Cancer is one of our scariest diseases. That’s because it is the second-leading cause of death in the U.S., killing nearly 600,000 people each year.

It is not surprising that many Americans do everything in their power to avoid exposure to toxins that might increase their chances of getting cancer. Millions have quit smoking. Others buy organic food to reduce exposure to herbicides and pesticides.

Despite such proactive steps, there may be cancer-causing chemicals in one of the least-suspected places – medications. Few people realize that many of the medicines they swallow every day may contain compounds that are under a cancer cloud.

The Food and Drug Administration often requires pharmaceutical companies to test their compounds for carcinogenicity in animal models. Not infrequently, medications come back positive for an elevated risk of tumors. That rarely disqualifies them from the drugstore.

Hormone replacement therapy (HRT) is a classic example. Studies have clearly linked estrogen plus progestin with breast cancer (Journal of the National Cancer Institute online, March 29, 2013). The FDA even has a black-box warning about this risk.

Most of the time, however, the FDA doesn’t seem to know what to do about cancer concerns. The popular diabetes drug Actos (pioglitazone) has been banned in France and Germany because of research suggesting an increased risk of bladder cancer. The official U.S. prescribing information describes reports of bladder tumors in rat studies and bladder cancer in humans, but leaves patients and doctors confused about whether the risk is worth worrying about. Other diabetes drugs (Byetta, Januvia, Onglyza and Victoza) have been linked to pancreatitis and precancerous cellular changes in the pancreas.

Sometimes FDA staffers fight among themselves about cancer risks. Dr. Thomas Marciniak is a patient-safety expert within the FDA. He recently raised a red flag about a potential cancer risk linked to popular blood pressure medications called ARBs (angiotensin receptor blockers). This includes medications such as Benicar (olmesartan), Cozaar (losartan) and Diovan (valsartan).

According to a May 30 Wall Street Journal report, Marciniak sent a message to high-ranking FDA officials stating: “The FDA needs to inform patients and physicians about the ARB lung-cancer risks. The FDA must act now.” A 2011 analysis by the agency found no increase in cancer. Marciniak’s boss told the Journal, “We have no reason to tell the public anything new.”

If the FDA cannot agree about drugs and cancer, it is hardly any wonder that physicians and patients get confused. Even when the agency requires cancer warnings, most people don’t know what to make of these cautions.

The popular heartburn medicine omeprazole (Prilosec) carries a notice about abnormal cell growth and stomach tumors (carcinoids) in rats. Does that pose a problem for people?

Dozens of other drugs carry some sort of cancer warning. They include skin creams for eczema (Elidel and Protopic) and powerful biotech drugs for rheumatoid arthritis (Cimzia, Enbrel, Humira and Remicade).

When the FDA requires a cancer warning in the prescribing information, it leaves everyone in a quandary. If there is a real hazard, such drugs should be restricted. If the danger is trivial, we should be told the risks are not relevant to people so patients won’t worry needlessly.

Joe Graedon is a pharmacologist. Teresa Graedon holds a doctorate in medical anthropology and is a nutrition expert. Their syndicated radio show can be heard on public radio. In their column, Joe and Teresa Graedon answer letters from readers. Email them via their Web site: