Background: Overweight and obesity affect approximately one-third of Canadian children and these children are at an increased risk of developing life-long obesity-related health consequences. Treatment efforts in the pediatric setting focus on behavioural modification – including modification of dietary intake. Eating behaviours encompass a modifiable aspect of dietary intake and could be a useful approach for weight reduction in the context of pediatric weight management (PWM) programs. Objectives: 1) To describe changes in family, child and parent eating behaviours over a 12-month period after enrolment in a PWM program. 2) To examine if changes in family eating behaviours over the first 6 months in a PWM program can predict child BMI z-score at 12 months, even when controlling for other factors related to changes in BMI z-score. Methods: Children ages 2-17 with a BMI >85th percentile were included in this analysis. Data was collected at baseline, 6 month and 12-month timepoints as part of the CANadian Pediatric Weight management Registry. BMI z-score was derived from height (cm) and weight (kg). Eating behaviours (eating related to hunger (ERH) and eating style (ES)) were measured in children, parents and families using the Family Eating and Activity Habits Questionnaire. Child and parental scores were summed to generate a family score. Decrease in scores over time indicated improvements in eating behaviours. Results: 807 children were included in the analysis (48.3% male, 51.7% female, mean age 12.08, mean baseline BMI z-score 3.55). Family ERH scores decreased significantly from baseline to 6 months by 1.63% (-0.16 points, SD=0.09, p<0.05) but by 12 months scores were the same as at baseline. Child ERH scores decreased significantly by 2.06% at 6 months (-0.14 points, SD=0.02, p<0.05), but at 12 months scores were the same as at baseline. Parent ERH scores did not change significantly. Family ES scores decreased from baseline to 6 months by 1.88% (-1.99 points, SD=0.21, p<0.05) and stayed approximately the same at 12 months. Child ES scores were similar, decreasing 2.86% (-1.68 points, SD=0.03, p<0.05) from baseline to 12 months. Parents ES score decreased significantly by 6 months, but after 12 months, scores were the same as at baseline. Change in either family eating behaviours from baseline to 6 months did not predict child BMI z-score at 12 months when adjusting for age, sex and BMI z-score at baseline. In subsequent exploratory analysis, increases in child eating behaviour scores at 6 months was a significant predictor of increases in child BMI z-score at 12 months (p=0.02) when adjusting for age, sex and BMI z-score at baseline. Conclusion: In the context of PWM programs, eating style and eating for hunger improved over the first 6 months after enrolment. Improvements in child eating behaviour after 6 months could predict improvements in child BMI z-score after 12 months. / Thesis / Master of Science (MSc)

Yes / Objective: To examine score validity and reliability of a Child version of the 21-item Three-Factor Eating Questionnaire (CTFEQ-R21) in a sample of Canadian children and adolescents and its relationship with body mass index (BMI) z-scores and food/taste preferences. Design: Cross-sectional study. Setting: School-based. Subjects: 158 children, 63 boys (11.5±1.6 years) and 95 girls (11.9±1.9 years). Results: The exploratory factor analysis revealed that the CTFEQ-R21 was best represented by four factors with item 17 removed (CFFEQ-R20) representing Cognitive Restraint (CR), Cognitive Uncontrolled Eating (UE 1), External Uncontrolled Eating (UE 2), and Emotional Eating (EE) and accounted for 41.2% of the total common variance, with good scale reliability. ANOVAs revealed that younger children reported higher UE 1 scores and CR scores compared to older children, and boys who reported high UE 1 scores had significantly higher BMI Z-scores. Children with high UE 1 scores reported a greater preference for high protein and fat foods, and high-fat savoury (HFSA) and high-fat sweet (HFSW) foods. Higher preference for high protein, fat, and carbohydrate foods, and HFSA, HFSW, and low-fat savoury foods was found in children with high UE 2 scores. Conclusions: This study suggests that the CFFEQ-R20 can be used to measure eating behaviour traits and associations with BMI z-scores and food/taste preferences in Canadian children and adolescents. Future research is needed to examine the validity of the questionnaire in larger samples and in other geographical locations, as well as the inclusion of extraneous variables such as parental eating or socio-economic status.

