Restrained eating in Lebanese adolescents: scale validation and correlates

Abstract

BackgroundRestrained eating disorder is prevalent worldwide across both ethnic and different cultural groups, and most importantly within the adolescent population. Additionally, comorbidities of restrained eating present a large burden on both physical and mental health of individuals. Moreover, literature is relatively scarce in Arab countries regarding eating disorders, let alone restrained eating, and among adolescent populations; hence, the aim of this study was to (1) validate the Dutch Restrained Eating Scale in a sample of Lebanese adolescents and (2) assess factors correlated with restrained eating (RE), while taking body dissatisfaction as a moderator between body mass index (BMI) and RE.MethodsThis cross-sectional study, conducted between May and June 2020 during the lockdown period imposed by the Lebanese government, included 555 adolescents aged between 15 and 18 years from all Lebanese governorates (mean age of 16.66 ± 1.00 years). The scales used were: Dutch Restrained Eating Scale, body dissatisfaction subscale of the Eating Disorder Inventory-Second version, Rosenberg Self-Esteem Scale, Beirut Distress Scale (for psychological distress), Hamilton Anxiety Rating Scale and Patient Health Questionnaire (for depression).ResultsThe Confirmatory factor analysis results were obtained as follows: χ2/df = 159.88/35= 4.57, CFI= 0.96, TLI= 0.95, RMSEA = 0.08 [0.068-0.093]. Female gender (B=0.19), higher BMI (B=0.49), higher physical activity index (B=0.17), following a diet to lose weight (B=0.26), starving oneself to lose weight (B=0.13), more body dissatisfaction (B=1.09), higher stress (B=0.18) were significantly associated with more restrained eating, whereas taking medications to lose weight (B=-0.10) was significantly associated with less restrained eating. The interaction BMI by body dissatisfaction was significantly associated with restrained eating; in the group with low BMI, high body dissatisfaction was significantly associated with more restrained eating. The factor analysis yielded a one-factor solution with Eigen values > 1 (variance explained = 59.65 %; αCronbach = 0.924). Female gender (B = 0.19), higher BMI (B = 0.49), higher physical activity index (B = 0.17), following a diet to lose weight (B = 0.26), starving oneself to lose weight (B = 0.13), more body dissatisfaction (B = 1.09), and higher stress (B = 0.18) were significantly associated with more RE, whereas taking medications to lose weight (B=-0.10) was significantly associated with less RE. The interaction body mass index (BMI) by body dissatisfaction was significantly associated with RE; in the group with low BMI, higher body dissatisfaction was significantly associated with more RE.ConclusionsOur study showed that the Dutch Restrained Eating scale is an adapted and validated tool to be used among Lebanese adolescents and revealed factors associated with restrained eating in this population. Since restrained eating has been associated with many clinically-diagnosed eating disorders, the results of this study might serve as a first step towards the development of prevention strategies targeted towards promoting a healthy lifestyle in Lebanese adolescents.