JULY – AUGUST 2021
See full newsletter
In the late 70s, when I worked at a research institute in Guatemala, I observed that the incidence of preeclampsia (a hypertensive disorder of pregnancy) was very low compared to other countries in the region. And it occurred to me that it could be attributed to a “protective” effect of calcium, given that women have a very high intake of the mineral due to the Mayan custom of “nixtamalization” (cooking the corn grain with water and quicklime before cooking. grinding) to prepare the tortillas.
It was a good guess, but it had to be proven. And we were able to do this over several decades in studies in animals and humans, culminating in a randomized clinical trial in pregnant women and a subsequent global recommendation from the World Health Organization (WHO) to incorporate supplementation in all women. with low calcium intake to prevent this pathology.
The story illustrates the importance of designing and conducting randomized studies that evaluate the effect of different interventions, in order to support the implementation of evidence-based health practices. It is a long and difficult path, but necessary and ethically imperative. You have to take strong steps. There is no need to generate false expectations. When a professional justifies making decisions “based on my experience,” he is using inappropriate words.
Observations generate hypotheses. And an association, of course, does not imply causation. There are many examples in the history of medicine in which observational studies led to the postulation of mechanisms and the adoption of recommendations that, when passed through the sieve of rigorous clinical research, did not demonstrate benefits in health outcomes. And they even proved to be counterproductive, as happened with certain antioxidant vitamins.
And that is because, when cohorts are studied and variables are related to certain results, there may be “confounding” factors that are mediating the effects. Perhaps it is not this or that nutrient or habit that produces the benefit, but rather people in better health may be more inclined to consume or adopt them.
The same occurs with routine practices sustained over time because some “eminence” once proposed it or because it is taught in faculties, but without any type of research to support it. An example would be episiotomy, a vaginal incision promoted since the 20s of the last century in women who were having their first birth: In the 90s, we were able to carry out studies in Rosario and Neuquén that showed that it did not have the supposed benefits and that it was better not to do it.
Another aspect that complicates the issue is that, often, patients themselves demand a certain intervention (unproven and even harmful) as a right, which reinforces the propensity of health professionals to provide it. It is a vicious circle, a deformation.
There are no simple solutions. But it is necessary to reaffirm our commitment to the indication of evidence-based therapies or interventions, adopting for our practice the same critical spirit that guides the best scientific researchers: what is the basis of a certain statement? Are there alternative explanations that can explain it? To what extent can biases operate when evaluating the results?
By Dr. Belizán, IECS researcher, senior researcher at CONICET and associate professor at the University of Chapel Hill and Tulane, United States.