Stanford University Home

Stanford News Archive

Stanford Report, July 2, 1997

Study explains Bay Area's higher breast cancer

Study explains Bay Area's higher breast cancer rate

BY TIM STEPHENS

Ever since a 1994 report identified white women in the Bay Area as having the world's highest incidence of breast cancer, many have worried that some unknown hazard in the local environment might be increasing their susceptibility to this disease. Now, a study by Stanford epidemiologists should help dispel that concern.

The higher incidence of breast cancer here is entirely explainable in terms of known risk factors prevalent among women who live in the Bay Area, the new study shows.

Childbearing patterns ­ specifically, the number of children a woman has had and the age at which she bore her first child ­ account for most of the difference in breast cancer rates between the Bay Area and the rest of the country, the researchers report in the July 2 Journal of the National Cancer Institute.

"Women in the Bay Area tend to have fewer children, and to have them at later ages, than women in the rest of the United States, and those are well-established risk factors," explained Dr. Anthony Robbins, a physician pursuing a PhD in epidemiology at Stanford.

Robbins performed the study with Jennifer Kelsey, chief of epidemiology and professor of health research and policy, and statistician Sonia Brescianini.

The annual breast cancer incidence in the Bay Area is 114.6 new cases for every 100,000 women who have never before been diagnosed with the disease, according to the National Cancer Institute's 1988-1992 estimates. This figure, which includes women of all ethnic backgrounds, is approximately 25 percent greater than that of women in New Mexico, who have the nation's lowest incidence.

Within the Bay Area (defined in the study as the counties of San Francisco, Alameda, Contra Costa, Marin and San Mateo) there are also differences between breast cancer rates for women of different ethnic backgrounds. For example, black women in the Bay Area have a breast cancer rate about 21 percent lower than that of their white counterparts, while the rate for local Asian women is about half the rate for white women.

A widely publicized report on the rate of breast cancer in the Bay Area was issued in 1994 by the Northern California Cancer Center, which gathers regional data for the National Cancer Institute's Surveillance, Epidemiology and End Results (SEER) program. Using data from an international study comparing cancer rates in 20 countries around the world, the report noted that white women in the Bay Area have a breast cancer rate 50 percent higher than women in most European countries and five times higher than women in Japan. The report was covered extensively by local media.

These regional differences in breast cancer incidence are not a new phenomenon, said Robbins. He and Kelsey based their analysis on incidence rates from the SEER program for 1978 through 1982, plus risk factor data from a large study conducted in roughly the same period, known as the Cancer and Steroid Hormone (CASH) study. The CASH study gathered detailed information on many known or suspected breast cancer risk factors from women in eight SEER regions, including the Bay Area. Compared with the other seven regions, the breast cancer rate in the Bay Area was about 14 percent higher for white women and 10 percent higher for black women.

"These differences have not changed much from the early '80s into the '90s, so San Francisco does have a persistently elevated breast cancer incidence," Robbins said.

The use of data from 1980-82 is a major strength of the new study, according to Robbins. Since that time, the use of mammography for early detection of breast cancer has increased dramatically, resulting in a large increase in breast cancer incidence rates. The increase in mammography screening, however, may not be uniform throughout the country or even among different groups in the same region, and this makes it harder to compare incidence rates.

"To a large extent, the incidence rate nowadays is determined by how aggressively women are encouraged to get screened, so we wanted to go back to a time when screening would not confuse the issue," Robbins said.

In many respects, he found, Bay Area women as a group ­ regardless of ethnic background ­ differed from women in the other seven SEER regions in ways that would be expected to increase their risk of breast cancer. A host of studies have linked a woman's risk of developing breast cancer with certain reproductive factors, such as her age at the time of her first menstruation, age at menopause, age at first pregnancy, and number of children. Bay Area women, on average, differ from the rest of the country in all of these characteristics.

Early menstruation, late menopause, late age at first live birth, and low number of pregnancies all may increase a woman's risk of breast cancer by affecting her lifetime exposure to the hormone estrogen. For example, the risk of breast cancer for a woman who bears her first child after age 30 is nearly double that of a woman who bears her first child before age 20. The average age at first full-term pregnancy in the Bay Area was 23.3 years, compared with 21.8 years in the other SEER regions. (These figures may seem low, but they do not represent current childbearing; rather, they represent the first pregnancies of women 20 to 55 years old who were surveyed in the early 1980s.)

Bay Area women, as a group, also consume more alcohol than women in other parts of the country, Robbins noted. The role of alcohol in breast cancer is controversial, however, so he recalculated the analysis without considering alcohol as a factor and found that this recalculation did not substantially alter the outcome. For both black and white women, adjusting for the prevalence of known risk factors statistically eliminated the increased risk of breast cancer associated with the Bay Area.

This study did not directly address the issue of ethnic differences in breast cancer incidence. Possible explanations for such ethnic variations, Robbins said, include not only differences in reproductive factors, but also genetic differences, dietary factors and differences in the use of screening mammography.

The new findings should be reassuring to women in the Bay Area, Robbins said. "The data," he concluded, "are not consistent with an environmental cause for the higher rate of breast cancer in the Bay Area, which is what many women had feared."

Unfortunately, aside from alcohol consumption, most of the risk factors that appear to account for the Bay Area's higher breast cancer rate are difficult or impossible for women to modify. In that respect, it is hard to take much comfort from the study, Robbins acknowledged.

"However," he said, "the take-home message is that there's no evidence the higher risk of breast cancer in Bay Area women is due to the external environment. Women don't need to fear that their risk will increase just because they live here. I think that's important for women to know."

While this study may explain why breast cancer is more common in the Bay Area than in other parts of the nation, the higher rate of breast cancer in the United States as a whole compared with other countries remains an issue of intense interest to epidemiologists. Differences in diet, culture, environmental factors, genetics and health care practices are among the many factors that may be responsible, Robbins said.

He also noted that since breast cancer rates throughout the United States are so much higher than in other countries, any environmental causes of breast cancer in this country would need to have a nationwide distribution in order to explain the international differences. "To be consistent with the data," Robbins said, "any serious environmental causes would have to be affecting women in all regions of the United States, not just women in the Bay Area." SR