Tuesday June 28 2011
The study did not directly measure vitamin D levels
“Women who spend just three hours a day in the sunshine can halve their risk of developing breast cancer,” reported the Daily Express. It said that a study has shown the benefits of vitamin D and that “exposure to sunlight for 21 hours a week between April and October significantly cuts the chances of developing a tumour”.
The news report was based on a study carried out in Ontario, Canada. The study compared the amount of time spent outdoors during four life periods – teens, 20s and 30s, 40s and 50s and 60–75 years old – in women who developed breast cancer and women of a similar age who did not.
The researchers compared the risk of breast cancer in women who spent less than 6 hours outside a week, with women who spent over 21 hours outside in each stage of life. They found that the women who spent longer outside had 26–50% lower odds of breast cancer.
This was a relatively large study, but it had several limitations related to its design. The average age of the women was 56 and they had to recall the amount of time they had spent out of doors over most of their lives, which increases the likelihood of error. In addition, vitamin D levels were not measured but estimated. Further research would need to establish whether vitamin D levels are associated with the effects observed.
Where did the story come from?
The study was carried out by researchers from Cancer Care Ontario. Funding was provided by the Canadian Breast Cancer Research Alliance. The study was published in the peer-reviewed Journal of American Epidemiology.
Both the Daily Mail and the Daily Express did not make it clear that the study looked at the relative odds of breast cancer rather than absolute risk, which may lead people to misinterpret the results. Additionally, the study did not measure vitamin D directly, so it is not possible to say that vitamin D is responsible for the effects seen, as the newspapers suggest. Other factors may have influenced the risk of breast cancer.
What kind of research was this?
This study investigated whether there was an association between vitamin D production from exposure to sunlight and breast cancer risk. The researchers said that recent studies have suggested that vitamin D may be associated with reduced risk of breast cancer, but that these studies have only looked at dietary vitamin D levels.
In this population-based case-control study, the researchers wanted to see whether there was an association between breast cancer risk and time spent outdoors, ultraviolet radiation levels where the person lived, skin colour and sun-protection practices.
What did the research involve?
The researchers used data from the Ontario Women’s Diet and Health Study. In this study, the Ontario Breast Cancer Registry was used to identify women who were 25–74 years old who developed breast cancer in 2002 and 2003. The researchers contacted 4,109 of these women, and 3,101 of them took part in the study in 2003 and 2004. As a control group, women of a similar age who did not have breast cancer were randomly selected from households in Ontario, and 3,420 of them completed the study.
The women were asked to complete a questionnaire about risk factors for breast cancer and to complete a food-frequency questionnaire to record their dietary habits. The researchers used questions on ethnicity or racial background as a substitute for skin colour. Ninety percent of the study participants were Caucasian, so skin colour was classified as Caucasian or non-Caucasian (6% were Southeast or South Asian, 2% black, 1% aboriginal, and less than 2% had other skin colours).
The participants were asked about variables related to sun exposure during four periods of their lives: teenage years, 20s and 30s, 40s and 50s and 60–75 years old. Women were asked how much time they spent outdoors each weekend or weekday, what sun protection (such as sunscreen or wearing long sleeves) they used and where they lived (the latitude and longitude was used to estimate how much UV light the participants were exposed to). The researchers said that between November and March, the sun in Ontario is not sufficient to produce vitamin D. As such, they only looked at sun-exposure frequency from May to September.
Each woman was given a solar vitamin D score for each of her four periods of life. This score took into account hours of ultraviolet exposure per week, skin colour and sun-protection practices.
In their statistical analysis, the researchers used a technique called logistic regression to calculate how much the solar vitamin D score was associated with risk of cancer at each age period. They also gave people a cumulative life score, by classifying sun exposure as high (greater than average) or low (less than average) and combining all the periods.
The researchers identified factors other than sun exposure that could be associated with breast cancer risk and could potentially influence the researchers’ calculation of how much the vitamin D score predicted breast cancer risk (confounders). These were: the women’s marital status, education, ethnicity, body mass index, smoking status, amount smoked, breastfeeding, lactation, age of first period, oral contraceptive use and duration of use, whether the women had given birth and their age at childbirth, age of menopause, hormone replacement therapy use, family history of breast cancer or a history of non-cancerous breast disease, screening mammogram uptake, alcoholic drinks, dietary fat and calorie intake, physical activity and the amount of vitamin D and calcium they acquired from food and supplements.
