Are cancer rates rising due to environmental factors or simply more reporting of cases?

Illustration by Slug Signorino

Cecil replies:

If I ever get this figured out, bub, I’ll invest the Nobel money in beer and comic books and spend the rest of my days on the beach — I’ll have earned it. The question is complicated by countless variables, and establishing the simplest facts can be maddeningly difficult. For now here’s what we know:

• For a while we thought some key cancer rates were flat or declining, but on further study it turns out they’re going (oops) up. This bombshell dropped in 2002 with the publication of a paper innocuously titled “Impact of Reporting Delay and Reporting Error on Cancer Incidence Rates and Trends.” The gist: A significant fraction of cases (3 to 12 percent, depending on cancer type) don’t get reported promptly to the leading U.S. cancer data registry — in fact, it can take anywhere from 4 to 17 years following initial diagnosis before most (99-plus percent) cancers are counted. The missing cases can make early trend reports misleadingly rosy. Melanoma incidence in white males, for example, was once thought to be easing; now analysts think it’s rising 4 percent per year.
• Notwithstanding such fine points, few dispute that (a) rates for many common cancers over the past 30 years are way up; (b) that’s partly a function of the fact that we’re living longer — cancer incidence rises rapidly with age between 40 and 80; (c) age having been adjusted for, environmental factors, including lifestyle choices such as diet, are the chief drivers behind the climb, accounting for anywhere from 65 to 90 percent of human cancer. Some alarming numbers: The white female breast cancer rate rose from 103 cases per 100,000 person-years in 1977 to 146 in 1998. The prostate cancer rate for black men shot up from about 141 in 1975 to 342 in 1993. The white male melanoma rate went from 9 in 1975 to 28 in 2001. Migrant studies show the importance of the environment in many of these increases. For example, Asian natives have a low incidence of prostate cancer, but for U.S.-born sons of Asian immigrants, who presumably have adopted Western habits, the rate doubles.
• That said, cancer rates don’t track environmental trends in any obvious way — in fact, the most striking characteristic of U.S. cancer charts since the mid-70s, when centralized data collection began, is how abruptly and often inexplicably the rates change, even after you smooth the curve to average out year-to-year fluctuations. For example, after rising steadily for years, prostate cancer spiked up a whopping 69 percent from 1989 to 1992, then dropped almost equally sharply. What happened? The current consensus is that a new test, prostate-specific antigen (PSA) screening, turned up many previously undetected cancers; presumably the rate fell once all the old cancers had been found. More mystifying is the sudden change in rates for colorectal cancer. After a steady rise, the rate leveled off for blacks after 1980 and began a prolonged drop for whites starting in 1985. (After the mid-90s the rate for whites flattened out or headed back up, depending on what numbers you believe.) The falloff has been credited to better diet, but does anyone really think U.S. eating habits improved or even stabilized during the era of supersizing?
• The message the cancer-prevention crowd harps on is that your chances of getting cancer are largely a matter of personal choice, so don’t smoke, exercise regularly, eat less fat and more fiber, etc. While this is surely true, people often worry more about the cancer dangers they don’t (or didn’t) know about, e.g., lung cancer due to asbestos. What percentage of cancer deaths arise from such insidious threats? Respected UK researcher Richard Doll offered the following estimates in 1998: Ionizing radiation, UV light, 5 to 7 percent; occupational exposure, 2 to 4 percent; air-, water-, and food-borne pollution, 1 to 5 percent. For comparison, tobacco accounts for maybe 29 to 31 percent of deaths and diet 20 to 50 percent. Conclusions: (1) the experts think cancer risk due to smoking and lousy eating habits dwarfs most of the stuff people get paranoid about; (2) expert opinion notwithstanding, given that 30-point spread for diet, many cancer risk estimates barely qualify as educated guesswork.

Cecil Adams

Send questions to Cecil via cecil@straightdope.com.