Predictors of girls' self-initiated dieting and patterns of weight control behavior in childhood and adolescence

Open Access
Balantekin, Katherine N
Graduate Program:
Nutritional Sciences
Doctor of Philosophy
Document Type:
Date of Defense:
February 24, 2015
Committee Members:
  • Leann L Birch, Dissertation Advisor
  • Jennifer Savage Williams, Dissertation Advisor
  • Barbara Jean Rolls, Committee Member
  • Kathleen Loralee Keller, Committee Member
  • Lori Anne Francis, Committee Member
  • dieting; weight control; adolescence; parenting; media; social n
Forty percent of pre-adolescent girls report self-initiated dieting, with that number rising to three-quarters among adolescent girls. Despite the wide variety of both healthy (e.g. increase vegetable intake) and unhealthy (e.g. use of diuretics) weight control behaviors that can considered part of “dieting,” it is unclear what children and adolescents are doing when they report dieting. The purpose of the present research was to extend our knowledge of predictors and patterns of girls’ self-reported dieting behavior in late childhood and adolescence. Participants included 197 non-Hispanic white girls and their biological parents in central Pennsylvania followed from age 5y-15y. Data were collected across the 10-year period, with assessments every 2 years. The aim of study 1 was to examine the influence of parental encouragement to diet on the emergence of early (by 11y) and adolescent (between 11y and 15y) dieting. All measures of parental encouragement to diet (maternal, paternal, combined) predicted the emergence of early dieting, but not adolescent dieting. In addition, parental encouragement to diet predicted increases in girls’ BMI percentiles between the ages of 9 and 15, but only among girls who reported dieting. In Study 2, latent class analysis was used to identify patterns of weight control behaviors in the current sample at age 15. Four patterns were identified, Non-dieters, and three dieting groups: Lifestyle, Dieters, and Extreme Dieters. The Non-dieters did not endorse using any weight control behaviors, and the three dieting groups increased in the number and severity of their weight control behaviors, from the Lifestyle group who only reported increasing exercise and fruit and vegetable intake, to the Extreme Dieters who reported using all the behaviors, with over half reporting use of at least one unhealthy behavior. The three dieting groups were also different on a number of concurrent individual characteristics (e.g. BMI, depression), with differences between the groups observed as early as 5y. One of the intended purposes of dieting is to influence dietary intake. Thus, different patterns of weight control behaviors may be associated with different changes in intake. When dietary intake was examined, girls in the Extreme Dieters group had the lowest self-reported intake but consumed significantly more in the laboratory. The aim of Study 3 was to extend the findings of Study 2 to examine the association between patterns of weight control behaviors and family, friend, and media factors. Significant family factors included family functioning, priority of family meals, and maternal weight-teasing. Significant friend factors included weight-teasing and dieting. Sensitivity to media was significant. When all the factors were examined in a combined model, family functioning, friends’ dieting, and media sensitivity remained significant predictors of membership in a dieting group relative to the Non-dieters. Overall, the results of the present study indicate that for adolescent girls, dieting is an umbrella term for a wide variety of behaviors, and that types of dieters (e.g. Lifestyle, Dieters, and Extreme Dieters) are different on a number of individual characteristics and family, friend, and media factors. Taken together, these findings emphasize the need for a multidimensional prevention and interventions, addressing risk factors for dieting and use of unhealthy and extreme weight control behaviors at the individual, family, friend, and community (e.g. media) levels.