Showing 2 ideas for tag "probative"

Goal 3: Advance Translational Research

Move toward more probative research

As some of us described in a recent publication, in the fields of nutrition and obesity, and perhaps in other fields as well, there is often a great deal of research which uses up resources, investigator time, journal pages, and attention span for questions that do not advance the field. We call for scientists, reviewers, and funding decision makers to collectively ask much more rigorously, "How will this proposed study... more »

Is this idea a Compelling Question (CQ) or Critical Challenge (CC)? Critical Challenge (CC)

Details on the impact of addressing this CQ or CC

A quintessential example of this state of affairs is the large number of studies involving breakfast consumption and obesity. On the order of 100 studies have been done testing the association of breakfast eating vs. breakfast skipping with obesity. This seemed to lead to the belief and widely stated public health message that breakfast should be consumed to prevent obesity or promote weight loss. While the first several such epidemiologic studies were reasonable to conduct, what was needed thereafter were RCTs to test for causal effect. Instead the scientific community provided itself with a large body of observational studies to the point where an association was established far beyond any reasonable doubt (P≈10-42). Only recently have a handful of investigators conducted the RCTs needed to advance knowledge further, i.e. the "probative" studies. These have failed to support the hypothesis that breakfast consumption vs skipping leads to better weight control.

Feasibility and challenges of addressing this CQ or CC

Similarly, we seem to be faced with a large stream of studies of varying quality testing the hypothesis that modest increases in physical activity in school settings for children will lead to important differences in weight outcomes. Yet, considerable research has already demonstrated to a reasonable degree of certainty that school-based programs with modest increases of physical activity do not have major effects on children's BMIs. This does not mean that physical activity is not important for other outcomes, or that there is not some way of inducing physical activity that would lead to major changes in BMI, but repetitively trying one minor variation on school-based programs after another is not the best use of our resources. These are just examples.

Name of idea submitter and other team members who worked on this idea David B. Allison, Ph.D.; Kevin Fontaine, Ph.D.; Kathryn A. Kaiser, Ph.D.; Andrew W. Brown, Ph.D.; Edward C. Archer, Ph.D.

Voting

1 net vote
2 up votes
1 down votes
Active

Goal 3: Advance Translational Research

Increased receptivity to probative programmatic trials

We believe there should be greater openness to large, simple trials that answer clear questions of interest (e.g. does giving children more fruits and vegetables while changing nothing else lead to weight loss?; does eating breakfast regularly lead to weight loss?; etc.). The conduct of such trials may sometimes be expensive but can sometimes be only modestly costly if they are kept simple. However, it is difficult to... more »

Is this idea a Compelling Question (CQ) or Critical Challenge (CC)? Critical Challenge (CC)

Details on the impact of addressing this CQ or CC

While we do not mean for a moment that we should drift away from mechanistic science, we believe that there should also be openness to addressing some questions that are simple and perhaps even slightly dull, but can be unequivocally answered with a trial. In this way, beliefs can be converted to facts.

Feasibility and challenges of addressing this CQ or CC

It is difficult to seek funding for such trials because reviewers want to see more testing of mechanisms, more physiologic outcomes, more testing of hypothesized mediating variables, and more exciting scientific hypothesis tests.

Name of idea submitter and other team members who worked on this idea David B. Allison, Ph.D.; Kevin Fontaine, Ph.D.; Kathryn A. Kaiser, Ph.D.; Andrew W. Brown, Ph.D.; Edward C. Archer, Ph.D.

Voting

0 net votes
1 up votes
1 down votes
Active