Sunshine is good medicine. The health benefits of ultraviolet-B induced vitamin D production

Authors

  • W B Grant,

    Corresponding author
    1. Sunlight, Nutrition and Health Research Center (SUNARC), 2107 Van Ness Avenue, Suite 403B, San Francisco, CA 94109, USA
    Search for more papers by this author
  • R C Strange,

    1. Clinical Biochemistry Research Laboratory, Keele University School of Medicine, North Staffordshire Hospital, Stoke-on-Trent, Staffordshire, England
    Search for more papers by this author
  • C F Garland

    1. Department of Family and Preventive Medicine, University of California, San Diego, 9500 Gilman Drive, 0631C, La Jolla, CA 93093, USA
    Search for more papers by this author

William B Grant, 1745 Pacific Avenue, Apt. 503, San Francisco, CA 94109, USA. E-mail: wbgrant@infionline.net

Summary

Most public health statements regarding exposure to solar ultraviolet radiation (UVR) recommend avoiding it, especially at midday, and using sunscreen. Excess UVR is a primary risk factor for skin cancers, premature photoageing and the development of cataracts. In addition, some people are especially sensitive to UVR, sometimes due to concomitant illness or drug therapy.

However, if applied uncritically, these guidelines may actually cause more harm than good. Humans derive most of their serum 25-hydroxycholecalciferol (25(OH)D3) from solar UVB radiation (280–315 nm). Serum 25(OH)D3 metabolite levels are often inadequate for optimal health in many populations, especially those with darker skin pigmentation, those living at high latitudes, those living largely indoors and in urban areas, and during winter in all but the sunniest climates. In the absence of adequate solar UVB exposure or artificial UVB, vitamin D can be obtained from dietary sources or supplements.

There is compelling evidence that low vitamin D levels lead to increased risk of developing rickets, osteoporosis and osteomaloma, 16 cancers (including cancers of breast, ovary, prostate and non-Hodgkin's lymphoma), and other chronic diseases such as psoriasis, diabetes mellitus, hypertension, heart disease, myopathy, multiple sclerosis, schizophrenia, hyperparathyroidism and susceptibility to tuberculosis.

The health benefits of UVB seem to outweigh the adverse effects. The risks can be minimized by avoiding sunburn, excess UVR exposure and by attention to dietary factors, such as antioxidants and limiting energy and fat consumption. It is anticipated that increasing attention will be paid to the benefits of UVB radiation and vitamin D and that health guidelines will be revised in the near future.

Introduction

Solar ultraviolet radiation (UVR) has well-known roles in the aetiology of basal cell carcinoma (BCC) and squamous cell carcinoma (SCC),1 immune system suppression,2,3 premature ageing of the skin,4–6 and cataract formation.7 However, the beneficial effects for human health are less well recognized. The observation of lighter human skin pigmentation with increasing latitude provides the clue that sunlight is beneficial. The current hypothesis on the evolution of skin pigmentation in ancestral peoples is that the amount of melanin in the skin as a function of latitude is a careful balance between opposing requirements of the skin. On the one hand, the skin must be dark enough to reduce the risk of melanoma and other skin cancers and prevent the destruction of folic acid. On the other hand, the skin must be light enough to permit the photoinitiation of vitamin D production.8 Vitamin D is generated in humans by the action of UVB radiation on subcutaneous 7-dehydrocholesterol (7-DHC) into pre-vitamin D3, after which it undergoes thermal conversion to 25-hydroxycholecalciferol (25(OH)D3).9 If there were not such trade-offs between different functions of the skin, all humans would be likely to have similar pigmentation. Such evolutionary pressures on skin pigmentation were exerted at a time when human populations spent substantial parts of the day outdoors. At present, the proportion of the workforce with outdoor jobs is relatively small, and UVR exposure is often obtained from recreation, which tends to involve shorter exposures.

This paper outlines what is known about the health benefits of UVB radiation and put them into perspective with the health risks of UVR exposure.

