I was recently involved in a randomised controlled trial targeting weight loss maintenance, illustrated here with a ridge walk, which is i think quite apposite (Image copyright steynard on flickr). The intervention was a type of counselling (motivational interviewing) which also included planning and self-monitoring which are known to affect weight loss. It was called the WILMA study (WeIght Loss Maintenance in Adults trial, which isn’t actually the most contrived acronym i’ve encountered). The full report has just recently been made available and it is a beast >400 pages which is basically comparable in length to a PhD thesis. It was a lot of work.
We originally wanted to recruit 950 participants but recruitment proved challenging for various reasons and the trial was closed after 170 participants were recruited. The study became a feasibility study rather than an effectiveness trial. Nevertheless we were still interested in estimating what size effect could be achieved for future studies. The primary outcome was Body Mass Index(BMI). This was a three arm trial, with two groups receiving intensive (6 sessions) or less intensive (2 sessions) motivational interviewing being compared to a control group (who received none). We found that the intensive group had BMIs roughly one point [95% CI:-2.16, 0.23 ](or 2.8 kilogrammes [95% CI:-6.09, 0.45]) lower than control on average. Moreover, as a feasibility study it was resoundingly successful with just under 84% of participants being successfully followed-up at one year, and over 85% of participants adhering to the intervention.
Interestingly, a pre-specified sensitivity analysis that accounted for adherence to the intervention (i.e. attendance at motivational interviewing sessions) (a Complier-Average Causal Effect (or CACE) analysis since you ask) found a statistically significant difference in the secondary outcome of weight (difference between groups 3.69 kgs 95% CI:[-7.08, -0.31]).
The key now is to try to use all of the learning from this study, in order to design another study that might be able to demonstrate definitively whether or not this intervention is useful in combatting the growing obesity problem.