I’m sure you’ve noticed this. The news covers a new breakthrough drug that shows great promise, doctors are interviewed, victims of the disease are thrilled, everyone is encouraged. However, a few years later, follow-up studies show the drug to be ineffective or to have serious side effects that didn’t show in the initial study.
The U.S. Food and Drug Administration has the same problem. There have been cases of popular drugs approved for use that later had to be pulled due to a lack of effectiveness or side effects that were missed initially. An example of this is Vioxx, a very popular drug that doctors worldwide gave to more than 80 million people. It was withdrawn from the market when it was discovered that the drug damaged the heart and caused strokes.
How could this happen? We are told medical care is science-based, that we can rely on research to show us how to best live our lives, which drugs are best for which disease, etc. How could research be so unreliable?
Researchers are now studying the quality of medical research itself, and the results are disappointing, to say the least. Dr. John Ioannidis is considered a leader in this field, and his article “Why Most Published Research Findings Are False” is one of the most downloaded articles on the National Institute of Health’s website.
He estimates that the simplest (and most common) type of research studies are wrong 80 percent of the time. The more sophisticated studies he estimates to be incorrect 25 percent of the time, and that as much as 10 percent of the platinum-standard, large randomized trials reach the wrong conclusion. That is just the trials themselves. With a bias as to which studies eventually get published, he estimates a whopping 90 percent of the published medical information that doctors rely on to make decisions every day is wrong. Yikes.
Most of the “wrong” results he covers are “false positives,” in which the results seem much better than they really are. According to Dr. Ioannidis, this is usually because the researchers have an incentive to get positive results. The following are common reasons for bias in research.
Funded and published
“Publish or die” is a phrase used to describe this situation. Research is more likely to get published if it has an interesting, dramatic conclusion. Also, you are more likely to be funded for more studies if earlier studies you performed stand out somehow. But realistically, most studies do not show dramatic results. Many drugs fail to get better results than sugar pills, and many dietary supplements do not make big improvements in health.
Not surprisingly, a drug is much more likely to get a positive result from a study if the study is sponsored by the drug manufacturer. There is a lot of conflict of interest in the research community; a surgical team produced several studies of a particular spinal implant, and the doctors’ reviews of the product were glowing. However, it was exposed that the supplier of the implant, Medtronic, paid the doctors doing the evaluations $210 million over 15 years. Some of the payments were indirect, and went to corporations owned by the doctors rather than directly to them. Follow-up studies showed serious problems with the implant.
“Cherry picking” studies
Drug companies have been taking a lot of heat lately for only releasing studies of their drugs that have positive outcomes. Studies that show the same drug to be ineffective never see the light of day. Sometimes several studies are run at the same time, and only the ones that show good results are released.
One medical writer describes this as “that which occurs when publications in medical journals present a biased or misleading description of the design, conduct or results of a trial.” The drug company, Merck, has been accused of distorting information about Vioxx’s safety for five years, all while it was still being prescribed to patients.
Testing for lab results rather than health
It is much easier to monitor a test result than a person’s overall health. For example, there is no question that statin drugs lower blood cholesterol; that can be monitored by a simple blood test. But studies do not show nearly as much improvement in death rates from these drugs. Studies that look at health and mortality are more complicated, more expensive and take longer.
Dr. Ioannidis’ advice? Ignore most research.
As a wellness provider, my experience is that there is no drug or surgery that will ever prove to be as beneficial as a healthy lifestyle. For health problems other than medical emergencies, nothing is more effective than natural, drug-free wellness care. And yes, this is supported by the research, but I don’t know if those studies have been reviewed by Dr. Ioannidis.
Dr. Michael Noonan practices chiropractic, acupuncture and other wellness therapies in Old Town.