Vitamin D reduces the risk of certain diseases

The recognition that there are important health benefits from solar UVB radiation through production of vitamin D has been slow in coming. It was not realized until the 1920s that rickets was a disease related to insufficient vitamin D.10 In the 1960s UVB was found to play a role in heart disease,11 and it was shown to be involved in osteoporosis and other musculoskeletal diseases.12,13 In the 1980s, it was found to reduce the risk of colon cancer,14 and to reduce blood pressure.15 In the 1990s it was found to reduce the risk of multiple sclerosis16 and the risk of being born with schizophrenia.17

Insufficient vitamin D is a significant health risk in the US and Northern Europe. This fact was underscored by the recent vitamin D conference held by the US National Institutes of Health.18 The impetus for the conference came from recent reports of rickets among breast-fed babies born to African–American mothers in the state of North Carolina.19 The goal of the conference was to help develop a research plan for improved guidelines for vitamin D. Some of the material presented here was developed for a manuscript relating to the topic of estimating the economic burden in the US due to insufficient vitamin D (Grant, submitted).

A list of diseases for which vitamin D is a risk-reduction factor and representative papers indicating some of the stronger evidence is presented in Table 1, whereas Table 2 indicates which types of evidence are satisfied for each disease (Grant and Holick, submitted). The list includes many diseases that are not ordinarily linked to vitamin D, such as diabetes mellitus, heart disease, hypertension, myopathy, psoriasis, and schizophrenia. There have been many good reviews published recently on the role of vitamin D in reducing the risk of disease.36–43

Table 1.  Summary of some of the stronger and/or most recent evidence indicating that UVB and/or vitamin D reduce the risk of various diseases.
DiseaseEvidenceReference
CancerGeographical variation with respect to solar UVB20
Serum 25(OH)D3 preceding colon cancer21,22
Diabetes mellitusHypovitaminosis D23
Correlation with vitamin D receptors23
Heart diseaseCorrelation with vitamin D receptors24
Inverse correlations of 25(OH)D3 with congestive heart failure25
HyperparathyroidismReduction in parathyroid hormone with UVB, vitamin D26
HypertensionGeographical and racial variations in blood pressure27
Infectious disease susceptibilityVitamin D and susceptibility to tuberculosis28
Multiple sclerosisGeographical variation29
Seasonal variation30
Risk from low childhood UVB31
MyopathyInverse correlations of 25(OH)D3 with body sway and muscle strength32
Association with hypovitaminosis D33
OsteoporosisUrban/rural difference in hip fracture rates34
Hip fracture prevention through calcium vitamin D supplements35
PsoriasisTreatment with UVB36
RicketsTreatment with vitamin D10
SchizophreniaVariation of risk with respect to sunshine during pregnancy17
Table 2.  Summary of evidence that vitamin D reduces the risk of specific diseases.
DiseaseLatitude or geographySolar exposureVitamin D receptorsMechanismsSerum 25(OH)DVitamin D analoguesClinical studiesAnimal or laboratory studiesVitamin D intake
Colon cancer+++++++++
Breast cancer++++++   
Ovarian cancer+++  +   
Prostate cancer++ +++ + 
Pancreatic cancer  +  + + 
Other cancers++ +   + 
Multiple sclerosis+++++ +++
Hypertension+++++++++
Psoriasis ++++++++
Diabetes mellitus Type 1++++   ++
Hyperparathyroid-secondary ++++++  
Myopathy, muscle weakness ++++ +  
Heart disease +++++++ 
Schizophrenia+ +++ ++ 
Renal disease end stage   ++ ++ 
Rheumatoid arthritis  + ++  +
Hyperparathyroid-primary++    +  
Tuberculosis  +++    
Graves′ disease  + +    
Diabetes mellitus Type 2  +     +
Periodontal disease +       

