When it comes to nutrition research, study designs need to be more holistic, says eminent scientist
When a scientist involved in landmark nutritional studies says there’s a problem with clinical trials, you sit down and listen.
The problem lies in how we take gender differences into account when designing clinical trials and interpreting their results, said Howard D. Sesso, ScD, MPH. Sesso speaks from experience. As Associate Director of the Division of Preventive Medicine at Brigham and Women’s Hospital and Associate Professor of Medicine at Harvard Medical School, he has participated in crucial nutritional studies such as the Physicians’ Health Study I and II and the Women’s Health Study. . Sesso shared his thoughts on clinical trial design on October 19 at the Council for Responsible Nutrition (CRN; Washington, DC) Science in Session conference.
âWhen we think of dietary supplements and the many large and small-scale clinical trials that have been done,â Sesso said, âthe notion of how gender plays a role in these results, I would actually say, was lost, Unfortunately. “
First, there’s the problem of restricting studies to a unisex population, Sesso said. While studies like the Physician Health Study and the Women’s Health Study, which are always well-designed and well-conducted clinical trials, focus on one gender, you lose the chance to study the same result in the other sex. Why is this important? “By limiting the results to mostly men, or only men, or only women, we lose the value of subgroup analysis,” Sesso explained.
Subgroup analysis is important because it gives us a window into how the intervention has unfolded in other populations. And it goes beyond genre, Sesso said. The same goes for differences in age, race and ethnicity, nutritional status, diet, body weight, and many other factors that help us contextualize the results in a rich way. By not diversifying and creating adequate subgroups in a study, âthe problem is that you end up with results that relate to one group but not the other, or [just] a certain age group, âhe said.
There are also the practical drawbacks of limiting a study – drawbacks involving wasted time and money that could attract the attention of policy makers. Suppose a study done in men shows promising results that the researchers say should also be explored in women. The researchers must launch an entirely new study, which they could have avoided by including both sexes from the start. Sesso said this happened in the case of the Doctors’ Health Study, which was conducted in men but was ultimately extended to women. It was “not the most efficient way to do it,” he said. âIt really should have been done at the same time to some extent. “
Then there is the problem of not digging deep enough into the data to show how the study results applied to different subgroups, for example how the interventions might have impacted differently on men and women. women. This is especially the case with meta-analyzes, Sesso said. “Where is the separation between men and women, by age, things like that?” These are things that just aren’t emphasized enough in these meta-analyzes. “
He lamented: “We focus on the big picture, not the details.” And these details are important. They are important to the way we understand and apply the results of studies. They are important in the way we make public health recommendations for the general population.
They are also important when comparing the results of different clinical trials in order to draw large-scale conclusions and, indeed, to advance the science of nutrition. If more studies were designed more similarly and to include more subgroups, their results could be more easily compared to each other. âWhether we’re looking at large-scale or small-scale trials, they need to complement each other in a much more direct and functional way,â Sesso said. âThere is just too much heterogeneityâ¦ between these trials over time, which made it difficult to develop a really strong recommendation. “
âIn the past, what we’ve done with our testing is we’ve been too restrictive, frankly. We focused on the intervention and the clinical outcomes and did not integrate the mechanisms simultaneously so that we could bring together a number of different trials, âhe concluded.
The good news is that we are learning to avoid these obstacles. As examples, Sesso cited VITAL (the VITAmin D and OmegA-3 trial) and COSMOS (the Cocoa Supplements and Multivitamin study). Here, the researchers deliberately adopted more specificity and more subgroups. âOur most recent trials that we did included both males and females so we could actually look directly at the change in effects by gender, age, and other important components,â he said. stated, “so that when you do a trial, you may actually be testing the same procedure on a larger number of people.”
Sesso hopes that we can âstart to reverse the scenario a bitâ and take these considerations into account when designing studies. âWe did it backwards,â he said. By designing studies more holistically, we can find promising results in some of the subjects, but not all, and then we can follow those who are likely to see benefits. But what if we don’t look at these topics in the first place? It is likely that we are missing a valuable piece of the puzzle.