Résumé : Problématique: Les comportements alimentaires à la petite enfance pourraient influencer le gain de poids. Alors que les enfants d’âge préscolaire passent beaucoup de temps dans des milieux de garde, peu d’études ne rapportent les liens entre leurs comportements alimentaires dans ces milieux et leur composition corporelle. De plus, les recherches dans ce domaine sont basées sur des données subjectives des parents. Les objectifs étaient donc de développer un outil de mesure objectif des comportements alimentaires et de l’utiliser afin d’examiner le lien entre ceux-ci en milieu de garde et la composition corporelle des enfants d’âge préscolaire. Méthodes : Cette étude transversale a été menée auprès de 309 enfants de 3 à 5 ans dans 24 milieux de garde. Le comportement alimentaire a été mesuré en effectuant l’évaluation des restes au repas du dîner. Les données de ces évaluations ont été utilisées pour développer un score représentant la réticence alimentaire et un représentant l’affinité. Des corrélations intra-classes ont été utilisées afin de déterminer la fidélité des nouvelles mesures. Des corrélations de Spearman ont été utilisées afin de comparer les nouvelles mesures aux comportements alimentaires rapportés par les parents. Des régressions linéaires multivariées ont été utilisées pour examiner la relation entre les scores de comportement alimentaire mesurées objectivement et le tour de taille et l’indice de masse corporelle (IMC) ajusté à l’âge des enfants. Résultats : La nouvelle mesure de réticence alimentaire a démontré une excellente fidélité inter-juge (ICC= 0,970 à 0,998, p<0.0001), intra-juge (ICC=0,975 à 0,998, p<0.0001) et une bonne fidélité test-retest (ICC=0,723, p<0.0001). Elle corrélait également avec une mesure subjective précédemment validée (rho= 0,534, p<0.0001). La nouvelle mesure d’affinité alimentaire n’était toutefois pas valide (rho= -0,182, p=0,2). Le score de réticence alimentaire était positivement associé à l’IMC ajusté à l’âge des enfants (bêta ajusté= 1,41, IC à 95%=0,15, 2,67), mais non avec leur tour de taille (bêta ajusté=0,60, IC à 95%= -0,86, 2,06). Conclusion : La nouvelle mesure de réticence alimentaire démontre une excellente fidélité et validité chez les enfants d’âge préscolaire en milieu de garde. Plus il y avait de la réticence alimentaire, plus l’IMC ajusté à l’âge abaisse. Les milieux de gardes pourraient donc être des milieux propices pour l’implantation d’interventions pour réduire la réticence alimentaire, et contrer l’insuffisance pondérale chez les jeunes affectés. / Abstract : Background: Eating behaviours could be associated with weight gain during early childhood. Although a majority of preschoolers spend most of their active day-time hours in daycare centres, associations between their eating behaviours at daycare and their body composition have been limitedly studied. Further, research on eating behaviours of children mainly relies on parent-reported measures. The objective of this study was to develop an objective measure of eating behaviours and to use it to assess the relationship between these behaviours and body mass index and waist circumference among preschoolers. Methods: This cross-sectional study was conducted among 309 children aged 3 to 5 in 24 daycare centres. Eating behaviours were measured through weighted digital plate waste analysis. Data from this evaluation was used to create a food reluctance score and a food affinity score. Intraclass correlations (ICC) were used to determine the reliability of the new measure. Spearman correlations were used to compare the new measures with parental report of eating behaviours. Multivariate linear regressions were used to examine the relationship between objectively measured food behaviours and children’s waist circumference and age-adjusted body mass index (BMI). Results: The new measure of food reluctance demonstrated excellent inter-rater reliability (ICC= 0.970 to 0.998, p<0.0001), intra-rater reliability (ICC=0.975 to 0.998, p<0.0001), and good test-retest reliability (ICC=0.723, p<0.0001). It also provided evidence of concurrent validity through correlation with a validated subjective measure (rho= 0.534, p<0.0001). The new measure of food affinity was however not valid (rho= -0.182, p=0.2). The food reluctance score was positively associated with children’s age-adjusted BMI (adjusted bêta; 95% CI= 1.41, 0.15, 2.67), but not with their waist circumference (0.60; -0.86, 2.06). Conclusion: The objective measure of food reluctance demonstrated evidence of reliability and validity. Greater demonstration of food reluctance at the daycare center was associated with a lower BMI. This suggests that daycare centers could represent promising settings for integrating strategies to counter food reluctance and counter inadequate weight gain among affected preschoolers.