Cancer

As early as 1936 there were reports in the literature that solar radiation was inversely related to cancer mortality rates.44–47 However, it was not until a publication by the brothers Cedric and Frank Garland in 1980 that recent interest in the protective role of solar UVB radiation against cancer was initiated. Using the ecological approach, the Garlands established a link between colon cancer mortality rates in the US and solar UVB radiation and the production of vitamin D.14 (In ecological studies, populations are treated as entities within geographical confines; measures of disease outcome and possible influencing factors are found for the populations in the various geographical units, and statistical correlations are determined.) Additional ecological studies also found inverse correlations between solar UVB radiation and breast cancer,48 ovarian cancer,49 prostate cancer,50 and non-Hodgkin's lymphoma.51,52

These ecological studies provided the primary impetus for further studies on the role of solar UVB radiation and vitamin D in reducing the risk of cancer. A number of case-control and cohort studies were subsequently conducted on breast, colon, ovarian and prostate cancer. Sunlight associated with residence and/or occupation and serum vitamin D levels were found to be associated with 20–50% reductions in breast cancer incidence rates between the highest and lowest quartiles or quintiles.53,54 Similar results were obtained for studies on the risk of colon cancer, colon adenomas, and ovarian cancer.55–60

Colorectal cancer and vitamin D

A cursory review of the literature regarding the relation between colorectal cancer and vitamin D suggests that there is a general inconsistency in the findings: ecological studies always find that UVB and vitamin D are significant risk-reduction factors, whereas case-control and cohort studies generally find that dietary vitamin D is not a significant risk-reduction factor, pre-diagnostic 25(OH)D3 is sometimes a significant risk-reduction factor, and total ingested vitamin D is generally a significant risk-reduction factor. A critical review of these papers concluded that dietary sources of vitamin D are, by themselves, insufficient to provide sufficient protection against colorectal cancer; additional sources such as supplements or natural or artificial UVB are required.61

Geographical variation of cancer mortality rates in the US: UVB and other factors

In the first comprehensive ecological study of cancer mortality rates with respect to UVB radiation in the US,20 UVB radiation for July 1992 was obtained using the Total Ozone Mapping Spectrometer.62 These data were digitized to correspond to the approximately 500 state economic areas of the US that comprise the mid-level geographical division for cancer mortality data in the Atlas of Cancer Mortality in the United States.63 Cancer mortality rates for all states except six rapid-growth states were used in regression analyses with the UVB data. Solar UVB radiation was confirmed as a risk-reduction factor for 12 cancers, including bladder, endometrial, gastric, oesophageal, pancreatic, and renal cancer.20

Critics of that study pointed out that other factors that might also explain the geographical differences in cancer mortality rates in the US, and that all contiguous states should have been included. Accordingly, the ecological study was extended using additional covariates with the cancer mortality data averaged by state, for all contiguous states plus the District of Columbia (Grant, submitted). The fraction of the population living rurally64 was included as an additional index of solar UVB radiation, since rural life is associated with more time spent in the sun.65 Lung cancer mortality rates were used to account for the long-term adverse health effects of smoking, since smoking accounts for 87% of lung cancer mortality rates in the US66 Data on the proportion of the population who were of Hispanic heritage64 were used to help take into account the cancers with high mortality rates in states with large Mexican and Latin American populations.67 Alcohol consumption for 198068 was also included. Finally, a measure of socio-economic status, the fraction of people living below the poverty level,69 was included.

The new ecological study links UVB to a total of 16 types of cancer, primarily those of the digestive and reproductive systems (Grant, submitted). Six types of cancer (breast, colon, endometrial, oesophageal, ovarian, and non-Hodgkin's lymphoma) were inversely correlated to solar UVB radiation and rural residence in combination. Another 10 types of cancer (bladder, gallbladder, gastric, pancreatic, prostate, rectal, renal, testicular, vulvar, and Hodgkin's lymphoma) were inversely correlated with UVB but not with urban residence. Ten types of cancer were significantly correlated with smoking, six types with alcohol, and seven types with Hispanic heritage. Poverty status was inversely correlated with seven types of cancer. For African–Americans, UVB was inversely correlated with breast, colon, and rectal cancer, whereas smoking was correlated with bladder, breast, colorectal, oral, and pancreatic cancer. Since the results for alcohol, Hispanic heritage, and smoking for white Americans agree well with the literature, they provide a high level of confidence in the approach and its results for UVB radiation.

