American Cancer Society Study Links 12% of Colorectal Cancer Deaths to Cigarettes

February 1, 2001

In the largest prospective study of cigarette smoking and colorectal cancer mortality, researchers from the American Cancer Society report finding strong evidence that cancers of the colon and rectum are, in part, smoking related.

In the largest prospective study of cigarette smokingand colorectal cancer mortality, researchers from theAmerican Cancer Society report finding strong evidence that cancers of the colonand rectum are, in part, smoking related.

An analysis of the American Cancer Society’s Cancer PreventionStudy II appeared in a recent issue of the Journal of the National CancerInstitute (92:1888-1896, 2000). It concluded that not only is long-termcigarette smoking associated with an elevated risk of dying of colorectalcancer, but that about 12% of 1997 deaths from this cancer may be attributableto cigarette smoking.

Lowest Rates Among Nonsmokers

Ann Chao, PhD, Michael J. Thun, MD, and colleagues at theAmerican Cancer Society report that colorectal death rates were highest amongcurrent smokers and lowest among those who never smoked. "Clear benefitswere also observed among those people who had quit smoking; the longer ago, thehigher the benefit," said Dr. Chao.

In the study, current male smokers had a 32% higher death ratethan nonsmokers, and current female smokers had a death rate of 41%. The higherdeath rates increased with duration of the habit and with the number ofcigarettes smoked daily.

The rates were higher for current smokers who began smokingcigarettes at younger ages, than among those who smoked longer and morecigarettes per day.

Study Data

The Cancer Prevention Study II began enrolling approximately1 million Americans in 1982. Dr. Chao and her coauthors analyzed data on312,332 men and 469,019 women, among whom 4,432 died from either colon or rectalcancer. Data from these participants were controlled for a number of potentiallyconfounding variables including alcohol use, physical activity, family historyof colorectal cancer, vitamin use, and dietary factors.

"The size of the study allowed us to examine, in detail,gradients in smoking behavior separately in former and current smokers,"said Dr. Chao. "It also permitted us to estimate the percentage ofsmoking-related colorectal cancer deaths in the general population, which wouldhave been about 12% in 1997, or more than 6,800 people for that year," shesaid.

"The smoking epidemic in women began decades later than inmen," said Michael J. Thun, MD, vice president of epidemiology andsurveillance research for the American Cancer Society. "This may explain,in part, why the trends in colorectal cancer incidence and death rates differedbetween genders during the 1950s, 1960s, and 1970s, with male rates being higherthan female rates. As the smoking rates increased for women, the colorectalrates became very similar for both genders," said Dr. Thun.