Positive Health Online

Noel Coward sang that only "mad dogs and Englishmen go out in the midday sun", "because the sun is much too sultry, and one must avoid its ultraviolet ray!" When the Europeans who colonized Africa and Asia went out in the sun they wore "solar topees" because, even a century ago, medical people were warning of the dangers of intense sunlight.

There was a debate about whether places with high levels of UV exposure, like the highlands of Kenya, could ever be "white man's country". Some doctors took the view that the "actinic rays" of the sun would sap the vitality of the European stock and lead to degeneration.

Today, when white people have lived in Africa for generations, the fears of degeneration seem to be groundless –and nobody wears a solar topee any more. All that has happened is that after a lifetime in the tropics some people have developed skin cancer, which is the least serious kind of cancer.

The history of medical attitudes to ultraviolet is interesting. In 1877 scientists discovered that sunlight had the effect of killing bacteria; research over the next thirty years showed that the UV component was responsible, and could kill off the bacteria of tuberculosis, cholera, anthrax and other diseases. In 1903 the Nobel prize was given to Niels Finsen for his demonstration that sunlight therapy was effective against tuberculosis, then "the captain of the armies of death". Sanatoria to treat tubercular patients by sunlight were set up, and UV therapy became widely used in the medical world for many other conditions.

Both children and adults in hospital were often taken out into the sun during the day, and many hospitals were designed with verandas and French windows to make this possible. It was generally believed that sunlight was beneficial to health, and, in fact, the modern popularity of sunbathing dates from this period. Then in the 1930s the first modern drugs were discovered, and the drug era began. Sunlight went out of fashion among doctors. Modern hospitals are multi-storey buildings sealed off from the outside world. Ultraviolet therapy is still used for some skin conditions, but drug treatment is preferred – it's cheaper. Doctors have lost sight of the benefits of UV exposure.

In recent years dermatologists have become concerned about the persistent world-wide increase in the incidence of malignant melanoma since the 1930s. In some ways melanoma is similar to other types of skin cancer, which are known to be caused in part by exposure to the UV in sunlight. In other ways melanoma is quite different. It is more dangerous and affects younger people. Typical melanoma sites are parts of the body normally covered by clothing, and the connection between sunlight exposure and melanoma is complex. Research studies have consistently shown that continual exposure to daylight, as in people who work outdoors, reduces the risk of melanoma! Some studies find that "intermittent" exposure, as in sunbathing, increases the risk; others do not. Melanoma, unlike other types of skin cancer, cannot be produced experimentally by exposure to UV. A study in 1974 showed that within the USA there is more melanoma the nearer the equator and the higher the sunlight intensity. In the UK, melanoma is much more common in Scotland than in either England or Wales – and much less common in Northern Ireland! By the way, none of the research papers mentions depletion of the ozone layer, which for scientists is a completely separate issue.

Despite the confusing nature of the evidence, a public health campaign since 1987 has made claims such as the Department of Health's statement that "if people avoid excessive exposure to sun, they can prevent melanoma." Because of this campaign, history has been rewritten to say that sunbathing has nothing to do with health, and was invented in the 1920s by Coco Chanel, making a suntan merely a fashion accessory. A popular belief has also been created that melanoma is connected with ozone depletion. The "hole" in the earth's ozone layer is over Antarctica, not at present a popular destination for sunshine holidays. If the hole were to expand greatly, all life on earth might be threatened, but this is not happening. Measurements of sunlight intensity are available from the Meteorological Office for several weather stations in the UK for the last 36 years, and there is no sign of an upward trend. Sunbathing is no more dangerous now than in the past.

Meanwhile, in another part of the medical forest, there is strong evidence that exposure to sunlight reduces the risk of internal cancers.

Recent research studies in the USA and the former USSR have shown that breast cancer mortality declines with increasing sunlight intensity. An independent study of prostate cancer in the USA showed a similar correlation, as did an earlier (1980) study of colon cancer. These are major cancers, which between them in the UK kill 35 times as many women as melanoma, and 25 times as many men. So advising people to avoid the sun and use high factor sunscreens may tend to increase total cancer deaths even if it achieves some reduction in melanoma, which is doubtful.

