[Researchers at the American Cancer Society compared the risk of dying among overweight people in two of its studies-one done from 1982 to 1991 and another from 1998 to 2002.] Instead of decreasing, ... the relative risk of dying among the overweight actually increased over time.

The group who had reported skin cancer in 1982 had about a 25 to 30 percent higher death rate from cancers generally during the next 12 years than people who did not report skin cancer.

If you have a short follow-up, you still have a problem with people who don't know they're sick, ... As the follow-up lengthens, that's less of a problem because the people who don't know that they're sick die.

If there's blame to go around, most of the blame falls on the tobacco companies.

It's still a very good idea to eat a diet that's high in fresh fruits and vegetables, both for its direct benefit on heart disease and reducing adult-onset diabetes.

Lung cancer death rates have fallen 17 percent in men from 1990 to 2002. Both incidence and death rates have leveled off in women, so we are turning the corner.

The CDC seriously underestimated the risk of dying in the overweight by not adequately dealing with the bias due to smoking and disease.

We estimate that 90,000 deaths due to cancer could be prevented each year in the United States if men and women could maintain normal weight.

It really comes down to whether a study is representative or valid, ... You can't get a valid answer from smokers or sick people because anyone who's ever practiced medicine knows that being sick affects weight. We want to know about weight before, not after, people get sick.