| pubmed-article:1959794 | pubmed:abstractText | The descriptive and analytical epidemiology of ovarian cancer is reviewed, starting from the substantial geographical differences, with high rates in North America and Europe and low rates in developing countries and Japan, although, on a worldwide scale, almost 50% of 140,000 total cases occur in developing countries. Over the past decades, incidence and mortality rates have remained approximately stable in high-risk areas, but have generally tended to rise in low-risk areas. In the past, ovarian cancer was more common in higher social classes, but sociocultural differences seem to have flattened off over recent decades. In etiological terms, the protection afforded by multiparity and oral contraceptive (OC) use is well established, with relative risks (RRs) of the order of 0.5 for multiparae and OC users. There is also consistent evidence that risk increases with late age at menopause. Less consistent and weak, if any, are the effects of age at menarche and first birth, although there is hint that a first birth over age 35 is not protective. It is conceivable that diet can play an important role in ovarian carcinogenesis, but only scattered data are available to suggest that high fat consumption may represent an indicator of risk. Available knowledge on ovarian cancer epidemiology is also discussed in relation to models of carcinogenesis and implications for prevention and public health. | lld:pubmed |
