Can results of research be replicated in the real world?


October 02, 1990|By B. D. Colen

Is it possible for medical research to be conducted too carefully? May researchers be so skilled and their investigation so carefully crafted that their findings cannot be realistically applied to the daily practice of medicine?

Even though a study is carried out under ideal conditions and by the best practitioners in the field, the results might end up being applied under less than ideal circumstances by some practitioners who are, at best, mediocre; the results of incredibly accurate research then become, in effect, inaccurate.

The question comes to mind from a recent study published in the New England Journal of Medicine -- a study whose findings may well alter the course of medical practice.

Researchers at Harvard University Medical School and Boston's Brigham and Women's Hospital -- perhaps the premier women's hospital in the country -- set out to determine if ultrasound can provide a safe and effective alternative to using amniocentesis to determine for certain if a developing fetus has a neural tube defect.

An amnio, as it is commonly known, involves inserting a long needle through the woman's abdomen, through the wall of the uterus, into the amniotic sac, where a sample of amniotic fluid is withdrawn in order to test for, among other things, neural tube defects.

The problem is that, depending on what figures you choose to believe, amniocentesis causes a miscarriage rate of 0.5 percent to 1 percent.

Now comes the report from the Harvard researchers, who say that by using ultrasound alone they successfully detected 51 of 51 neural tube defects. Dr. Beryl Benacerraf, the chief researcher, believes that, overall, ultrasound would prove between 94 percent and 100 percent effective in detecting neural tube defects.

Given that the ultrasound process is not known to pose any risk to the developing fetus or the woman carrying it, and given that the amnio miscarriage rate is 10 times greater than the ultrasound failure rate, ultrasound would certainly appear to be the technology of choice for screening for neural tube defects.

If . . .

If every ultrasound technician were as technically skilled as Beryl Benacerraf. If every ultrasound was done at Brigham and Women's Hospital. If every ultrasound was done with the sophisticated equipment used in this study.

And that brings us back to the original question of whether a study can be done too carefully.

Dr. Benacerraf is widely acknowledged to be the Isaac Stern of ultrasound, and the equipment he uses is the ultrasound equivalent of a Stradivarius. Compared with him, the average ultrasound technician is the second violin in a small city symphony, and his equipment is only a reasonably good professional instrument.

Clearly, then, what Dr. Benacerraf and his colleagues have done is raise interesting questions for other researchers to pursue in the real world of obstetrics. What, one wonders, would the results look like if a similarly designed study was carried out in a number of suburban medical offices?

Rather than have a direct effect on the practice of fetal-maternal medicine, what such "perfect" studies should do is remind us that the best conducted studies of mice and men are often too well conducted to have any meaning in the real rat race of the world in which we live.

*B. D. Colen is the senior correspondent for science and medicine at Newsday.

*Los Angeles Times Syndicate

We welcome your opinions about health issues. All columns must be typed and 650 words or less. Include your name and daytime phone number.

Baltimore Sun Articles
Please note the green-lined linked article text has been applied commercially without any involvement from our newsroom editors, reporters or any other editorial staff.