The number of premature cancer deaths prevented annually by vitamin D or ultraviolet exposure from 1970 to 1994, based on this multivariate analysis, was estimated to be 20 000–25 000, which agrees closely with the estimate of premature deaths due to insufficient solar UVB radiation, 16 000–23 000.20 However, the number of premature cancer deaths due to living in an urban residence, determined by plotting the mortality rate vs. the regression rate twice, once as calculated, and once with the fraction of urbanization set equal to zero, was about 25 000, bringing the total number of premature deaths to 45 000–50 000 per year. This number is about five times the number that die annually from melanoma and other skin cancers annually in the US, approximately 9800.69

Mechanisms of vitamin D for cancer prevention

Vitamin D may reduce the risk of cancer by mechanisms such as inducing cell differentiation, increasing cancer cell apoptosis, reducing metastasis and proliferation, and reducing angiogenesis.70–74 In addition, vitamin D down-regulates parathyroid hormone (PTH),75,76 which has been linked to cancer cell growth.75 The role of vitamin D in reducing the risk of cancer is so compelling that a considerable effort is being expended to find vitamin D analogues that have the effectiveness of vitamin D in fighting cancer without the problems of disregulating calcium metabolism.77 A recent MEDLINE search identified approximately 1000 papers reporting on vitamin D or its metabolites and cancer as major subjects of the reports.

Many organs have been shown to convert the inactive form of vitamin D, 25(OH)D3, to the active, cancer-reducing form, 1,25(OH)2D3. This ability has been shown for the prostate78 and for the brain, colon, lymph nodes, pancreas, placenta, and skin.79

Prostate cancer

Luscombe et al.80 recently examined the association between UV exposure and prostate cancer risk using a case-control approach in Northern European Caucasians (210 prostate cancer cases and 155 patients with benign prostatic hypertrophy BPH). Exposure was assessed using a validated questionnaire. Chronic exposure was assessed by: (i) daily sun exposure (weekdays and weekends, considered separately and combined) in three age categories (20–39, 40–59 and over 60 years old) (ii) proportion of working life spent outdoors and (iii) history of residence abroad in a hot country for over 6 months. Acute exposure was assessed by: (i) childhood sunburn (erythema for more than 48 h or blistering) recorded as yes/no and number of recalled sunburn events (ii) history of foreign holidays with average weeks abroad/year (iii) sunbathing calculated as never, rare, occasional, or frequent (scored as 1, 2, 3 and 4, respectively) in the three age categories above. Factors related to response to UV including skin type, hair, and eye colour were also recorded.

The cancer cases had less cumulative exposure than the BPH patients (P = 0.006). In particular, subjects with the lowest 25% of exposure (below 1639 days or 1.9 h/day) were at greatest risk of the cancer. Thus, compared with the upper three quartiles, patients with the lowest 25% of exposure had a 2.5-fold increased risk of prostate cancer (P = 0.001). There were no significant associations with outdoor work or history of living abroad. For acute exposure, a positive history of childhood sunburn was protective (P < 0.0001) and increasing numbers of childhood sunburn events increased this effect (OR = 0.64 per event, P < 0.001). Other factors associated with acute UV exposure (cumulative sunbathing score, history of regular holidays), were also significantly associated with cancer risk. There was no demonstrable effect from the use of sunscreens. Susceptibility was not associated with hair colour, eye colour or skin type. There was a trend for individuals with skin type 4 (tans but never burns) to have an increased risk relative to other skin types, although this was not significant (OR = 1.49, P = 0.143). Indeed, further analysis of the data showed that among men with low levels of exposure, skin type 1 conferred protection compared with skin types 2–4 (OR = 4.78, 95% CI 3.01–8.25, P < 0.0009).81 These findings indicate that susceptibility to prostate cancer is in part determined by extent of exposure to UVR and that the ability to pigment mediates this effect. Importantly, these data were confirmed in a new group comprising 242 prostate cancer cases and 157 BPH patients in the UK.82

