INTRODUCTION
The World Health Organization's International Agency for Research on Cancer recommends
avoiding outdoor activities at midday, wearing clothing to cover the whole body, and
daily use of sunscreen on usually exposed skin [1]. The American Cancer Society advocates
Slip! Slop! Slap! and Wrap! to make sure skin is covered in clothing or sunscreen
and to avoid exposure to the sun between 10 am and 4 pm
[2]. The U.S. Surgeon General has issued a Call to Action focused on reducing ultraviolet
(UV) exposure, whether from indoor UV or from the sun [3]. Though these recommendations,
all focused on reduction of skin cancer, are accompanied by brief acknowledgement
of the importance of vitamin D for health, they persist in urging avoidance of the
sun at the precise times when vitamin D can be synthesized in the skin—the hours between
10 am and 3 pm—and suggest that all necessary vitamin D can be obtained through food
and dietary supplements.
These recommendations are understandable from the viewpoint of preventing the 3.5
million new cases of and 2000 deaths from nonmelanoma skin cancer in the United States
each year [4], but they neglect the fact that we have a long cultural history of appreciation
of the sun and use of UV radiation for healing purposes. Moreover, they neglect that
we have evolved with physiological adaptations to help protect the skin from the sun
[5] when we are mindful of our exposure and do not burn. They neglect the fact that
increased sun exposure, based on latitude, has been associated with protection from
several different types of cancer [6–15], type 1 diabetes [16], multiple sclerosis
[17,18], and other diseases [19–23]. They also neglect the fact that exposure to the
sun induces beneficial physiological changes beyond the production of vitamin D. Though
adherence to the current sun-protective recommendations would likely result in the
reduction of nonmelanoma skin cancer, that reduction would likely be overshadowed
by the potential reduction in deaths from other cancers and from cardiovascular disease,
which could be achieved by doubling average blood concentrations of 25-hydroxyvitamin
D (25(OH)D) to 40 ng/mL through a combination of sun exposure and supplements [24].
The potential harm of sun avoidance and the neglect of its positive effects on human
health led to a seminar, Vitamin D for Public Health: Integrating Sunshine, Supplements
and Measurement for Optimal Health, presented by GrassrootsHealth at the University
of California San Diego to inform and to help initiate an action plan to restore a
more balanced approach to solar radiation based on input by the conference speakers.
HELIOTHERAPY
The healing power of the sun and its use in medical treatment (heliotherapy) have
roots extending back into antiquity. In the modern era, particularly the first half
of the 20th century, heliotherapy was widely used in both Europe and North America,
particularly for the treatment of cutaneous tuberculosis, for which Niels Finsen garnered
the Nobel Prize for Medicine in 1903. Much of this work was done prior to the discovery
of vitamin D and of its synthesis in the skin by UV radiation, which would have been
a principal factor in the recovery from disease reported a century ago. However, with
the discovery of antibiotics, the era of drug treatment of tuberculosis began in the
1950s, and heliotherapy fell into disuse and is today virtually forgotten. A major
advantage of antibiotics was the ability to avoid prolonged hospitalization with its
associated expense and disruption of individual lives. But that was a matter of efficiency,
not efficacy.
Tuberculosis currently afflicts 30% of the world's population. The effects—and perhaps
the benefits—of heliotherapy in this disorder, though much less well studied today,
extend beyond the synthesis of vitamin D [25]. We do not know (1) the relative efficacy
of antibiotic treatment and heliotherapy for various manifestations of tuberculosis
and (2) whether vitamin D, by itself, is sufficient to explain the therapeutic efficacy
of heliotherapy in this disease.
Physiological Responses to Sun Exposure
The best recognized response to sun exposure is elevation of vitamin D status. Two
African tribes, the pastoral Masai and the hunter–gatherer Hadza, have been shown
to have serum 25(OH)D concentrations averaging 46 ng/mL [26]. Both tribes live in
equatorial East Africa, where humans are thought to have originated, and have daily
sun exposure approximating that of ancestral humans.
Physiological responses go beyond production of vitamin D. When the skin is stimulated
with UVA radiation, nitric oxide is released, stimulating vasodilation and lowering
of blood pressure. During active exposure to UVA, diastolic blood pressure in one
study fell by roughly 5 mmHg and remained lower for 30 minutes after exposure [27].
A reduction of diastolic blood pressure by 5 mmHg decreases risk for stroke by 34%
and coronary heart disease by 21% [28].
Another physiological response of skin exposure to sunlight is the thickening of the
stratum corneum (the outermost layer of the epidermis) and increased skin pigmentation
through production of melanin. This paired response actually protects the skin and
deeper tissues from the deeper penetrating and damaging UVA rays while retaining benefits
from UVB exposure [29]. Though both UVA and UVB exposure result in increased skin
pigmentation, the mechanisms are different, with UVB being responsible for the up-regulation
of melanin synthesis and thus the protective effects against UV damage to DNA [30].