There is no mystery about how this works. Vitamin D3 is known to be effective in inhibiting the growth of tumours, and UV is essential to the synthesis of D3 in the body. This synthesis is far more important as a source of this vitamin than the diet, except in countries like Japan whose typical diet includes fish, very rich in vitamin D. Probably the majority of people in Europe and America are marginally D-deficient, especially older people whose ability to synthesize vitamin D declines with age. Diet can be an effective source of vitamin D for cancer prevention, 400-800 IU daily being a relatively safe dose. But excess vitamin D, unlike vitamin C, is not easily eliminated from the body and too much of it can be toxic. The body's own synthesis, on the other hand, is self-regulating. Regular UV exposure is likely to promote long-term health for most people, especially the elderly. Everyone knows that sunshine makes you feel healthier - now we have hard evidence for one of the reasons why.

For too many people, the dermatologists' message comes across as "the sun is bad for you". It's not true: the sun is good for you. It's not so much that we over-expose ourselves on holiday as that we under-expose ourselves for most of the year. We should get as much daylight as we can every day, especially in Britain's cloudy climate, and through the winter. We should spend more time out of doors, because window glass absorbs a large fraction of the UV in natural light. For people who have to work under artificial light, full spectrum lighting (including UV) is very helpful. It's even possible that those sunbeds that dermatologists despise so much may have health benefits for some people during the winter. It's true that sunlight can be harmful in excess – but then, so can oxygen. This is not a good reason to prescribe the elimination of all exposure to oxygen, or to sunlight.

Notes & References

1 J M Elwood: Melanoma and Sun Exposure: Contrasts Between Intermittent and Chronic Exposure. World J Surg 16: 157-165, 1992.

Professor Elwood, one of the world's leading melanoma experts, surveys 23 studies of melanoma and UV exposure from 1979 to 1990. He remarks that "despite these reports, the ideal study of melanoma has yet to be done". Although he concludes that "the relationship between melanoma and sun exposure is now supported by strong evidence", he bases this conclusion on a very selective view of the evidence, and even the selected results show modest risk factors with wide confidence limits. These seem far too small to account for the observed increase in melanoma incidence. Dr Elwood refers to a "dilemma about public education", admitting that "if there is a complex curvilinear relationship with total dose, public education campaigns aimed at reducing total dose could, for some individuals with high exposures, increase the risk." He doesn't mention the ozone layer once. This paper is well worth reading in full.

2 F C Garland, C F Garland, E D Gorham and J F Young: Geographic Variation in Breast Cancer Mortality in the United States: A Hypothesis Involving Exposure to Solar Radiation. Preventive Medicine 19: 614-622, 1990.

These researchers found that breast cancer mortality declines with increasing sunlight intensity at 87 locations across the USA. After considering possible causal factors including diet, they conclude that "lack of exposure to UV sunlight can increase the prevalence of vitamin D deficiency [which] may place some populations at higher risk for breast cancer".

The same authors (minus Young) also did a study of " Sunlight and breast cancer incidence in the USSR" (Int J Epidemiol 1990 Dec; 19(4): 820-4) with similar results. Garland and Garland did an earlier study "Do sunlight and vitamin D reduce the risk of colon cancer?" (Int J Epidemiol 1980; 9: 227-231) which concluded that they do. They point out in the 1990 paper that "populations at high risk for breast cancer are also at high risk for colon cancer, and those at low risk for breast cancer are typically at low risk for colon cancer". Breast cancer is the no 1 cause of female cancer deaths.

3 Schwartz & Hulka. Anticancer Research 10: 1307-1310, 1990.

The authors found that prostate cancer mortality in the US reduces with increasing sunlight intensity; it is lower in white-skinned people than in blacks. They point out that Vitamin D3 is known to be effective in inhibiting the growth of tumours, and that UV is essential to the synthesis of D3 in the body. They consider dietary factors: "despite the supplementation of foods the vitamin D obtained from the diet generally is a negligible portion of the body's supply of this vitamin." They show that older people are more likely to be D-deficient because the ability to synthesize vitamin D declines with age. Prostate cancer is the no 2 cause of male cancer deaths in the US.

4 H G Ainsleigh: Beneficial Effects of Sun Exposure on Cancer Mortality. Preventive Medicine 22(1): 132-140, 1993.

This author surveys a substantial body of relevant research, including the Garlands' work (he seems to have missed Schwartz & Hulka), and evidence that the use of sunscreens can increase melanoma incidence. He concludes that "advising the public to seek regular moderate solar exposure is supported by a broad view of the available scientific research as an effective means of lowering cancer mortality".