More recent results from Scandinavia indicate that a moderate concentration of 25(OH)D3 (40–60 nmol/L) is correlated with the lowest risk of prostate cancer.83 The authors suggested that low serum 25(OH)D3 concentration leads to a low tissue concentration and to weakened mitotic control of target cells, whereas a high vitamin D level might lead to vitamin D resistance through increased inactivation by enhanced expression of 24-hydroxylase. This result is not peculiar to Scandinavia; a similar finding was made in an ecological analysis of the geographical variation of prostate cancer mortality rates in the US. Unlike many cancers such as breast, colon, and ovarian cancer, which have their highest mortality rates in the north-east and lowest in the south-west,63 prostate cancer has a fairly pronounced latitudinal gradient in mortality rates with the highest values at the highest latitudes. In the ecological analysis, it was determined that latitude had the highest correlation with prostate cancer mortality rates, with the square of UVB being more weakly correlated, and urban residence being weakly inversely correlated.84 This result suggests that wintertime UVB levels (minimum values of 25(OH)D3) are most important in reducing the risk of prostate cancer, whereas summertime UVB levels (highest 25(OH)D3 levels) are a risk factor. Thus, moderate levels may be associated with the lowest risk.

A role for genetic polymorphisms

The link between prostate cancer risk, UV exposure and vitamin D synthesis suggests that an individual's ability to initiate pigment synthesis may mediate the harmful and beneficial effects of UV.85 Allelism in genes associated with ability to pigment following exposure may influence prostate cancer risk.85 Thus, under conditions of moderate exposure common in Northern Europe, individuals with lighter skin and little ability to pigment (skin type 1) will synthesize more vitamin D than subjects with darker skin.86 Accordingly, risk of prostate cancer will be lowest in men with light skin who fail to pigment. This risk will be moderated by extent of exposure. In particular, individuals with skin type 1 often develop sun avoidance strategies to avoid burning. Genetic factors in the synthesis of melanin need to be considered, because melanin largely determines skin colour. The rate-limiting steps in melanin synthesis are catalysed by tyrosinase (TYR) under the influence of melanocyte-stimulating hormone. This hormone acts via the melanocortin-1 receptor (MC1R). Both TYR and MC1R have polymorphisms with functional consequences.85 Vitamin D itself is also clearly important and some but not other studies have shown links between vitamin D receptor (VDR) genotypes and prostate cancer risk. Luscombe et al.85,87 found that polymorphisms in TYR (codon 192 variants) and MC1R were associated with prostate cancer risk. Homozygosity for MC1R Arg160 was associated with increased risk (OR = 2.18), whereas homozygosity for the TYR A2 allele was linked with reduced risk of cancer (OR = 0.42). Importantly, the protective effect of TYR genotypes found in the total group reflects an association with risk in subjects with the highest quartile of exposure. Similar associations between VDR polymorphisms and prostate cancer risk and the level of exposure to UVR have also been recently reported; in men with UVR exposure above the median (11.00 h/year), the CDX-2 GA (odds ratio = 2.11), CDX-2 AA (odds ratio = 2.02), and Fok1 ff (odds ratio = 2.91) genotypes were associated with increased prostate cancer risk.88 These data show for the first time, that allelism in genes linked with skin pigment synthesis is associated with prostate cancer risk.

Multiple sclerosis

The story of how it was realized that vitamin D is an important risk-reduction factor for multiple sclerosis (MS) is interesting, especially since the data required to make this connection have been available since the 1920s, but the interpretation did not come until 1997. Data on prevalence of MS in the various US states were developed for veterans of World Wars I and II and of the Korean War. In both data sets, there were very strong latitudinal gradients, with MS prevalence increasing rapidly with latitude.29 In 1997, the first paper appeared suggesting that vitamin D explains this gradient.16 A strong case for UV radiation in reducing the risk of MS was made on the basis of a case-control study in Australia in which it was determined that childhood sun exposure, especially in winter, was associated with a significant reduction in risk.31 More recently, a study based on the Nurses’ Health Study found that total ingested vitamin D was a significant risk-reduction factor,89 and a study in the UK found that MS among those with non-melanoma skin cancer, an indication of time spent in the sun, was at half the value for the general population, unlike the association for other diseases among this group.90 Furthermore, vitamin D will also reduce the symptoms of MS. The mechanisms for the effect of vitamin D on MS are known.91 Interestingly, recent studies have reported associations between polymorphisms in genes associated with skin pigmentation and MS risk.92 Thus, the evidence available indicates that that MS rates in the US and the UK could be reduced significantly through adequate vitamin D. In addition, there is evidence from the seasonal cycle of lesions associated with MS that UVB and vitamin D can reduce by about half the number of lesions that occur for low serum levels of 25(OH)D3.30

Psoriasis

Psoriasis and other skin diseases benefit from UVB. An uncontrolled study of the use of commercial indoor tanning facilities to treat those with psoriasis found 30–50% improvements in symptoms.93 Analogues of vitamin D3 have been used as a topical therapy for psoriasis.94

Serum vitamin D levels and sources of vitamin D

One problem with the current guidelines regarding solar UVB exposure and vitamin D supplementation in many countries is that many people are not getting adequate amounts of vitamin D. Vitamin D insufficiency is a serious problem in the US due to a variety of factors. Winter doses of UVB radiation are insufficient to produce vitamin D in all but the most southern parts of the country. In addition, the modern lifestyle includes little time spent outdoors, and when people are in the summer sun, they often use sunscreens, which block the UVB radiation and reduce serum 25(OH)D3 production.95 Examples of vitamin D insufficiency can be found readily in the health literature for dark-skinned people in the US,19 Australians,96 and Canadians.97 Hypovitaminosis D is common in the UK and USA and is associated with various abnormalities in bone chemistry among elderly residents in these countries. This reason alone is a sufficient rationale for these countries to adopt a vitamin D supplementation programme, with 10 micrograms of vitamin D recommended.98

There is considerable evidence that levels of serum 25(OH)D3, the intermediate compound between cholecalciferol (vitamin D3) and 1,25(OH)2D, are often inadequate in residents of European countries. For example, the prevalence of subclinical vitamin D deficiency decreases with latitude in winter in Europe, falling from 50 to 80% in Greece to 20–30% in Norway.99,100 This finding is counterintuitive, but is probably related to a higher intake of vitamin D from diet and supplements in northern Europe to compensate for lower annual levels of UVB radiation, and the fact that there is insufficient UVB to produce vitamin D in winter even in southern Europe.101 There have been reports that vitamin D consumption and serum 25(OH)D3 are inadequate in Austria.102 Although serum 25(OH)D3 levels were similar for both genders in an adult population in Finland, serum parathyroid hormone (PTH) levels for women started to increase at half the serum vitamin D levels for men.103 Based on measurements of PTH, 86% of the women and 56% of the men were determined to have insufficient vitamin D status. Similar results were found for male adolescents in France, where serum 25(OH)D3 fell from approximately 59 nmol/L (24 ng/mL) in summer to approximately 21 nmol/L in winter.104 Pre-school children in the UK were found to be prone to low 25(OH)D3 levels in winter unless they were taking vitamin D supplements.105 Half of the pregnant women from the non-European ethnic minority population in South Wales had serum vitamin D levels below 8 ng/mL.106

Vitamin D supplementation at moderate dosages of 400–600 IU per day appear to be without any significant risk.107 It has been argued108,109 that daily intakes of 100 µg (4000 IU) of vitamin D3 per day is safe. However, serum vitamin D3 levels vary widely by individual for the same intake. Dosages in children should correspond to body mass and should be determined with greater caution. Oral doses from supplements in excess of 2000 IU/day may be associated with adverse effects such as increased calcium loss from bones in some individuals and should be avoided until further data are available. Vitamin D status would be best assured by periodic measurement of serum 25(OH)D3 levels, a simple test that is widely available.