The best time for creating this response coincides with the time of maximal UVB availability
(10 am–3 pm).
Additionally, human skin produces beta-endorphin in response to UVB exposure [31];
these opioid peptides have the result of increasing a feeling of well-being, boosting
the immune system, relieving pain, promoting relaxation, wound healing, and cellular
differentiation [31–33]. Light signals received through the eye regulate production
of melatonin and serotonin for circadian rhythm control and also play a role in seasonal
affective disorder [34].
Impact of Sunlight and/or Vitamin D on Specific Health Conditions
Cancer
Studies of the relationship between cancer, sun exposure, and vitamin D began decades
ago with geographic associations with cancer mortality. In 1941, Apperly reported
an association between latitude and cancer mortality based on sun exposure although
vitamin D was not yet explicitly implicated [6]. In 1980, Garland and Garland reported
the association between latitude and colon cancer using sun exposure as a proxy for
vitamin D status [7]. By 1990 it had been hypothesized that deficiency of vitamin
D was the main cause of breast cancer [35–37].
In the last decade, analyses of UVB irradiance and cancer incidence in countries worldwide
have shown a regular pattern of higher rates for countries that are further away from
the equator compared to those near the equator. This pattern has held true for cancers
of the colon [8], breast [9], pancreas [10], ovary [11], brain [12], bladder [13],
kidney [14], and multiple myeloma [15].
In the United States, mortality rates for 15 types of cancer for white Americans are
highest in the northeast and lowest in the southwest; these rates are inversely correlated
with solar UVB irradiance [38]. Similar findings have been reported for Australia,
China, Japan, and Spain [39].
In addition to these epidemiological studies, other studies using serum concentrations
of 25(OH)D have found strong inverse associations for cancer risk and vitamin D status.
A 2011 study in 10 European countries reported that individuals with the lowest concentrations
of 25(OH)D (averaging 8 ng/mL) had almost 3 times the risk of colon cancer as those
with the highest (averaging 50 ng/mL) [40]. For breast cancer, Lowe et al. found a
50% lower incidence for women who had 25(OH)D concentrations at 48 ng/mL compared
to those at 10 ng/mL [41]. Similar findings from Mohr et al. in 2011 showed a 50%
reduction in short-term incidence at 45 ng/mL versus 12 ng/mL [42]. A more recent
meta-analysis of 11 case–control studies of breast cancer incidence rate versus 25(OH)D
concentration near time of diagnosis found a 70% lower incidence rate for 45 ng/mL
compared to 5 ng/mL [43].
Though higher latitudes and greater cloud cover predictably decrease vitamin D status,
the inverse association is also true: lower vitamin D status is a marker for reduced
sun exposure. This distinction is important because it is not possible in the epidemiological
studies cited to distinguish the effects, if any, due to reduced vitamin D status
and those due to other actions possibly produced by sun exposure. The mechanism of
heliotherapy action must be recognized as increased vitamin D synthesis plus other,
inadequately characterized spectral effects.
Two intervention studies have reported vitamin D effects on cancer. Lappe et al.,
using a randomized controlled trial (RCT) design, showed an approximate 70% reduction
in all-cancer risk in postmenopausal women given calcium and vitamin D in a dose sufficient
to raise serum 25(OH)D from 29 to 38 ng/mL, whereas those given only calcium showed
an approximate 40% reduction in all-cancer risk [44]. In another study, men with low-risk,
biopsy-proven early-stage prostate cancer were given 4000 IU/day of vitamin D3
[45]. Over the year of treatment, mean circulating concentrations of 25(OH)D rose
from 33 to 66 ng/mL. A second biopsy showed that the number of cores positive for
malignancy was reduced for more than half of those enrolled in the study. Patients
from the same practice who did not receive supplementation showed an increase in positive
cores over the same period of time. There were no adverse events as a result of vitamin
D supplementation in either study.
A proposed model for how vitamin D deficiency is related to a wide array of cancers
is the DINOMIT model of cancer progression, which outlines a mechanism whereby vitamin
D might restrain cancer development and spread [46].
Type 1 Diabetes
Improved vitamin D status has been associated with a lower risk of type 1 diabetes.
