On our listserv this month there was some discussion of a recently published article entitled Financial Incentives for Extended Weight Loss: A Randomized, Controlled Trial. In thinking about my next blog for ASDAH there was simply no way I could pass this one up. Here was a chance to put to rest (again) two of the most pervasive oxymorons of traditional health promotion: incentives for behavior change and extended weight loss. No one could have published research that lends itself to this task better if they had set out to do so.
Before discussing the details of the study, a few comments about these two concepts are in order. The easier concept to grasp of the two is clearly extended weight loss. We are all quite aware of the lack of significant evidence that weight loss programs result in long-term weight loss for all but a small minority of participants. The authors of the article acknowledge this in their opening paragraph, saying:
Weight loss interventions have generally been unsuccessful in achieving sustained weight loss.
As for the use of incentives for behavior change, it turns out that the situation is surprisingly similar. Like weight loss programs, incentives often result in short term change, and like weight loss programs, incentives rarely result in long term success. This is true for a wide variety of applications and is certainly true in the area of health behavior change.
The use of incentives for behavior change has its roots in the mid 20th century work of B.F Skinner. Skinner believed that all human actions were merely “repertoires of behaviors” that could be fully explained by the environmental consequences that followed them. He saw humans as brainless machines responding to their environment claiming that:
There is no place in the scientific analysis of behavior for a mind or a self. (American Psychologist, 1990)
We answer the telephone after it rings because someone on the other end (hopefully not a telemarketer) answers. Theoretically, if we responded to the ringing phone enough times without someone answering, we would eventually stop doing so. While this makes intuitive sense, Skinner went way beyond this by concluding that all human behaviors were thusly determined. His take on love, for example, is that when two people meet:
[o]ne of them is nice to the other and predisposes the other to be nice to him, and that makes him even more likely to be nice. It goes back and forth, and it may reach the point at which they are very highly disposed to do nice things to the other and not to hurt. And I suppose that is what would be called being in love. (Walden II).
Pretty romantic, don’t you think? Anyway, to make a long story short, the behavioral approach to motivation (external rewards and punishments) is ubiquitous in our culture and rarely questioned in spite of the fact that decades of studies have convincingly demonstrated that it does not work for most people in most situations. Furthermore, the research strongly suggests a number of significant iatrogenic consequences, not the least of which is a reduction in internal motivation associated with its use. (see Punished By Rewards, Alfie Kohn) This conclusion holds true in a wide variety of applications and certainly for health behavior change – as multiple studies with weight loss, smoking cessation, seat belt use, etc. have demonstrated (see Robison, To Reward?…Or Not To Reward?: Questioning the Wisdom of Using External Reinforcement in Health Promotion Programs, and Kohn, Incentives and Health Promotion: What Do the Data Really Say?).
So, the research in this article incorporates not one, but two supposed health promoting approaches that result in short term change, rarely succeed in the long run and engender iatrogenic consequences – a 32-week, randomized, controlled trial of the use of financial incentives for extended weight loss.
Three groups of men between the ages of 30 and 70, with BMIs between 30-40 from a VA Hospital were assigned to one of three groups (n=22 each group). Subjects in the two experimental groups met with a dietitian and put up money (matched by the program) which they could then win or lose based on their self-report of whether or not they met their weight loss goals (1 pound per week) over the course of the next 24 weeks. One of the experimental groups was told that weeks 24 through 32 constituted a maintenance period, while the other was not. The control group met with a dietitian and then received no incentives, but was followed for the same period of time. The main measure was weight loss at 32 weeks. As a secondary measure, weight loss was again assessed 36 weeks after the initial 32 weeks.
Subjects in the two experimental groups lost an average of about 8 or 9 pounds, as compared to the control group’s 1 pound average loss over the course of the first 32 weeks. At the end of the “maintenance period” (additional 36 weeks), “participants regained weight post-intervention” and there was no difference between the incentive and control groups. The results are not overly surprising given the decades of research and scores of studies which have reached the same conclusions. Of course iatrogenic consequences were not assessed in either case.
So, what lessons can we take from this study and decades of similar findings? Weight loss interventions work in the short term for many and in the long term for very few. And financial incentives do about the same. Do we really need one more study to tell us this? Apparently we did and we still do. Here are the authors’ conclusions.
This is the first study to demonstrate that deposit contact incentives can successfully help keep off weight for 32 weeks.
Future research is needed to devise techniques that promote sustained weight loss over longer periods of time.
Amazing – no?
P.S. I actually helped develop and run a program like this back in the late 80s for employees at Michigan State University. More about that and why incentives are contraindicated for health behavior change and most other changes in my next blog.