Ima be a little pragmatic here.
For an endeavor to succeed, there must be two things. First, the action must be easy and simple (all those words that describe a thing that becomes second nature at some point, i.e. does not require special training a priori) to do. For example, going hungry is not easy. Then, the result (the reaction) must be effective (all those words that describe a thing that is, for example, reliable, predictable, controllable, satisfactory, etc). For example, buying a lottery ticket is not effective.
I'll use my golf lessons to illustrate. I wrote golf lessons based on my personal experience with extensive study and practice. I designed the lessons to be simple to understand, easy to perform, logical in their progression, with a bit of higher level knowledge to complement though not strictly needed. They're written in segments, each with a simple specific goal in mind, all easy to achieve by players of all level. However, like the Atkins diet, it's a bit unconventional so that's pretty much the only obstacle to overcome, but once past it, it's smooth sailing because it's all easy to perform and all outcomes from all segments are predictable, i.e. do this to do that and so forth.
The point is that a "diet" doesn't need to be explained why it works, it merely needs to be explained how to do, so long as the instructions produce the intended result. For example, eat less move more is a set of instructions that does not produce the intended result in a predictable manner, so it's not effective. But, telling people to eat less carbs is effective because it actually produces the intended result reliably. Granted, it's possible an underlying condition prevents it from working as it should, but then again this same condition will also prevent any other dietary intervention from working as they should. Accordingly, any dietary intervention should come with this "advisory" that says "if it don't work, it's not the diet (because it should work as intended, right?), it's something else, find it, fix it, move on".
In that last paragraph you quoted, Teaser, it suggests the researchers are blaming the subjects for the failure of their own intervention, as if they could not do anything wrong. Instead, they should acknowledge that their intervention must be tailored to produce the intended results, regardless of the subjects' apparent starting conditions. They seem to believe that the intervention, if followed, will produce intended results, without regard to how these instructions are conveyed to the subjects. If you've ever read any assembly instructions for furniture for example, you'll know exactly what I mean by that.