5 D P Strachan, K J Powell, Aruna Thaker, F J C Millard and J D Maxwell. Thorax 1995; 50: 175-180.

These authors report that "a vegetarian diet is an independent risk factor for tuberculosis in immigrant Asians [in the UK]. The mechanism is unexplained. However, vitamin D deficiency, common among vegetarian Asians in south London, is known to affect immunological competence." The authors mention "speculation about a possible link between vitamin D deficiency and impaired host defence to human Mycobacterium Tuberculosis", citing a paper in Tubercle 1985; 66: 301-4.

6 Dr Frank Ryan: Tuberculosis: The Greatest Story Never Told, 1994, p322.

The TB researcher Karl Jahnke contracted the disease in the lymph glands of his neck, from contact with patients. He was treated with streptomycin without success, and was then sent to a sanatorium in the Alps. 'Here his doctor permitted him to walk out of doors but with the caution that he must at all costs avoid direct sunshine. "From then on, I walked only in the shadows. One day, in overcast weather, I walked to the top of the Nebelhorn, in the Allgäu. Suddenly, to my consternation, the sun came out from behind the clouds and, looking round, there wasn't a tree at this altitude to offer me the slightest shade. I got well and truly sunburnt. But from that moment on my tuberculosis began to improve. The enlarged glands grew smaller and smaller until they finally healed up altogether." '

The sanatorium movement for the treatment of TB started in the mid-19th century, and the regime was based on fresh air, good food and complete rest; patients were advised to avoid the sun. Betty MacDonald, The Plague and I gives a vivid account. Treatment by sunlight came later: see A Rollier, Heliotherapy.

7 Ozone depletion and intensity of sunlight

I have analysed yearly means of daily total global irradiation (calories per sq cm) from the Meteorological Office for several weather stations in the UK, calculating the gradient of the least squares straight line, and the results are shown in the table below.

Stations are listed from North to South. As you can see, there is no sign of an overall upward trend. Direct measurements of ultraviolet irradiation have been made only since 1988, but since UV is the most energetic part of the spectrum, one would expect any increase in UV intensity to be reflected in an increase in the total calories.

I had a telephone conversation on 16 June 1995 with Dr Colin Driscoll at NRPB (the UK National Radiological Protection Board), who confirmed that their UV monitoring since 1988 shows no upward trend. Measurements of UV in the Swiss Alps have been made over a much longer period, and Dr Driscoll says that no significant upward trend has been found there either. His explanation is that pollution of the troposphere in the Northern hemisphere has reduced UV penetration, masking the effect of ozone depletion - the two effects presumably being equal and opposite, which seems a remarkable coincidence. The absorption of UV by acid haze over cities is well documented, however (Gorham, Garland & Garland, Can J Public Health 1989; 80: 96-100), and may explain why sunlight intensity in London is lower than would be expected from its latitude.

Dr Driscoll went on to say that the observed UV penetration figures have to be weighted for health monitoring purposes by the action spectrum of the disease of interest; the NRPB weights its figures by the action spectrum for sunburn. Dr Driscoll is satisfied that the action spectrum of non-melanoma skin cancer is similar to that of sunburn, but says that the action spectrum for melanoma is unknown because there is no animal model for melanoma. Of course, most cancer research is carried out on animal models, usually a strain of laboratory mice which develops tumours very easily; by the way, much of the research on the effects of ultraviolet is done using UVC, which is all filtered out of sunlight before it reaches the earth (Radiological Protection Bulletin, Jan 1995 p17). I asked Dr Driscoll about the research of the Norwegian scientist Dr Moan, who has used a fish called Xiphophorus to calculate an action spectrum for melanoma. Dr Driscoll didn't seem to take this too seriously. Dr Moan's paper says that UVA is more important in the action spectrum than UVB, and therefore "the fluence of melanomagenic radiation from the sun at the surface of the earth will increase only moderately even if the ozone layer is completely removed" (J Photochem Photobiol B, Aug 1994; 24(3): 201-3).

J E Lovelock: The Ages of Gaia 1988, p164-70 gives a masterly account of the "ozone war" which followed his own discovery of the presence of traces of chlorofluorocarbons (CFCs) in the atmosphere in 1972. He points out that exposure to UV has both benefits and dangers, and remarks in passing "it seems that the incidence of multiple sclerosis varies with latitude reciprocally to that of skin cancer". Of the thinning of the ozone layer over the south polar regions, he says: "This event is entirely unexpected and in great contrast to the fact that over most of the world the level of ozone is either unchanged or even slightly increased."

