The Medical Minute: Risk -- it's all relative

January 12, 2005

By John Messmer
Penn State Family and Community Medicine
Penn State Milton S. Hershey Medical Center
Penn State College of Medicine

WARNING: This product can cause abdominal pain, intestinal bleeding, anemia, bloating, weight gain, rash, constipation, diarrhea, asthma and death.

Would individuals who read this warning on a product think twice about using it? Why is such a product even allowed on the market? Not only is this product available in stores everywhere, it is promoted vigorously in print ads and on television, despite the risks.

The product is cow's milk. Most people are aware that the risk of moderate milk consumption is low -- so low many people are not aware there are any risks. Most people would not consider milk so dangerous that it should be removed from the market. The reason is most of us have daily experience with milk and have a general concept in our minds of the risk versus the benefit of milk consumption. We may not be able to discuss lactose intolerance, the science of milk allergy or how cow's milk causes iron deficiency in infants, but we have a general sense of its risk versus benefit ratio.

On the other hand, most nonmedical people have little day-to-day experience with prescription medications. For most of us, the first time we think about medication risks is after our doctor prescribes something for us and warns us of potential side effects. Usually, the pharmacist includes information with the filled prescription about the potential hazards of the medication prescribed. Then there are the Internet Web sites which give ominous warnings about certain medications.

Many Americans assume the Food and Drug Administration (FDA) has determined whether a drug is safe enough for people to use. However, the withdrawal of the cholesterol lowering drug Baycol, the diabetes drug Rezulin and the anti-inflammatory Vioxx has shaken that faith a bit. This fear of adverse reactions may prevent many people from using appropriate treatments.

When a new drug is tested in people, any symptom they experience during testing is reported and listed as a possible side effect. The frequency of this symptom in people who took the drug is compared to the frequency in those on placebo. If it happens significantly more often in those on the drug, it is considered a side effect, even if it happens only 1 percent of the time.

Serious adverse events (SAEs), such as death, heart attack or stroke, also are studied. It takes a large number of people in a study to analyze this kind of problem. The more people in the study, the easier it is to find problems.

If a study has 1,000 people on a study drug and 1,000 on placebo and two people on the drug have a heart attack but only one on the placebo, you might say the drug causes the a heart attack twice as often as placebo, even though only one more person per thousand has one. The relative risk is doubled, but the absolute risk increase is low.

Recent news about Vioxx and naproxen were the result of this type of analysis. The news reported that heart attacks occurred in twice as many people on Vioxx as those on naproxen. That sounds like a lot, but there were 17 people per thousand treated who had heart attacks on Vioxx compared to seven on naproxen. On the other hand, it is generally accepted that Vioxx caused less stomach damage than naproxen. Therefore, if an individual is at high risk for stomach ulcers but low risk for heart attack, using Vioxx probably would have been a better choice than naproxen.

Just like deciding to drink milk despite its risks, the decision to take a medication must consider the benefits. Statin medications lower cholesterol and reduce the risk of heart attack. Many people believe that statins cause liver damage. In reality, true damage occurs no more frequently with statins than occurs naturally. Liver enzyme elevations occur in about seven per 10,000 people, compared to six per 10,000 on placebo. Compare this to the 42 percent reduction in heart disease deaths by simvistatin in one study (111 versus 189 out of 4,444 people).

When the press reports that a drug can cause some adverse effect, be wary. Do not assume it means the drug is dangerous or that the reported problem will definitely occur. Every medication has risks, but knowing the absolute risk versus the relative risk of the medication compared to the risk of the disease being treated can help individuals feel more comfortable taking it or requesting an alternative therapy.

(Media Contacts)

Last Updated March 19, 2009