Another way to obtain vitamin D in winter or when confined indoors is through use of UV lamps. A study in the UK found that the use of low-intensity UV lamps turned on 15 min per day and yielding a summertime dose of UVB for ambulatory people raised serum 25(OH)D3 levels from a mean near 12 nmol/L to about 32 nmol/L after about a year.110 The end values are still not optimal, but do represent a substantial improvement. It should be noted that the efficiency of vitamin D production in skin decreases with age.

Mean values of serum 25(OH)D3 in Boston are 35 ± 10 ng/mL at the end of summer and 30 ± 10 ng/mL at the end of winter.111 Taking multivitamins reduced vitamin D insufficiency significantly at the end of winter. These values are for a region of the USA where mortality rates for eight types of cancer are about twice those in the south-western states.63 Thus, values for 25(OH)D3 in the range 60–70 ng/mL might be required for optimal protection against cancer and several other chronic diseases. What is not well understood is the amount of casual or intentional UVB dose required to generate adequate levels of serum 25(OH)D3. The amount varies considerably depending on a number of factors, and there has been little systematic study for any of the various conditions linked to vitamin D.

Some changes in public health policy regarding vitamin D intake are being considered. There have been suggestions that vitamin D supplementation be increased in Denmark112 and Boston.113 In Europe, there is a programme named OPTIFORD underway to investigate if fortification of food with vitamin D is a feasible strategy to remedy the insufficient vitamin D status of large population groups.114

Adverse health effects of UVR

Melanoma and other skin cancers

It is worthwhile to examine whether the risk of melanoma and other skin cancers can be minimized while at the same time increasing the production of vitamin D from solar UVB radiation. The risks that have been identified for melanoma include light hair, skin, and eye colour, a history of heavy freckling in adolescence, and a tendency to burn readily and tan poorly.115 Intermittent sunburns, such as on weekends or vacations, are more commonly associated with melanoma than is daily sun exposure.116

The UVA spectral region appears to be more strongly associated with melanoma than is UVB radiation.117–121 UVA radiation penetrates the skin deeper than does UVB radiation, where UV generates free radicals that subsequently damage DNA.120,121 UVB seems therefore to be more involved in melanoma indirectly through temporarily reducing the protective layer of skin through sunburn rather than directly through DNA damage or free radicals. Although UVB does generate free radicals, their concentration at the basal epithelium is only 1/70th that of the more common and deeper-penetrating UVA photons.121 Vitamin D present in the epidermis may actually reduce the risk of melanoma.117,122 The ratio of UVA to UVB increases with latitude, which seems to be linked to the increase in melanoma mortality rates with latitude in Europe.122

Further evidence for UVA comes from the recent meta-analysis of studies that investigated whether use of sunscreen reduced the risk of melanoma – the finding was that it did not.123 This finding is probably due to the fact that sunscreen is much more effective at blocking UVB than UVA.

The dietary links to melanoma and other skin cancers are also important. High-fat diets are thought to be risk factors for melanoma and other skin cancers.124,125 Increased height, weight, and body surface area are associated with increased risk of melanoma among males in Washington State.126 A low-fat diet was found to increase the survival rates of patients with advanced melanoma.127 Vitamin E is inversely correlated with BCC.128 Vitamin A is a risk-reduction factor for melanoma.129 Smoking is a risk factor for BCC and SCC.130 Thus, UVR is not the only risk factor associated with skin cancers, and the risk factors may act synergistically.

Thus, an overall recommendation to minimize, or even avoid, time in the sun may not be the best way to reduce the risk of melanoma and other skin cancer.131 A better recommendation may be to seek limited but regular solar UVB exposure for vitamin D production and normal seasonal skin accommodation in summer, but to avoid sunburns and excessive tanning. When solar UVB radiation is not sufficient for vitamin D production, which could be for 5–6 months of the year in the UK, based on results in Boston,101 then the possible use of artificial UVB lamps or vitamin D supplements or fortification of food needs to be considered.