Professor Rona MacKie is head of the Department of Dermatology at Glasgow University, chairman of the UK Skin Cancer Prevention Working Party and principal author of the research paper on which the 1987 public health campaign was based. To give Dr MacKie credit, she says: "At the present time the increasing incidence of all types of skin cancer including malignant melanoma is attributed to changing habits of sunlight exposure, not to any possible excess UV as a result of ozone depletion."

8 The aetiology of melanoma

The 'official' theory of melanoma has two parts to it. The first is that 'intermittent' exposure to strong sunlight increases the risk; Dr Elwood considers the evidence for this part. The second is that such exposure has increased over the last 60 years. There is no evidence for this part – no one has measured the incidence of sunbathing, and any assessment must be subjective. One could argue on the contrary that the average white person's total UV exposure has greatly declined over this period. I think that the risk factors for intermittent exposure are far too small to account for the steady increase in melanoma incidence and mortality. By the same token, I think that the reduction in risk for 'chronic' exposure is too small to account for this increase via UV deprivation.

So what are the causes of melanoma? The best evidence for the official line remains the epidemiological study by Lee, Elwood et al (Int J Epidemiol 3: 325-52, 1974). If we accept the Garland's work we must accept this too; the method is the same. There must be some link between UV exposure and increased melanoma. But I think there must be other factors that are more relevant to the historical increase – factors that no one is looking at. The one I'd like to see investigated is TV exposure, which has undoubtedly increased over the period in question, and which John Ott (Health & Light) did some work on. The only other reference I have found so far is to an experimental study by Omura & Losco in Acupuncture & Electrotherapy Res. 18(1): 33-73, 1993 Jan-Mar. They found substances linked to cancer in the (live!) bodies of humans during exposure to TV.

At one time it looked as if artificial light could be a factor. The paper by Dr Elwood which I cite reviews what few studies there are, and concludes that lighting is innocent. I think that the evidence is inconclusive, too few studies have been done, and most research is bedeviled by the assumption (clear to see in Dr Elwood's paper) that if artificial lighting is harmful, it must be because it emits UV radiation! Normal fluorescent lighting emits hardly any UV. John Ott suggested that harmful effects could be caused by the mercury lines in the emission spectrum, but no one seems to have followed this up.

9 Incidence of melanoma

Just how serious is the world-wide increase in melanoma? A figure often quoted is that incidence is doubling "in most areas" every 10 years; for example, this estimate is given by Dr MacKie in "Ultraviolet radiation and the skin" in Radiological Protection Bulletin, June 1993. It is difficult to see what is really going on, because there is a long delay in the appearance of national cancer incidence statistics. Doctors in the UK are encouraged to report all new cases of cancer to their regional cancer registries, and these data are collected by OPCS (the Office of Population Censuses and Surveys) for England & Wales, and by separate centres for Scotland and Northern Ireland. OPCS publishes cancer incidence statistics for England and Wales several years in arrears; the latest available are for the year 1989! The centre for Scotland works a lot faster; cancer registration statistics for Scotland up to 1990 were published in 1993. Cancer mortality statistics are more up to date, the figures up to 1992 being available early in 1994, and perhaps more reliable.

The Cancer Research Campaign's "Malignant Melanoma Factsheet Feb 1994" analyses these figures. The number of new cases increased by 156% over the fourteen years to 1988, the graph showing a marked acceleration from 1984 on, and "in Scotland over the sixteen year period 1974 to 1990 the number of registrations increased by over 240%". Newton & Redburn (BMJ, 310: 502-3, 1995) point out: "There is evidence that public health campaigns give rise to a short term increase in the apparent incidence of melanoma. The rise in incidence before 1988 in both sexes was probably partly the result of campaigns that took place mainly in 1987." Their study of the Oxford region showed a sharp decline in melanoma incidence in 1989, a result since confirmed at national level. With the help of the Scottish increase and the 1988 peak, the UK total has doubled over the ten years 1978-88, just like Dr MacKie's estimate; the previous ten years go over the horizon, but judging from the graph the increase in those years was less than double.

The mortality figures show a different picture again; the graph from 1974 to 1992 looks close to linear, with no increase for women from 1988 on. It has taken about 24 years to 1992 for mortality to double. Furthermore, mortality per million population in Scotland in 1992 is a little below that for England & Wales, while incidence per million in Scotland in 1988 is 29% above England & Wales! No one seems to have noticed the discrepancy between the increases in incidence and mortality, or proposed any explanation.