@Article{info:doi/10.2196/jmir.3131,
author="Svensson, Madeleine
and Hult, Mari
and van der Mark, Marianne
and Grotta, Alessandra
and Jonasson, Josefine
and von Hausswolff-Juhlin, Yvonne
and R{\"o}ssner, Stephan
and Trolle Lagerros, Ylva",
title="The Change in Eating Behaviors in a Web-Based Weight Loss Program: A Longitudinal Analysis of Study Completers",
journal="J Med Internet Res",
year="2014",
month="Nov",
day="03",
volume="16",
number="11",
pages="e234",
keywords="behavior; counseling; diet; eating; method; questionnaires; Internet; weight loss; TFEQ",
abstract="Background: Eating behaviors are essential components in weight loss programs, but limited research has explored eating behaviors in Web-based weight loss programs. Objectives: The aim was to evaluate an interactive Web-based weight loss program on eating behaviors using the 18-item Three-Factor Eating Questionnaire Revised (TFEQ-R18) which measures uncontrolled eating, emotional eating, and cognitive restrained eating. Our Web-based weight loss program is comprised of information about healthy lifestyle choices, weekly chats with experts, social networking features, databases for recipe searches, and features allowing members to self-report and track their weight, physical activity, and dietary intake on the website. Methods: On registering for the weight loss program, 23,333 members agreed to take part in the research study. The participants were then asked to complete the TFEQ-R18 questionnaire at baseline and after 3 and 6 months of participation. All data collection was conducted online, with no face-to-face contact. To study changes in TFEQ-R18 eating behaviors we restricted our study to those members who completed all 3 TFEQ-R18 questionnaires. These participants were defined as ``completers'' and the remaining as ``noncompleters.'' The relationships between sex, change in eating behaviors, and total weight loss were studied using repeated measures ANOVA and Pearson correlation coefficient. Results: In total, 22,800 individuals participated (females: 19,065/22,800, 83.62{\%}; mean age 39.6, SD 11.4 years; BMI 29.0 kg/m2; males: 3735/22,800, 16.38{\%}; mean age 43.2, SD 11.7 years; BMI 30.8 kg/m2). Noncompleters (n=22,180) were younger and reported a lower score of uncontrolled eating and a higher score of cognitive restrained eating. Over time, completers (n=620) decreased their uncontrolled eating score (from 56.3 to 32.0; P<.001) and increased their cognitive restrained eating (from 50.6 to 62.9; P<.001). Males decreased their emotional eating (from 57.2 to 35.9; P<.001), but no significant change was found among females. The baseline cognitive restrained eating score was significantly and positively associated with weight loss for completers in both men (P=.02) and women (P=.002). Conclusions: To our knowledge, this is the largest TFEQ sample that has been documented. This Web-based weight loss intervention suggests that eating behaviors (cognitive restrained eating, uncontrolled eating, and emotional eating) measured by TFEQ-R18 were significantly changed during 6 months of participation. Our findings indicate differences in eating behaviors with respect to sex, but should be interpreted with caution because attrition was high. ",
issn="1438-8871",
doi="10.2196/jmir.3131",
url="http://www.jmir.org/2014/11/e234/",
url="https://doi.org/10.2196/jmir.3131",
url="http://www.ncbi.nlm.nih.gov/pubmed/25367316"
}