Problematic weight regulation as found in obesity and Anorexia Nervosa (AN) are chronic conditions which require long-term management. In order to develop long-term strategies to manage these conditions, a clearer understanding of the factors that can contribute to the development and also recovery from these conditions are a necessity. Although obesity and AN are at the opposite end of the bodyweight spectrum, some shared psychological processes may drive these states. One factor that has been suggested to contribute to problematic weight regulation is psychosocial stress whilst positive affect systems and affect regulation processes are important for regulating stress-related experiences. Gilbert (2005) describes an affect regulation system which consists of two positive affect systems known as social rank and attachment. Whilst the latter affect system refers to the attachment bond that develops between an infant and its caregiver (which extends to adult relationships), the former is used to form relationships that allow us to compete for limited resources and maintain our status in the social environment. Affect regulation processes in the current research are self-criticism and self-reassurance. Whilst self-critical thoughts and feelings can be triggered by perceptions of being low rank, the idea that people can be self-reassuring or being able to self-reassure at times of difficulty is nested in the positive infant-caregiver attachment bond and a consequence of internalizing parental soothing (Gilbert, 2006). Hence, as stress is suggested to be an important factor in problematic weight regulation and affect systems and processes are central to the regulation of emotional responses to stress-related experiences, the current series of studies examined these factors in relation to problematic weight regulation. The current research consisted of four studies designed to examine the role of stress and affect regulation in relation to weight change, weight regain following weight loss and recovery versus symptom maintenance in AN in women. A longitudinal study (Study One) was conducted to examine the change trajectories of stress, eating pathology and bodyweight, how these changes influence each other and the role of affect regulation systems and processes on these changes in a community based sample (N = 1157). Study Two examined the role of stress and affect regulation as predictors of weight regain in those who have lost weight (N = 42) and Study Three used a measure of life events and difficulties to investigate the role of stressful life changes and affect systems on recovery and relapse following AN (N = 30). Finally, in Study four, an expressive writing task which has been demonstrated to have a positive impact on stress-related health outcomes was used to explore the role of stress, affect systems and processes on problematic weight regulation and eating at times of stress (N = 57). The findings of the research studies demonstrated that there is a concurrent link between stress and the regulation of bodyweight and eating in a community-based sample of women. However, the proposed relationship between stress, bodyweight and eating behaviours was not confirmed when examined longitudinally in a community-based sample, over a 7-month period in women who have lost weight or when examined retrospectively as contributing to symptom maintenance in women with AN. However, the main finding of the current series of studies suggested that affect systems and affect regulation processes do have important implications for regulating stress-related experiences, bodyweight and eating behaviours. Perceived low social status, greater insecurity of attachment, more self-critical and less self-reassuring thoughts and feelings were related to increases in stress levels, higher bodyweight and higher levels of dysfunctional eating patterns. In addition, whilst expressive writing did not reduce stress, influence bodyweight or improve affect regulation at times of difficulty, writing about positive experiences had a positive impact on reducing dietary restraint behaviours during a stressful period. In conclusion, these findings suggest that it may not be stress per se that contributes to unhealthy changes in bodyweight and eating behaviours but how we use our affect systems and processes to manage our emotions at times of difficulty. Consequently, these findings have important implications for practice as weight loss programmes, Eating Disorder prevention programmes and stress management interventions should address the issues of perceived low social status, self-criticism and attachment insecurities.

