1 2 Research report 3 Influence of social factors on weight-related behaviors according to gender 4 in the French adult population q 5 Marie Pigeyre a,⇑ Q1 , Alain Duhamel b , Jean-Pierre Poulain c , Julien Rousseaux b , Pierre Barbe d , 6 Sandrine Jeanneau c , Laurence Tibère c , Monique Romon a 7 a Nutrition Unit, EA2694, UDSL, University of Lille Nord de France, Lille University Medical Center, CHU Lille, F-59000 Lille, France 8 b Department of Biostatistics, CERIM, EA2694, UDSL, University of Lille Nord de France, Lille University Medical Center, F-59000 Lille, France 9 c Research Center for Work, Organization and Power, CERTOP, UMR-CNRS 5044, Toulouse II – Le Mirail University, Toulouse, France 10 d Endocrinology Unit, Rangueil University Medical Center, Toulouse, France 11 12 14 article info 15 Article history: 16 Received 12 March 2011 17 Received in revised form 29 November 2011 18 Accepted 30 November 2011 19 Available online xxxx 20 Keywords: 21 Socioeconomic position 22 Perception of weight appropriateness 23 Eating behavior 24 Smoking status 25 TV viewing 26 Physical activity 27 28 abstract 29 Although the prevalence of obesity is higher in low socioeconomic position (SEP), the relationship 30 between SEP and body mass index (BMI) differs according to gender. The purpose of this study is to inves- 31 tigate the relationships between BMI and SEP according to gender and explore the weight-related behav- 32 iors. In a cross-sectional survey, 1646 French adults were weighed and answered a questionnaire about 33 eating behavior (DEBQ), SEP markers, ideal weight perception, physical activity and smoking. Our study 34 showed that BMI was inversely correlated with SEP score in women only, independently of other BMI- 35 associated factors (age, restrained eating, smoking status, TV viewing and physical activity). The SEP gra- 36 dient was the same in both genders for some weight-related behaviors, such as restrained eating, physical 37 activity and TV viewing, but differs for others, such as smoking and weight consciousness. There was an 38 interaction between the SEP score and the actual BMI on the ideal BMI in women only, thus the difference 39 in ideal body weight according to SEP is mainly due to difference in obese women. Our study concluded 40 that gender differences in the relationship between BMI and SEP could be mainly due to the subjects’ per- 41 ception of weight appropriateness and their weight-related behaviors. 42 Ó 2011 Published by Elsevier Ltd. 43 44 45 Introduction 46 The relationship between socioeconomic position (SEP) and 47 obesity is well established (McLaren, 2007). In Western countries, 48 the prevalence of obesity is inversely associated with SEP and this 49 relationship tends to be more pronounced in women than in men 50 (Wardle, Waller, & Jarvis, 2002). Although different aspects of life- 51 style such as diet, physical activity, sedentary and smoking, have 52 clear influences on weight, associations with SEP and gender merit 53 further investigation. Associations between total energy intake or 54 dietary fat intake and SEP are variable; most researchers have re- 55 ported a higher fruit and vegetable intake in higher SEP popula- 56 tions (Drewnowski, 2009). Smoking is usually more common in 57 low SEP populations (James, Nelson, Ralph, & Leather, 1997) and 58 physical activity is higher among male manual workers (Wardle 59 et al., 2002); these two factors would tend to promote a lower 60 BMI in lower SEP groups. However, leisure-time physical activity 61 is associated with higher SEP, especially in women (Ball, Mishra, 62 & Crawford, 2003). Moreover, it has been suggested that TV view- 63 ing is associated with obesity and SEP in women but not in men 64 (Crawford, Jeffery, & French, 1999). 65 The contribution of life style factors to social gradient of obesity 66 is thus different between genders. However it has been suggested 67 that these life style factors explain only 20% of social gradient 68 (Molarius, 2003). One contributory, gender-related factor in the 69 relationship with SEP is the differing attitude to weight gain and 70 the use of various practices to weight control (Wardle & Griffith, 71 2001). Dieting is the most common response to weight gain in 72 higher SEP (Ball et al., 2003). In modern societies, characterized 73 by an abundance of easily available food, cognitive restraint may 74 be a frequent, adaptive eating behavior for limiting weight gain 75 (de Lauzon-Guillain et al., 2006). 76 These differences in weight-related behaviors as a function of 77 SEP and gender may be related to the normative evaluation of 78 weight status. It has been suggested that differences in weight con- 0195-6663/$ - see front matter Ó 2011 Published by Elsevier Ltd. doi:10.1016/j.appet.2011.11.031 q Acknowledgments: M.P. conceptualized and wrote the manuscript, J.P.P. have conceptualized the SEP score, P.B. participated in the conception and design of the study and reviewed the paper, S.J. and L.T. were involved in data collection and data analysis and the SEP score design, A.D. and J.R. performed the statistical analyses and M.R. participated in conceptualizing the study and writing the manuscript. David Fraser’s help with editorial revision of the manuscript was greatly appreci- ated. The writing group takes sole responsibility for the content of this article. None of the authors have any personal or financial conflicts of interest with respect to the present work. ⇑ Corresponding author. E-mail address: marie.pigeyre-2@univ-lille2.fr (M. Pigeyre). Appetite xxx (2011) xxx–xxx Contents lists available at SciVerse ScienceDirect Appetite journal homepage: www.elsevier.com/locate/appet APPET 1388 No. of Pages 7, Model 5G 12 December 2011 Please cite this article in press as: Pigeyre, M., et al. Influence of social factors on weight-related behaviors according to gender in the French adult pop- ulation. Appetite (2011), doi:10.1016/j.appet.2011.11.031