The role of implicit attitudes towards food and physical activity in the treatment of youth obesity
Introduction
The increasing prevalence of childhood and adolescence overweight and obesity in Western society is a serious problem (Lissau et al., 2004). It is generally acknowledged that the causal factor next to genetic predispositions is an interplay of an unhealthy diet and a sedentary lifestyle (Morgan, Tanofsky-Kraff, Wilfley, & Yanovski, 2002).
For youngsters with severe obesity, family-based cognitive behavioural interventions that focus on increasing physical activity and decreasing unhealthy diets are recommended (Braet, Tanghe, De Bode, Franckx, & Van Winckel, 2003). In general, these programs have promising results. Nevertheless, a lot of youngsters regain a considerable amount of their weight after treatment (e.g., Braet et al., 2003, Braet et al., 2004). Therefore, longitudinal research is necessary to investigate whether the underlying dynamics of the food and physical activity behaviour are affected by the treatment, and to what extent their change is related to weight loss and weight regain.
Several characteristics have been found to be important predictors of weight loss in obese youngsters following treatment, under which baseline degree of overweight and the child's age (Braet, 2006). Further, an amount of theories have dealt with possible explaining psychological determinants, such as the learning history of the child and a heightened responsiveness to food cues (for a review, see Braet, 2005). From a healthy psychology perspective, attitudes towards food and exercise have been put forward as crucial factors in the development and maintenance of obesity (e.g., De Bourdeaudhuij et al., 2005, Dennison and Shepherd, 1995).
In most studies, attitudes are asked directly, for instance using self-reports. Recently, however, it has been argued that their indirect assessment, by which cognitions are inferred from behaviour performance other than self-report, should have a better explanatory and predictive value because of two reasons (Greenwald et al., 2002). First, indirect measures should be more resistant for demand characteristics that often affect self-reports (see Schwarz, 1999). Second, indirect measures are supposed to catch cognitive automatic associations that guide spontaneous behaviour, and that can not easily be tapped by self-reports. The outcome of the indirect assessment of these associations is often called implicit cognitions. Implicit attitudes then are the measure of automatic associations between an attitude object and a certain valence (Greenwald et al., 2002).
In the last decades, several indirect response time based paradigms have been constructed to assess implicit attitudes (for a review, see Fazio & Olson, 2003), amongst which the affective priming task (Fazio, Sanbonmatsu, Powell, & Kardes, 1986), the Implicit Association Task (IAT; Greenwald, McGhee, & Schwartz, 1998) and the Extrinsic Affective Simon Task (EAST; De Houwer, 2003). The core principle of these tasks is that implicit attitudes are inferred from the reaction time (and percentage of errors) performance on a computer sorting task. More specifically, it is assumed that response latencies will be shorter (and less errors will be made) when participants have to use the same response bottom to sort stimuli that carry the same valence (congruent stimuli; e.g., names of insects and negative nouns vs. names of flowers and positive nouns), than when the same response bottom has to be pressed for stimuli with an opposite valence (incongruent stimuli; e.g., names of insects and positive nouns vs. names of flowers and negative nouns). Implicit attitudes are calculated then by comparing the performance on the congruent stimuli with the performance on the incongruent stimuli.
Indirect measures have been applied in several domains of clinical and health psychology (for a review, see Fazio & Olson, 2003). As described elsewhere (Craeynest et al., 2005), we used the EAST in a cross-sectional study investigating implicit food and exercise attitudes in children and adolescents with obesity at the beginning of an intensive treatment. Our results revealed a borderline significant effect, indicating that youngsters with obesity implicitly preferred both unhealthy and healthy food, whereas normal-weight controls evaluated them as neutral. No effects were found for physical activity.
Longitudinal research investigating whether implicit food and physical activity cognitions in youth obesity change over a treatment period, and to what extent they are related to treatment success (i.e., weight loss or weight gain), is scarce. We only know of the study by Barton, Walker, Lambert, Gately, and Hill (2004) in which a sentence-completion questionnaire was used to investigate whether automatic exercise and food cognitions change over the course of a residential weight loss camp for adolescents with obesity. They found a reduction in negative and an increase in positive thoughts about exercise, but not in those about eating. Further, cognitive change was largely accounted for by the reduction in weight.
This manuscript describes a longitudinal study in which implicit and self-reported food and exercise attitudes are assessed in youngsters with obesity during a six month inpatient treatment and at a one year follow up. Self-reported attitudes are assessed with a questionnaire; implicit attitudes are assessed with an EAST. Three main questions are addressed: First, do implicit and self-reported food and exercise attitudes change during an inpatient treatment for obesity? Second, to what extent does a change in these attitudes explain the variance in weight change during treatment? And third, to what extent does a change in these attitudes explain the variance in weight loss or regain at follow up?
Section snippets
Participants
Participants were thirty-eight obese children and adolescents at the beginning of their hospitalization in 2002–2003 in a medical-paediatric centre (MPC) that is specialized in the treatment of youth obesity. Fourteen of them dropped out during the therapy because of having reached their therapy goal (n = 4) or because of suspension due to behavioural problems breaking the rules of the centre such as smoking (n = 10). Five could not be contacted at follow up. These youngsters did not differ from
Mean implicit and self-reported attitudes at baseline
A 3 (word category: sedentary activities vs. moderate intense PA vs. high intense PA) × 2 (extrinsic response valence: positive vs. negative) Repeated Measures ANOVA on the baseline reaction times of the physical activity task showed no significant effects, all Fs < 2.85. Also the same 3 × 2 Repeated Measures ANOVA conducted on the baseline percentage of errors revealed no effects, all Fs < 2.19. However, a 2 (word category: unhealthy vs. healthy food) × 2 (extrinsic response valence: positive vs.
Discussion
This manuscript presented a longitudinal study, investigating whether implicit and self-reported food and exercise attitudes changed during a six month inpatient treatment for youth obesity and at a one year follow up. Further, it was examined whether this change explained the variance in overweight change during treatment and at follow up.
A first finding was that the youngsters lost weight during the treatment, that seemed not to be regained at follow up. This lack of weight regain is in
Acknowledgements
This research was supported by grant B/03814/01 from the Ghent University. We are very grateful to the MPC Zeepreventorium and the children for their cooperation to this study. In addition, we wish to thank Jan De Houwer for his ideas on conceptualizing the experiment. Furthermore, many thanks go to all colleagues and students who assisted in data collection, especially Els Persijn, Leen Van Vlierberghe, Katrien Verhoeven and Mieke De Roo.
References (27)
- et al.
Explicit and implicit attitudes toward food and physical activity in childhood obesity
Behaviour Research and Therapy
(2005) - et al.
Differences in psychosocial variables by stage of change for fruit and vegetables in older adults
Journal of the American Dietetic Association
(2004) - et al.
Cognitive change in obese adolescent losing weight
Obesity Research
(2004) The malleability of automatic stereotypes and prejudice
Personality and Social Psychology Review
(2002)- et al.
Environmental and societal factors affect food choice and physical activity: Rationale, influences, and leverage points
Nutrition Reviews
(2001) Psychological profile to become and to stay obese
International Journal of Obesity
(2005)Patient characteristics as predictors of weight loss after an obesity treatment for children
Obesity Research
(2006)- et al.
Inpatient treatment of obese children: A multicomponent programme without stringent calorie restriction
European Journal of Pediatrics
(2003) - et al.
Inpatient treatment for children with obesity: Weight loss, psychological well-being, and eating behaviour
Journal of Pediatric Psychology
(2004) - et al.
Establishing a standard definition for child overweight and obesity worldwide: International survey
British Medical Journal
(2000)