Wednesday, March 11, 2015

The DIET Score - What Every Weight Loss Study Needs

By Jon Rawlinson (The Long Road Ahead) [CC BY 2.0)], via Wikimedia Commons
Re-upping/tweaking this post from 2012 as I'm not aware of anyone having taken me up on the idea, and rightly or wrongly, I think it would add value to weight loss literature were someone to take it on.
Most diet studies stink. They stink for a pile of reasons not the least of which are their generally short or medium term durations. Given the real trial is keeping the weight off, these short and medium term studies' value added are small in that short term losses in no way, shape, or form guarantee their long term maintenance. Given that people seem capable of putting up with the most god-awful and inane diets around, being able to put up with a study protocol for a few months seems a given, so too does the fact that if someone loses an extreme amount of weight in a short period of time, they're not likely to gain it all back before the conclusion of even a medium term diet study.

Now I realize that research dollars don't materialize out of thin air, and that the likelihood is that good idea or not, we're still not going to see an automatic shift from short term, to long term data reporting. To that end, I'd like to offer a potential work around.

Before I get to it, some non-evidence based experiential theory,
If you don't like the life you're living while you're losing weight you're almost certainly going to gain it back
The ability of a body to gain weight isn't something that's "curable", and therefore if weight's the outcome being measured, treatment must be continued forever if its results are to be maintained. If treatment's too much of a misery, if the diet's too strict, too sparse, too confusing, too anything, well I'm betting its results aren't going to be too lasting.

Even more simply put,
Weight lost through suffering will likely be regained
So if my purported truisms are in fact true, why is it the that I have never seen any consideration of dietary quality of life in a weight loss study's methodology and evaluation?

The way I see it, if there were a measure with which dietary satisfaction could be evaluated, even if it's a short term study, you might get a sense as the liveability of the diet and the lastability of its affects. If everyone in the study felt the intervention was a misery you might gather there'd be a damn good chance the intervention won't be long lasting. On the other hand, if the majority reported an intervention as being enjoyable, you might think more folks will continue with it long term.

So I'd like to propose the Diet Index Enjoyability Total (DIET) score whereby using a series of simple Likert scales (descriptive scales from 1-10), researchers could set out to evaluate a particular weight loss approach's DIET score where high scores might represent lives that could actually be enjoyed and low scores the usual under-eating, over-exercising, highly restrictive, quality of life degrading, misery that are most modern day diets. Ideally these scores would be gathered during a diet's early days (the honeymoon period), but also at intervals up to two years out to ensure that data are collected long after the scale stops whispering sweet nothings in a person's ears.

What sorts of "enjoyability" items could be scored?
  • Hunger
  • Cravings
  • Feelings of fullness/satisfaction
  • Need to cook special meals for other family members
  • Ability to still eat out with friends and family
  • Energy levels and feelings of general well-being
  • Complexity of dietary requirements
  • Dietary flexibility vs. monotony
  • Rigidity of dietary requirements (ie forbidden foods/food groups and impact on quality of life)
  • Expense/cost of dietary requirements (ie expensive foods, supplements, etc.)
10 items yielding a score of 100. Now scores wouldn't necessarily correlate with degree of loss, but I'd be willing to wager, the higher the score, the greater the long term utility of the diet which might well matter more than the amount of initial loss. This of course could be studied in turn, and if validated, included as a descriptor for all studies' diet protocols.

Any of my obesity researcher readers willing to take this on?  I'll happily help!