La problématique de l’excès de poids représente un enjeu de santé publique. En effet, l’obésité, et, en premier lieu, le surpoids, concerneraient (à eux deux) 45% de la population française. Au-delà de facteurs génétiques, hormonaux et socio-économiques, les habitudes de vie des personnes (i.e., alimentation, dépenses énergétiques, etc.) sont souvent mentionnées pour expliquer l’origine d’un telle pathologie. Or, si ces premiers facteurs trouvent un appui scientifique, les habitudes de vie sont, quant à elle, peu documentées et relèveraient davantage du mythe. Néanmoins, les travaux menés en psychologie sociale ont démontré que, parfois, le mythe peut devenir réalité… En effet, les recherches réalisées sur le phénomène de menace du stéréotype ont permis de montrer que les stéréotypes négatifs dont certains groupes sociaux sont la cible peuvent nuire à leurs comportements dans un sens conforme à ces stéréotypes. Plus précisément, par peur de confirmer un stéréotype négatif à l’égard de leur groupe social, des membres peuvent être paradoxalement conduit à émettre des comportements consistants avec ce stéréotype, dans un contexte où ce dernier est saillant et applicable. Le présent travail visait, par conséquent, à mettre en exergue l’impact de la stigmatisation relative au poids sur les conduites alimentaires des personnes en surcharge pondérale vivant en France. Pour ce faire, 633 personnes en surpoids/obèses et 1278 normopondérés ont participé aux six expérimentations (et cinq pré-tests) présentées dans cette thèse. Dans un premier temps, nous nous sommes assurés de l’existence d’une stigmatisation relative au poids en France, et nous nous sommes attachés à étudier son contenu. Dans un second temps, nous nous sommes intéressés au ressenti des personnes obèses face à cette stigmatisation, en terme de vécu, fréquence, stratégies de coping mises en œuvre pour y faire face, bien-être psychologique et comportements de santé (e.g., alimentation, activité physique, etc.). In fine, à travers le phénomène de menace du stéréotype, nous avons examiné les effets de cette stigmatisation relative au poids sur les intentions/comportements alimentaires d’une population normopondérée et en excès de poids. Les résultats de ces études ont, tout d’abord, permis de montrer qu’il existait bien une stigmatisation relative au poids en France. Cette stigmatisation attribuait notamment de mauvaises habitudes alimentaires et un manque d’activités physiques aux individus en excès de poids. Ensuite, nous avons démontré que le vécu de stigmatisation relative au poids de personnes obèses et leurs stratégies de coping adoptées pour y faire face corrélaient avec un faible bien-être psychologique, un fort niveau d’impulsivité et des comportements alimentaires inadaptés. Pour finir, nous avons mis en évidence que, dans certaines situations, cette stigmatisation relative au poids, via le phénomène de menace du stéréotype, conduisait des personnes en surpoids à consommer davantage d’aliments caloriques. Ces résultats sont discutés au regard de la littérature portant sur la menace du stéréotype, notamment auprès des personnes en surpoids. Des pistes de recherches et d’interventions, pour diminuer les effets délétères de la stigmatisation relative au poids et ceux du phénomène de menace du stéréotype qui en résulte, sont également proposées. / Excessive weight is a public health issue. In fact, obesity and overweight affect around 45% of the French population.Beyond genetics, hormonal and social economics factors, lifestyle (i.e., feeding, physical exercise, etc.) is often cited as a possible cause of the disease. However, if these initial factors find scientific supports, way of life, however, is hardly substantiated and would rather reflect a myth. Nevertheless, social psychology has demonstrated that the myth can sometimes become reality… Indeed, research on the stereotype threat phenomenon shows that members of social groups can be harmed by negative stereotypes. More precisely, even if they fear it, individuals can adopt behaviours consistent with these activated and applicable stereotypes. Therefore, the present work aimed to highlight the impact of weight stigma on the French overweight people’s eating behaviours. To this end, 633 overweight/obese people and 1278 normal-weight people took part in the six experiments (and five pre-tests) presented in this thesis. Firstly, we ensured that the French weight stigma exists and we explored its content. Secondly, we took an interest in obese