Just as vitamin D status varies by season and latitude because of availability of
sun exposure [47], type 1 diabetes incidence rates peak annually in the winter/spring
[48] and risk varies directly with distance from the equator (just as with many cancers)
[16]. In Finland, with one of the highest rates of type 1 diabetes, the frequency
and dosage of vitamin D supplementation during the first year of life have been associated
with type 1 diabetes rates. Specifically, there is an almost 90% lower risk of type
1 diabetes by age 31 for individuals regularly given vitamin D supplements in infancy,
versus those who were not given supplements. Among those who received supplements,
those who received at least 2000 IU/day had an 80% lower risk than those who received
less than 2000 IU/day [49]. From 1965 to 2005, Finland had a dramatic increase in
the incidence of type 1 diabetes; over that same period of time, the recommended vitamin
D intake decreased from 4500 IU/day to 400 IU/day [50]. Though association does not
equal causation, this phenomenon is difficult to explain in any other way. A qualitatively
similar difference in diabetes risk is reported from the EURODIAB study [51] in which
countries recommending vitamin D supplementation in infancy had lower type I diabetes
incidence rates by age 15 than those countries not recommending vitamin D.
In a nested case–control study of U.S. service members, those with 25(OH)D concentrations
<14 ng/mL had 3.5 times the risk of type 1 diabetes compared to those at 40 ng/mL
or higher [52]. In a large cohort study, both insulin resistance and fasting insulin
levels were inversely associated with serum 25(OH)D concentration, providing biological
plausibility for a contributory role of vitamin D in diabetes [53]. Evidence suggests
that improving the vitamin D status of the population could lead to a marked decrease
in type 1 diabetes incidence.
Pregnancy
Only recently have we started to understand the developmental origins of disease and
how the perinatal environment affects lifelong health. Worldwide there is profound
vitamin D deficiency among pregnant women, yet the role of vitamin D in pregnancy
has largely been ignored [54]. Epidemiological data have shown that deficiency during
pregnancy causes higher risk of maternal preeclampsia [55–58], gingivitis, and periodontal
disease in the mother [59,60] and impaired fetal growth [61,62], impaired dentition
[63,64], and increased risk of respiratory syncytial virus infection [65] in the infant.
A recent RCT in India showed that women who were given vitamin D supplementation during
pregnancy had a 61% lower risk of preterm labor and a 47% lower risk of hypertensive
complications compared to participants who were not given supplementation [66]. In
another RCT, vitamin D supplementation of 4000 IU/day was shown to be safe and effective
in achieving sufficiency (≥32 ng/mL) for pregnant women and their infants, whereas
400 IU/day was ineffective [67–69]. In fact, women who achieved at least 32 ng/mL
had a lower risk of gestational diabetes, preterm birth, preterm labor, preeclampsia,
hypertensive disorders of pregnancy, and infection [69]. Overall, there is approximately
a 50% reduction in preterm birth when 25(OH)D serum concentrations of 40 ng/mL are
attained [70]. With a preterm birth rate in the United States of 11.4% and an associated
cost of $26 billion per year [71], achieving an optimal 25(OH)D concentration of 40 ng/mL
in pregnant women would greatly reduce this human and financial burden.
Vitamin D for Optimal Health
In common with many other micronutrients, vitamin D is a necessary but not sufficient
factor for key cell-biologic processes. That is, it is an enabler; it must be present
for those processes to occur, but it does not, itself, stimulate or cause them. In
brief, low vitamin D status does not so much cause disease or dysfunction as it impairs
cellular response to both internal and external signals. It is now recognized that
essentially every tissue and cell in the body has vitamin D receptors. Furthermore,
most cells also have the capability of converting 25(OH)D to its active form, 1,25-dihydroxyvitamin
D [1,25(OH)2D], and most of our daily vitamin D consumption occurs in this way [72].
This conversion in the cell allows each tissue to use vitamin D as it is needed. It
also follows that, in the absence of vitamin D, none of our body systems can work
at their optimal potential. Thus, it becomes clear that low vitamin D status would
inevitably be involved in a wide range of dysfunctions and diseases.
To address the issue of how much vitamin D is needed for optimal health, it is necessary
first to consider the desired endpoint. If the goal is to avoid a known disease, such
as rickets, the amount required will be less than if the goal is to optimize bone
health. Moreover, shifting focus to the physiological need for vitamin D offers several
criteria for determining need [73], including (1) the intake that minimizes the need
for functional adaptation or compensation; (2) the status found in our hominid ancestors;
or (3) the amount necessary to support a critical life function. These criteria are
elaborated below:
One manifestation of adaptive compensation is the elevated parathyroid production
that occurs when vitamin D status is low. Vitamin D–mediated intestinal calcium absorption
reduces parathyroid activity, and the point at which vitamin D ceases to be a limiting
factor in calcium absorption is the point at which parathyroid hormone concentrations
are lowest. That occurs when serum concentrations of 25(OH)D are in the range of 48–52 ng/mL
[74].
Approximating the vitamin D status of ancestral humans, as noted above, has been done
for 2 African tribes. Their average 25(OH)D concentrations (around 46 ng/mL) [26]
provide the best available estimate of the level to which human physiology has been
fine-tuned by natural selection over the millennia of human evolution.
