What should be done, if anything, about the flood of medical information from news stories, popular magazines, TV shows, advertising, and even our own doctors?
Is information overload even a bad thing?
Moreover, are we missing the forest while looking at the trees? Very often, I'd say yes.
It's often best to ask, "What are we really concerned about?" For example, too much cholesterol in the blood can lead to hardening of the arteries, which can lead to plugged-up arteries in the heart or brain, resulting in a heart attack or stroke.
The effect of the abnormally elevated cholesterol level is the concern, not the cholesterol itself, because lowering abnormal cholesterol levels can reduce the likelihood of disability or death from blocked arteries.
So the goal is to prevent illness, not to reduce cholesterol for its own sake. And, there are always tradeoffs. Taking medicine takes time, money, and other resources away from other uses. Plus, the medicine might not work or even cause additional medical problems.
Part of the problem is that some of the most scientifically sound information comes from the most focused and therefore most limited experiments or observations. A researcher might investigate how a particular treatment affects 47-year-old left-handed Icelandic women. But what those results mean to natives of the Philipine islands or America?
News reporting of results of scientific studies is also usually limited, partly because most reporters have little background in science and because the "news" naturally favors new, sensational, or weird results.
Most reporters also have limited knowledge about statistical analysis. So when a scientific study reports "statistically significant" numbers of cancer cases in a given area, the resulting news story usually doesn't mention that this can mean there's one chance out of 20 that the result is just a statistical fluke or a random variation and doesn't mean anything at all.
So, if a researcher found that five kinds of cancer had a "statistically significant" higher-than-expected incidence in a certain neighborhood, it might make for spectacular headlines. But if that researcher analyzed 100 different cancers, and "significance" was at the usual 95 percent level, we would actually expect to find a random 5 percent of the cancers in statistically significant numbers.
In other words, the "Garbage In, Garbage Out" process is probably at work. And, there would likely be five cancers with a lower-than-expected incidence. But that's not as newsworthy.
Even these focused and limited medical studies cost a lot of time and money. And the results are limited to the particular treatment and the particular patient population.
For example, finding a new drug to heal human illness requires a huge effort. A drug candidate has to be isolated or created in the lab, tested with animals under controlled laboratory conditions, and then survive a gauntlet of multiple levels of clinical testing with real human beings.
After all this, the Food and Drug Administration (FDA) and other agencies typically take many months or years before they can review the data and officially "approve" the drug for more general human use. The process takes uncountable hours of work and analysis, plus about a billion dollars, for just one drug.
Complicating things further, many people have several medical conditions. And many medical conditions and drugs interact.
For a simple example, a bleeding stomach ulcer can result in loss of blood and anemia. Both the ulcer and the anemia need medical treatment. How do the medicines and treatments for these different conditions interact? Very few scientific studies investigate the relationship of two such variables.
There are hardly any scientific studies to guide the doctor and patient when a patient is taking four or more different medicines. In other words, each such patient is a unique experiment.
Understanding these limits and at the same time relating the findings of medical science to treating the individual patient makes medicine as much an art as a science.
If medicine were only a biological science, we could program computers to make diagnoses and prescribe treatment, across the board.
This doesn't happen because although humans share a common nature, every one of us is unique in many ways, such as in our DNA information, susceptibility to illness, and responsiveness to medicines. The doctor uses his best medical judgment, often an educated guess, on how to treat complicated patients.
Today, with so much more information readily available via the Internet, medical options are greater than ever before.
You're best off if you do find a knowledgeable doctor you can trust to help analyze all the available information and choices.
Cookie-cutter approaches are dandy for cutting cookies but not for dealing with human beings.
Editor's Note: Robert J. Cihak wrote this week's column.
