantly steeper slope than the others. Between the extremes, skin type II and skin type VI, the inhibiting factor from melanin was around 1.3 to 1.4.

The authors conclude from this: the braking effect of melanin on vitamin D production is small compared with its protective effect against sunburn. Small enough to be barely noticeable during a week of holiday. Large enough to explain the epidemiological differences over a whole lifetime in Northern Europe.

Matsuoka and colleagues gave 32 untanned fair skinned people with skin type III graded, suberythemal UVB doses between 3 and 27 millijoules per square centimetre and then measured vitamin D3 in serum.

The rise followed the dose in an exponential relationship. The lowest dose that significantly raised the vitamin D3 level was 18 millijoules per square centimetre. That roughly corresponds to the threshold at which previtamin D3 is formed from the precursor molecule in the test tube.

Important: these were suberythemal doses. In other words, below the amount that produces reddening. The authors conclude from this that at northern latitudes in winter, the radiation does not even reach this threshold.

This question was examined in a very well designed study. Polish volunteers spent a holiday week on Tenerife at a very high UV index. Two groups were given sunscreens with sun protection factor 15 and instructions for correct application, one group applied them at their own discretion, and a control group stayed at home at 51.8 degrees north.

Both instructed groups got no sunburn, the group using their own discretion did. Even so, the 25-hydroxy vitamin D level rose highly significantly in the instructed groups, by 13.0 and 19.0 nanomoles per litre. Those who stayed at home lost 2.5 nanomoles per litre in the same period.

An intriguing side finding: the product with high UVA protection allowed more vitamin D production than the one with low protection, because it lets through relatively more UVB. The choice between protection and vitamin D is therefore less sharp than many assume.