The world is on the cusp of a major epidemic of breast cancer, said Peter Boyle, head of the international prevention research institute in Lyon, France, at a conference on breast cancer in Oman last weekend. The incidence of breast cancer has tripled in the past 30 years and is expected to double again by 2030 with most of the burden falling on low and middle income countries. In 2008 there were 1.4 million new cases of breast cancer and 460 000 deaths. Mortality rates are coming down in high income countries but not across the world.

Breast cancer is the commonest cancer in women in every region of WHO, apart from in the South East Asia region where cervical cancer is commoner but declining. Although breast cancer occurs everywhere and there is no group of women at low risk (as there is with cervical cancer), it is not the same disease everywhere—because women in high income countries present with small tumours but women in low income countries often present with locally advanced inoperable breast cancer. Rajendra Badwe from the Tata Memorial Centre in Mumbai said that 80% of breast cancers in women from rural India were inoperable.

The increase in breast cancer is driven by urbanisation, rising obesity, adoption of Western diets, declining physical activity, changing patterns of reproduction, and the increase in life expectancy. Rising obesity is important said Boyle, but declining physical activity is more important. The role of tobacco remains controversial, but alcohol is very important—even with low consumption. One drink a day causes a 10% increase in incidence, and each extra drink causes a 7% further increase. The effect of stopping drinking is not known.

It’s now clear that the contraceptive pill does increase breast cancer, but the overall effect is probably small as the pill is taken by younger women, who have lower rates of breast cancer, and the increase diminishes quickly when women stop taking the pill. In contrast, hormone replacement therapy taken by women at the time of the menopause is more important, although its effect also falls quickly when women stop the therapy.

Worldwide women are tending to enter menarche younger, have their children later, have fewer children, and experience the menopause when older—all of which are factors leading to an increase in breast cancer. But these are not modifiable factors, in contrast to length of breast feeding, which is protective. Boyle said that more emphasis should be put on the benefits for the woman of longer breast feeding.

Boyle and others calculated the attributable risks for women in France in 2000, and hormone replacement therapy accounted for 11% of breast cancers, physical inactivity for 10%, alcohol for 9%, reproductive factors for just over 5%, and obesity for 5%. The importance of hormone replacement therapy has declined, but the fact that physical activity is twice as important as obesity is not well enough known.

Mammography will reduce mortality rates by a fifth, said Boyle, but BMJ readers will know that this claim is not free from controversy. Equally controversial is the information given to women about mammography, which critics say is heavily biased in favour of mammography. We learnt at the meeting that Oman plans to introduce mammography for women over 40, an age at which sensitivity is low and false positives high. Will Omani women, I wondered, be given unbiased information?

Mammography is of little use in India, said Badwe, because it is unaffordable and because most cases of breast cancer occur in younger women. As throughout the world, breast cancer is commoner in older women than younger women, but many women with breast cancer in India—and Oman—are young simply because there are many more younger than older women. Badwe also observed that older women in rural areas have no idea of their age, it’s just not important. Doctors have to try and work out their age by asking how many children and grandchildren they have.

Physical examination and breast awareness will be more important than mammography, in India, said Badwe, and a huge randomised trial is underway of promoting breast awareness and self examination and direct examination of the cervix.

There is strong evidence, said Boyle, that tamoxifen will prevent breast cancer in some women at high risk of breast cancer by as much as 50%, but it is hardly ever prescribed for prevention. Is this a block with doctors or women? Nobody seems to know.

Richard Smith was the editor of the BMJ until 2004 and is director of the United Health Group’s chronic disease initiative.