(Center for the Advancement of Health) - A new study estimates that the average 50-year-old woman has about a 1-in-18 chance of developing breast cancer in her next 20 years--half the risk implied by the oft-repeated statistic 1-in-9.

Many women overestimate their risk because of a faulty understanding of the numbers, which become more accurate when broken down by race and ethnicity. This study also suggests that projections over a shorter time period, 5 to 20 years, may be more meaningful than the lifetime risk statistic.

Lifetime risk is the likelihood that a newborn girl will develop breast cancer sometime during her entire life. The 1-in-9 statistic does not apply to women of all ages, explains Cyllene R. Morris DVM, PhD, of the Public Health Institute's California Cancer Registry in Sacramento, California.

Although breast cancer is much more common in older women than younger women, the risk of developing breast cancer during the remainder of an older woman's life is actually lower than the risk for a newborn.

An older woman has already survived several decades free of breast cancer, while a newborn is still at risk of developing breast cancer (or dying of other causes) until she reaches that particular age, Morris says in the paper.

Several studies have shown that women generally overestimate their risk of developing breast cancer. Even women who are counseled or educated on breast cancer may still misinterpret the 1-in-9 statistic as a short-term probability rather than as the lifetime risk of a newborn girl.

Morris and colleagues at the California Department of Health Services compiled data from the registry on 109,165 women in the state diagnosed with breast cancer between 1993 and 1997.

The study will be published in the April issue of the American Journal of Preventive Medicine.

They found that, on average, a 50-year-old woman has a 1-in-86 risk of developing breast cancer in the next five years. Most guidelines call for women to start getting screened regularly at about age 50, a recommendation the authors don't question.

In comparison, the average 40-year-old woman has a 1-in-189 risk of developing breast cancer in that time period and a 1-in-26 risk within 20 years. A 60-year-old woman still has only a 1-in-61 risk over 5 years and 1-in-14 over 20 years. These numbers assume the woman has not already had a breast cancer diagnosis.

At 50 years old, a Caucasian woman has a 1-in-15 risk of developing breast cancer in the next 20 years of her life, while an African-American woman has a 1-in-20 risk, an Asian/Pacific Island woman has a 1-in-26 risk and a Hispanic woman a 1-in-27 risk.

Within a 5-year period, a 50-year-old Caucasian women has a 1-in-75 risk of developing breast cancer. In comparison, an African American has a 1-in-98 risk, an Asian/Pacific Islander a 1-in-107 risk and a Hispanic a 1-in-133 risk in 5 years.

Risk estimates are projections of current rates into the future and are, therefore, valid only if these rates remain stable over time. In the short term, such an assumption may hold true, but is not realistic over a long period of time, Morris and colleagues say.

Incidence rates of breast cancer are unlikely to remain stable over time because of advances in screening technologies, which lead to more tumors being detected earlier, and improvements in prevention strategies, which reduce the number of women who develop breast cancer, they say.

In contrast, risk estimates based on current age and calculated for the next decade or two may be more accurate, less prone to misconceptions and, therefore, more meaningful to the public, they say, adding the cautionary not that breast cancer is still poorly understood and that age and race/ethnicity alone cannot predict the risk of developing cancer for an individual woman.