What were the basic results?
The average age of the women in the study was 56 years old. Most women were postmenopausal (68% of cases and 64% of controls).
The researchers compared the chances of having cancer between the women who spent the most time outdoors (more than 21 hours a week) and those who spent the least time outdoors (less than six hours).
- Women who spent more time outdoors in their teenage years had a 29% lower risk of breast cancer than those who spent less time outdoors (odds ratio [OR] 0.71, 95% confidence interval [CI] 0.60 to 0.85).
- Spending more time outside in their 20s and 30s gave women a 36% lower risk of breast cancer than those who spent less than six hours outdoors (OR 0.64, 95% CI 0.53 to 0.76).
- Women in their 40s and 50s who spent more time outside had a 26% lower risk (OR 0.74, 95% CI 0.61 to 0.88).
- Spending more time outside between the ages of 60 and 75 lowered the odds of breast cancer by 50% (OR 0.50, 95% CI 0.37 to 0.66).
These results were not adjusted for confounders.
The researchers then looked at the risks associated with solar vitamin D score. They compared women with solar vitamin D scores that were in the top 25% with women whose scores were in the lowest 25%. These calculations were also adjusted for age.
The results showed that:
- Women who had higher solar vitamin D scores in their teenage years had a 21% lower risk of breast cancer compared to women with lower scores (OR 0.79, 95% CI 0.68 to 0.91).
- Women who had higher scores in their 20s and 30s had a 24% lower risk of breast cancer compared to women with lower scores (OR 0.76, 95% CI 0.65 to 0.89).
- Women who had higher scores in their 40s and 50s had a 25% lower risk of breast cancer compared to women with lower scores (OR 0.75, 95% CI 0.64 to 0.88).
- Women who had higher scores between the ages of 60 and 75 years old had a 41% lower risk of breast cancer than people with lower scores (OR 0.59, 95% CI 0.46 to 0.76).
How did the researchers interpret the results?
The researchers said that time spent outdoors during multiple periods of life and their proxy measure of vitamin D from sun exposure were associated with a reduced risk of breast cancer. They said: “it is plausible that vitamin D production mediates the inverse association observed between sunlight exposure and breast cancer risk, yet future studies are needed to confirm this risk."
This was a relatively large population-based case-control study. The findings indicated that women who spent a lot of time outdoors had a lower risk of breast cancer than those who spent very short periods of time outdoors.
The large size of this study is a strength, but the study also had several limitations, which affect how it could be interpreted for the UK population:
- Most (90%) of the study’s participants were Caucasian. All other non-Caucasian skin colours were grouped together. Further studies would be needed to see whether the findings vary for women of different ethnic groups.
- The researchers identified numerous potential confounders, but did not adjust for them in their analysis. They said that they did not do this because, individually, the confounders did not change the odds ratio by more than 10%. However, it is possible that the women’s risk may have been influenced by several confounders, each of which may have had a small effect but which together may have significantly affected the results. Factors that are thought to influence breast cancer such as family history of breast cancer, a younger age when the women began their periods, older age at menopause and decreased physical activity levels were linked with breast cancer risk, but their influence did not change the odds ratio by over 10% and so the results were not adjusted for these factors.
- The researchers point out that their solar vitamin D score was only a proxy measurement of vitamin D derived from the sun. To determine actual vitamin D levels would have required a blood test. Therefore, as the researchers point out, it is not possible to say for certain whether vitamin D is responsible for the effect seen.
- The women were asked to recall their past sun exposure throughout most of their lives. This raises a strong possibility that error may have been introduced. Ideally, a cohort study that follows women up over time would allow this sort of risk factor and vitamin D levels in the blood to be measured from the start.
This type of study can identify possible factors that may be associated with the risk of disease. However, the study’s limitations, particularly its reliance on the women to recall their sunlight exposure, means that further research will be needed to see whether exposure to sunlight affects breast cancer risk.
Analysis by Bazian.