A critical function of vitamin D is passage of the nutrient from mother to child in
breast milk. In order to fully support the infant's need for vitamin D, the mother
must have a blood concentration of cholecalciferol (vitamin D3) above 10 ng/mL, which
happens only when serum 25(OH)D is above 48 ng/mL [75]. At today's prevailing vitamin
D status values, no D3 is transferred into breast milk. Hence, currently recommended
25(OH)D concentrations are not adequate to support this critical physiological function.
In brief, all 3 physiological criteria converge on blood concentrations around 48 ng/mL.
Though a 25(OH)D concentration of 20 ng/mL may be sufficient to avoid clinically evident
rickets, it is not sufficient to sustain physiological functions and promote optimal
health.
Vitamin D may come from UV exposure, dietary intake, or supplements. The input from
all sources required each day to meet physiological needs and to support optimal health
is estimated to be roughly 6000 IU/day [76,77]. However, because of variations in
individual ability to produce vitamin D from UV exposure or to absorb it from dietary
sources, as well as variations in individual requirements, testing serum concentrations
of 25(OH)D remains important.
Cost–Benefit Analysis
Of the 30 leading causes of death in the United States in 2010, 19 have been linked
to low vitamin D status, including various forms of cardiovascular disease, various
cancers, diabetes mellitus, Alzheimer's disease, and falls and fractures in the elderly
[78]. If the population of the United States were to increase their vitamin D status
to 40 ng/mL, we could expect to see a potential reduction of as much as 336,000 deaths
each year (out of 2.1 million deaths attributed to the diseases concerned) [24]. This
includes estimated reductions of 180,000 deaths from cardiovascular disease, 20,000
from colorectal cancer, 12,000 from breast cancer, 70,000 from other cancers, and
15,000 from Alzheimer's disease. In addition to this annual reduction in deaths, the
direct costs of care for the associated diseases would be reduced by roughly $130
billion each year. Raising 25(OH)D concentrations appears to be the most efficient
and cost-effective way to reduce the burden of disease and increase life expectancy
in the United States [24].
Among the reasons vitamin D deficiency is so widespread are the public health messages
from the U.S. Surgeon General, the Institute of Medicine, and the World Health Organization,
all of whom promote avoidance of sun exposure and covering the skin with clothing
or sunscreen when out in the sun. It should be noted that these messages focus mainly
on reducing nonmelanoma skin cancer. With a total of 5 million cases of skin cancer
treated each year at an annual cost of $8.1 billion, skin cancers result in 13,000
deaths annually. Melanoma, by far the most deadly form of skin cancer, accounts for
70%–75% of those deaths [2,3] and 40% of the costs [3]. Despite public health messages
to the contrary, not all skin cancers, particularly melanomas, are directly attributable
to moderate sun exposure. Though painful sunburns before the age of 20 seem to be
a strong predictor of all types of skin cancer, chronic or lifetime sun exposure is
associated with an increased risk of nonmelanoma skin cancers but a decreased risk
of malignant melanoma [79].
CONCLUSIONS AND RECOMMENDATIONS
The full solar spectrum is essential to optimal health and well-being. Humans are
physiologically adapted to produce vitamin D in response to sun exposure, specifically
UVB radiation; other regions of the spectrum seem to confer benefit as well. Though
some vitamin D comes from our diet (and more recently from supplements), we should
not ignore the natural capacity that we possess to produce our own. We are of the
opinion that moderate sun exposure (less than the time required to burn) to the arms,
shoulders, trunk, and legs should be sought rather than avoided. Once that limited
time has been achieved, we agree that covering the skin or seeking shade may be appropriate.
The benefits of such exposure go beyond production of vitamin D and include other
physiological responses to sunlight, still inadequately explored, including release
of nitric oxide, production of beta-endorphin, and regulation of circadian rhythms—all
important components of lifelong health and well-being.
The current policy of sun avoidance is creating probable harm for the general population.
Ignorance of the effects of portions of the solar spectrum at wavelengths longer than
the ultraviolet is due mainly to lack of suitable measurement tools for cutaneous
and systemic responses to those regions. We propose therefore that the U.S. Surgeon
General's office, the World Health Organization, the Institute of Medicine, and other
health entities, together or separately, engage in an immediate effort both to define
and quantify comprehensively the benefits and harms of sun exposure and to develop
the measurement methods needed for their detection and quantification. Following this
effort, concrete recommendations for exposure at an individual level that are both
safe and beneficial should be created. We also recommend, as an interim strategy,
that both sun exposure and vitamin D supplementation be concomitants of drug therapy
for tuberculosis so as to garner both whatever benefits may be due to vitamin D and
those of heliotherapy that extend beyond its effect on vitamin D status.
AUTHORS' NOTE
Videos of all presentations from the seminar upon which this article are based can
be accessed at http://ucsd.tv/vitamin-d-public-health.