What Happens to Dropouts?
Dropouts in obesity treatment studies and programs present a significant problem. Long-term outcomes are what matter. But the individuals know very early on if things are working for them. And many other reasons contribute to high attrition rates. One analysis found that more than a quarter of participants in a large sample of obesity treatment studies dropped out of the studies.
For a new analysis published by Frontiers in Nutrition, Kathryn Kaiser and colleagues from the University of Alabama at Birmingham (UAB) pooled individual data from ten randomized controlled trials of lifestyle modification in people with obesity.
They found that dropout rates are higher for blacks, Hispanics, older subjects, and subjects with higher BMIs. Coincidentally, these are subgroups for whom treatment might be of particular importance. African Americans and Hispanics bear a disproportionate burden from obesity. The benefits of successful treatment for people with high BMIs are substantially greater. And the consequences of untreated obesity are greater with increasing age.
Other research has suggested that both early results and a long history of prior experience with weight management efforts have much to do with the likelihood of discontinuing a weight management program early on. Interestingly, one recent study found that some patient characteristics — such as a standardized measure of disposition to anger — could predict early dropouts.
For the people in these programs, dropping out might be absolutely the right thing. There’s no need to waste time on something that’s not working. For people running programs and conducting the research, it’s clear that accounting for the forces that drive dropouts is essential to obtaining credible findings.
Better outcomes and more robust analyses will be the result.
Click here to read the study by Kaiser et al.
One Dropped Out, photograph © chriscom / flickr
Subscribe by email to follow the accumulating evidence and observations that shape our view of health, obesity, and policy.
