Genetic predisposition to obesity, restrained eating and changes in body weight over adulthood


  • H. Konttinen
  • C. Llewellyn
  • K. Silventoinen
  • A. Joensuu
  • S. Männistö
  • V. Salomaa
  • P. Jousilahti
  • J. Kaprio
  • M. Perola
  • A. Haukkala


Background: There is no consensus on whether cognitive control of eating (i.e. restrained eating) is helpful, merely ineffective or actually harmful in weight management. We examined the interplay between genetic risk of obesity, restrained eating and changes in body weight and size. Methods: Participants were Finnish 25- to 74-year-olds who attended the DILGOM study at baseline (n=5024) in 2007 and follow-up (n=3735) in 2014. Questionnaires (e.g. Three-Factor Eating Questionnaire-R18; self-reported weight at age 20 years) and clinical measurements (e.g. height, weight, waist circumference) were completed during the study. We calculated 7-year change in body mass index (BMI) and waist circumference (WC), and annual weight change from age 20 to baseline. Genetic risk of obesity was assessed by calculating a polygenic risk score of 97 known BMI-related loci. Findings: Cross-lagged autoregressive models indicated that baseline restrained eating was unrelated to 7-year change in BMI and WC (β=0.00-0.01, P=0.485-0.767). Instead, higher baseline BMI and WC predicted greater 7-year increases in restrained eating (β=0.08, P<0.001). Polygenic risk score had a small positive correlation with restrained eating. Additionally, individuals with a higher genetic risk of obesity tended to gain more weight from age 20 to baseline and this effect was more pronounced in unrestrained than in restrained eaters (P=0.038 interaction). Discussion: Restrained eating appears to be an indicator of susceptibility to weight gain rather than a factor that causes weight gain in middle-aged adults. There was tentative evidence that restrained eating may be helpful in reducing genetic influences on long-term weight gain over adulthood.