Here's a table from the study I cited previously. Below is the author's explanation for that one data point that fell out of range.
"It was noticeable that those patients who maintained weight loss (case 11, table II, [that 7kg/15lbs in the last column] ) were those who bought new clothes that fitted at their reduced weight and warned them when their weight increased."In fact, the waist cord and abdominal studies were conducted to test the theory that clothes worn snugly around the midsection can assist in weight management.
So, even though I'm kind of titillated by the idea of a functional waist ornament, it was always a clothes issue, or as Critser (author of Fat Land) calls it, the Theory of the Belt:
"There is what might be called the theory of the belt, which holds that people will watch and maintain their weight better if they are warned that they are gaining weight by clothing that makes them slightly uncomfortable."