Behavioral counseling that is computer automated can enhance weight loss for individuals following an Internet-based weight loss program, according to a study by researchers at the University of North Carolina at Chapel Hill and the Miriam Hospital in Providence, R.I.

“As more consumers seek information on the web, it’s essential that we identify and improve on the components of Internet weight loss programs that make these programs successful,” said Dr. Deborah Tate, lead study author and assistant professor in the UNC School of Public Health.

More than 65 percent of Americans are overweight or obese. Earlier studies by the research team found that Internet weight loss programs are more effective when they include e-mail counseling from an expert. Automating such counseling has the potential to reach larger populations at a fraction of the cost of more intensive interventions, the researchers said.

“Compared to e-mail counseling provided by a human, a computer pre-programmed with messages based on specific criteria is a highly efficient treatment approach that can be more widely disseminated among the population in need,” said co-author Dr. Rena Wing, director of the Weight Management and Diabetes Research Center at The Miriam Hospital and Brown Medical School.

The study appears in the Aug. 14 issue of the Archives of Internal Medicine. Funding was provided by the Slim-Fast Nutrition Institute. The Institute provided login data from their Web site to the researchers, but had no role in the design or conduct of the study; in data collection, analysis, or interpretation; or in preparation or approval of the manuscript.

For the study, 192 overweight or obese adults were randomly assigned to one of three treatment groups: a group receiving no e-mail counseling, a group receiving computer-automated counseling, or a group receiving counseling via e-mail from a
human. Participants in all groups were encouraged to follow an Internet weight loss program; the automated and human groups also had access to an electronic diary and message board. The human group also received weekly e-mail feedback from a counselor, while the automated group received computer-automated feedback messages.

All three groups were evaluated at the start of the program, three months and six months. The primary measure of success was change in body weight, though physical activity, dietary intake and login frequency to the program’s Web site was also monitored.

“We found that both the human and computer-automated counseling groups lost significantly more weight than the group without counseling after three months,” says Tate.

While average weight loss was similar for both groups receiving e-counseling at three months, the researchers found that at six months, the human group experienced the greatest overall weight loss, averaging a drop of 13 pounds to 15 pounds.

The computer automated group averaged an overall weight loss of 8 pounds to 10 pounds, and 34 percent achieved a weight loss of more than 5 percent of their initial weight - considered a clinically meaningful weight loss. In addition, participants in the computer automated group achieved weight loss equivalent to human e-counseling for the first three months.

“Computer-automated guidance could be advantageous to individuals looking to lose weight through the web,” Tate said.

Wing adds that further research is needed to improve the computerized approach to promote continued adherence over a longer period of time, but said that from a public health perspective, “This is a promising area to explore.”

University of North Carolina at Chapel Hill