Cancer risk greater for baby boomers> Environmental causes suggested

WASHINGTON — A white man of the baby-boom generation has about twice the risk of developing cancer as his grandfather, and a white woman of the same age has about a 50 percent greater risk than her grandmother, according to a new study.

Even when cancers caused by smoking are disregarded and the aging of the population is accounted for, an upward trend of malignant disease this century is still evident in both sexes, researchers report in Wednesday’s Journal of the American Medical Association.

The findings “strongly suggest there are preventable causes (of cancer) out there that remain to be identified,” said Devra Lee Davis, an epidemiologist at the Department of Health and Human Services who headed the study. One possible cause, she and her co-authors speculate, is the presence of unspecified cancer-causing chemicals in the environment.

The rise in cancer has been concurrent with a steady fall in death from cardiovascular disease over the past four decades. The researchers are confident, however, that a person’s greater risk of getting cancer now is not simply a function of the decreasing chance that he or she will get heart disease. Whether the findings hold for other racial groups is not known.

Davis, Gregg E. Dinse of the National Institute of Environmental Health Sciences, and David G. Hoel of the Medical University of South Carolina used data collected by the National Cancer Institute’s Surveillance, Epidemiology, and End Results program. SEER gathers cancer statistics from nine regions in the United States that together encompass about 10 percent of the country’s population. Data on death from heart disease, stroke, and nonmalignant diseases came from the National Center for Health Statistics.

The researchers looked specifically at cancer statistics for the period 1973 through 1987 for persons aged 20 to 84. In addition, they reconstructed the cancer rates for various age groups in decades preceding that 15-year period, using statistical manipulations.

Cancer is predominantly a disease of old age, and as people live longer they are more likely to contract it. The cancer incidence and death rates used in the JAMA study, however, took this into account. They were calculated as simple fractions: The number of new diagnoses (or deaths) for a given age group divided by the size of the population in that group. This made the rates “age-adjusted.”

The researchers divided cancer into two groups: those types (such as lung, larynx, esophagus and mouth) likely to be caused in part by smoking; and all other types (such as colon, breast, cervix, lymph node) in which smoking plays little role.

They then looked at two trends: the change in cancer rates during the 15-year period; and changes in the risk of getting cancer among different age-group “cohorts” of people born between 1888 and the mid-1950s. Only data on whites were used because cancer statistics for other racial groups for 1973-1987 is not reliable, the researchers said.

Among the findings:

In the 15 years between 1973 and 1987, the overall mortality rate dropped 19 percent for people in the 65- to 74-year-old age group. Deaths from cardiovascular disease dropped 36 percent, but deaths from cancer rose 8 percent during that period.

Put another way, in 1973 more than one-half of the deaths in the 65-74 group were from heart disease, and one-quarter from cancer. Fifteen years later, when there were substantially fewer deaths per 100,000 persons in that age bracket, fewer than half the deaths were from cardiovascular causes and about one-third were from cancer.

During the 15-year period, the rate of new smoking-related cancers in all women rose by 50 percent, but stayed steady in men. The rate of cancer not related to smoking, in contrast, stayed stable in women but rose about 20 percent in men.

Women born in the 1920s and 1930s had six times the risk of developing a smoking-related cancer as did women born between 1888 and 1897 — a finding that reflects the explosive increase in women’s smoking in midcentury.

This trend appears to have peaked, with women born in the late 1950s now showing about a fivefold risk compared to those born at the end of the 19th century.

Men born between 1948 and 1957 have three times the chance of contracting a nonsmoking-related cancer as did men born shortly before the turn of the century, and about two times the chance of developing any type of cancer.

For women, the risk of nonsmoking-related cancers peaked with the cohort born between 1913 and 1922. It has remained steady among women born in the three decades thereafter — a rate about 30 percent higher than that seen for women born in the 1890s.