people’s feelings and how they were dealing with the weight stigma in terms of real-life experiences, frequency, coping strategies, psychological well-being and healthy behaviours (e.g., feeding, physical activity, etc.). Thirdly, through the stereotype threat phenomenon, we examined the effects of this weight stigma on the eating intentions/behaviours for normal-weight and overweight people. First, the results showed the existence of weight stigma in France. Poor eating habits and physical inactivity are attributed mostly to overweight people. Then, we demonstrated that obese people’s weight stigma experiences and their coping strategies correlated with a low psychological well-being, a high level of impulsiveness and inappropriate eating behaviours. Finally, according to stereotype threat, we provided experimental evidence that weight stigma can lead overweight people to consume more high-calorie food in certain circumstances. These results will be discussed with regard to the stereotype threat literature, and particularly studies related to overweight people. Suggestions for further research and possible interventions to reduce or eliminate the harmful effects of stereotype threat resulting of weigh stigma will also be proposed

The objectives of this thesis were to examine the relationship between the built environment and obesity-related behaviours in Canadian youth, and to address measurement issues relevant for studying these relationships. The thesis is comprised of four manuscripts. The first manuscript determined the accuracy of the food retailer locations surrounding 34 schools in Ontario as provided by two geographic information system (GIS) databases. The GIS locations of 25% of the food retailers were located within 15 m of their actual location, 50% were within 25 m, and 75% were within 50 m. These values did not differ by GIS database source. The second manuscript examined the association between the presence of food retailers surrounding schools and students’ lunchtime eating behaviours. It also determined whether a circular or road network buffer provided a better geographic boundary of the food retail environment surrounding schools. It was carried out in 6,971 9th and 10th grades students from 158 schools. The presence of food retailers near schools was strongly associated with students regularly eating their lunch at a snack-bar, fast-food restaurant or café. The road network buffers provided the better measure of the food retail environment. The third manuscript determined the optimal buffer size of the food retail environment surrounding schools. Six road network buffers were created around 158 schools. These ranged in size from 500 m to 5000 m. Results indicated that the 1000 m buffer was the optimal size. The fourth manuscript explored associations between urban sprawl and active transportation, moderate to vigorous physical activity (MVPA) and overweight/obesity in 7,017 youth aged 12 to 19 who resided in one of Canada's 33 Census Metropolitan Areas. Increasing levels of sprawl were modestly associated with increased active transportation (only among 12 to 15 year olds) and MVPA. No associations were found for overweight/obesity. Taken together, the results from this thesis provide a better understanding of the role of the built environment on obesity-related behaviours in Canadian youth. Important measurement issues in this topic area were also addressed. / Thesis (Ph.D, Community Health & Epidemiology) -- Queen's University, 2012-01-30 19:36:47.354

Obesity, a common health condition in adolescence leading to severe medical complications, is assumed to be influenced by temperament factors. This paper investigates associations between reactive and regulative temperament, problematic eating behaviours, and excess weight. Several self-report instruments were completed by 130 adolescents (mean age 14.13 ± 0.61 years), including 27 overweight and obese individuals (20.8%). Bootstrap analysis revealed a mediating effect of restrained eating on the relation between reactive temperament and body mass index percentile, which differed according to gender: Restrained eating, which predicted weight gain, was more present in girls having a higher sensitivity to reward and in boys showing a higher sensitivity to punishment. No effect of regulative temperament was found. These results have important implications for weight management programs, as they suggest that reducing restrained eating by working on temperament may help to control weight.