Other adverse health effects from UVR

There are several other adverse health effects from UVR, especially from high doses. One is cataract formation. In the US, the prevalence of cataracts increases by 3% per degree of latitude to the south.132 One way to reduce the risk of cataract formation is to wear UV blocking glasses when exposed to UVR. Another way is to include lutein-rich fruit and vegetables or supplements.133,134

Premature skin ageing is another major concern with respect to UVR exposure.6 Excess UVR exposure should be avoided. However, a good way to reduce these effects of UVR exposure is through consuming plenty of antioxidants.135

There are some conditions for which the best policy is near total avoidance of UVB radiation. One of these is systemic lupus erythematosus (SLE). In the US, a high correlation was found between SLE mortality rates and solar UVB radiation for July.136

Sunbeds

That the careful use of sunbeds may be an appropriate way to obtain vitamin D can be supported from several directions. First, the lamps used in sunbeds today have nearly the same ratio of UVB to UVA as sunlight incident at mid-latitudes – about 0.04. Second, in Europe, although use of sunbeds has been associated with a 50% increase in risk of melanoma,137 this is not the case in the US and in a recent UK study.138 The two studies that investigated this link in the US found no significant risk.139,140 The UK study found that the only significant associations in this study were with 10 or more sunburns and the use of a sunbed in young subjects with fair skin.138 This study also found a risk reduction for melanoma for the greatest total hours of sunbed usage, and pointed out that many studies of melanoma and sunbed use had failed to demonstrate the dose–response relationship that is required to show causality. It is suggested that the difference may be that the use of sunbeds is more carefully regulated in the US than in Europe, especially in regard to initial dose, maximum dose, and frequency of use. However, other confounding factors such as smoking and types of lamps used may also play a role. These important issues need to be addressed. Third, even if there were a 50% increased risk of melanoma, the health benefits from indoor tanning would mirror those from solar UVB exposure. A preliminary study of the economic burden in the US in 2003 associated due to impaired health or mortality due to insufficient UVB, the primary source of vitamin D in the US141 found that it was approximately $50 billion (range $25–$75 billion), which was much larger than the $3 billion attributed to the health risks of BCC, SCC, melanoma, cataracts, and premature skin ageing (Grant, in preparation). Fourth, it is noted that melanoma is much more related to recreational UVR exposure than to occupational UVR exposure.116 One of the advantages of sunbeds is that if properly used, they could provide a tan in a controlled manner more in accordance with occupational exposure, so that when one does take that vacation trip to the beach, one is much less likely to sunburn.

Discussion

It appears that the concern with the adverse effects of solar UV radiation exposure, namely increased risk for melanoma, basal cell and squamous cell carcinoma, premature ageing of the skin, and cataracts, may have led to public health recommendations that also have unintentionally reduced serum 25(OH)D3 levels. The health benefits of UVB seem to outweigh the adverse effects by a ratio of 15 : 1 in the US (Grant, in preparation), with a higher ratio likely in the UK, since solar UVB levels are lower there. We recognize the need for public health recommendations that protect the public from undue harm, but current guidelines regarding solar and artificial UVB radiation exposure and vitamin D fortification and supplementation appear to be inconsistent with new data on UVB and vitamin D. All findings should be reviewed and new guidelines developed that would provide a better balance between the health benefits and risks of sun exposure.

Conclusions

There are many health benefits from UVB radiation, which is an important source of vitamin D for most people on Earth. The health benefits include reductions in risk of 16 types of internal cancers, of diabetes mellitus, heart disease, hypertension, multiple sclerosis, myopathy, osteoporosis, psoriasis, rickets, schizophrenia, and tuberculosis.

Acknowledgements

William Grant is forming an organization called Sunlight, Nutrition and Health Research Center (SUNARC), which will have as its goals the continued research into the health benefits of vitamin D and UVB radiation and the health effects of diet and nutrition, the collection of information on these topics, and advocacy of revised health guidelines based on the findings. It is anticipated that the indoor tanning industry will be providing some of the funding for SUNARC. The North Staffordshire Medical Institute provided Richard Strange with some of the funding for work described in this paper.

Ancillary