Page 98 - Vitamin D and Cancer

4 The Epidemiology of Vitamin D and Cancer Risk 85

were ³ 400 IU (summary RR = 0.92, 95%CI = 0.87–0.97; p(heterogeneity) = 0.14).
One of the studies with high intakes, the Nurses’ Health Study, is of interest because
vitamin D intake was updated every 2–4 years, which allowed for an improved esti-
mate of long-term intake [71]. That study, which was based on 3,482 cases of breast
cancer, found that total vitamin D intake (dietary plus supplementary intake) was
inversely associated with the risk of incident breast cancer (multivariate RR = 0.72;
95%CI = 0.55–0.94) for >500 versus £ 150 IU/day of vitamin D. Notably, similar
inverse associations were observed with other components of dairy foods, including
lactose and calcium, indicating the difficulty of teasing out the independent effects.
Nonetheless, total vitamin D intake had a stronger inverse association than did either
dietary or supplemental vitamin D intake individually, which suggested that vitamin
D was indeed the relevant causal factor.

4.5.3 Sun Exposure

The death certificate-based case–control study of cancer mortality described above
found that greater residential exposure to sunlight (RR = 0.74; 95% CI, 0.72–0.76)
and occupational exposure to sunlight (RR = 0.82, 95% CI, 0.70–0.97) were associ-
ated with reduced mortality from female breast cancer (n = 130,261 cases) [43]. The
study also found that the magnitude of the association between outdoor employ-
ment and reduced breast cancer mortality was strongest in regions of greatest resi-
dential sunlight (OR = 0.75, 95% CI, 0.55–1.03), suggesting that sun light exposure
was the primary reason underlying the reduced risk with outdoor employment.
A population-based case–control study of 972 cases and 1,135 controls con-
ducted in Canada, examined self-reported sun exposure behaviors at different age
periods in relation to risk of breast cancer [72]. The study found a significantly
reduced risk of breast cancer associated with increasing estimated sun exposure
from ages 10 to 19 (RR = 0.65; 95% CI, 0.50–0.85 for the highest quartile of out-
door activities versus the lowest; P for trend = 0.0006). Notably, the associations
from ages 20 to 29 years were weaker, and no evidence was observed for exposures
for ages 45–54 years. These results suggest that the relevant time for vitamin D
exposure and reduced breast cancer risk occurs primarily or solely during
adolescence.
A population-based case–control study was conducted based on 1,788 incident
cases of advanced breast cancer and 2,129 controls over the years 1995–2003
among Hispanic, African-American, and non-Hispanic White women from
California [73]. In this study, among women with light constitutive skin pigmentation,
those with high sun exposure index based on reflectometry had a reduced risk of
advanced breast cancer (RR = 0.53, 95% CI: 0.31, 0.91). However, among women
with medium or dark pigmentation, high sun exposure index was not associated
with risk. To explain these discordant findings, the investigators posited that these
measures based on reflectometry may reflect vitamin D status better in more lightly
pigmented women than in darker skinned women. Finally, in a relatively small