Overweight, a common health condition in adolescence, has been linked with difficulties in emotional processing. This study investigates associations between emotional processing, conceptualised through the model of Emotional Openness (EO), problematic eating behaviours, including Eating in the Absence of Hunger and disinhibited eating, and overweight in adolescents. Several self-report instruments were completed by 160 youngsters (mean age: 14.36 ± 0.61 years) from the community, including 39 overweight and obese adolescents (24.5%). In girls, bootstrap analyses supported a mediating effect of restrained eating on the relation between three EO dimensions and body mass index percentile, in particular the communication of emotions, the cognitive-conceptual representation of emotions, and the perception of bodily indicator of emotions. No mediating effect was found in boys. These results have important implications for psychological weight management interventions, as they underline the relevance of work on emotional processing in order to reduce problematic eating behaviours.

Overweight, a common health condition in adolescence, has been linked with difficulties in emotional processing. This study investigates associations between emotional processing, conceptualised through the model of Emotional Openness (EO), problematic eating behaviours, including Eating in the Absence of Hunger and disinhibited eating, and overweight in adolescents. Several self-report instruments were completed by 160 youngsters (mean age: 14.36 ± 0.61 years) from the community, including 39 overweight and obese adolescents (24.5%). In girls, bootstrap analyses supported a mediating effect of restrained eating on the relation between three EO dimensions and body mass index percentile, in particular the communication of emotions, the cognitive-conceptual representation of emotions, and the perception of bodily indicator of emotions. No mediating effect was found in boys. These results have important implications for psychological weight management interventions, as they underline the relevance of work on emotional processing in order to reduce problematic eating behaviours.

A considerable percentage of the UK population are overweight (BMI≥25kg/m2) or obese (BMI≥30kg/m2). However, despite living in the same culture and exposed to a similar “obesogenic” environment, some individuals gain weight while others do not (French et al., 1995). This variability in weight control has been suggested to be associated with individual differences in eating behaviours (French et al., 2012). Certain factors, such as motivation (hedonic hunger and hunger status) as well as cognition (impulsivity and memory) may have an impact on eating behaviours and their relationship with weight control. Thus, the objective of this thesis was to explore individual differences in eating behaviours and investigate their relationship with motivation, cognition and weight control. The first experiment (Chapter 2) investigated the relationship between eating behaviours, motivation (hedonic hunger) and food consumption during an ad-libitum buffet. This study suggests that restrained eating behaviour was associated with higher overall energy intake, greater energy intake from unhealthy foods and greater energy intake from both high and low energy dense foods. However, no interactions between restraint and disinhibition or hedonic hunger was seen. Following this, the second experiment (Chapter 3) examined whether eating behaviours, such as disinhibition, restraint and hunger, change during a weight loss and weight maintenance period and whether they could predict changes in weight during these periods. Indeed, the results suggest that lower baseline restraint could predict greater weight loss during a low-energy liquid diet and interventions which increase restraint and decrease disinhibition may be beneficial for longer term weight maintenance. The third experiment (Chapter 4) was designed to investigate whether motivation and cognition influences eating behaviours. The results suggest that hedonic hunger, restraint and impulsivity may lead to higher levels of disinhibited eating behaviour. This study was also able to replicate the findings of previous literature suggesting that episodic memory is negatively associated with BMI (Cheke et al., 2016). Finally, following on from the previous study results, the fourth experiment (Chapter 5) included a more diverse sample of participants including dieters. The results provide evidence that individuals on a diet have poorer episodic memory ability than those currently not on a diet. This study also extended previous results suggesting that hedonic hunger (but also episodic memory and hunger) are important factors in disinhibited eating. Hedonic hunger was also shown to be important in levels of hunger.