Obesity Research & Clinical Practice (2008) 2, 125—131
Our perception of weight: Socioeconomic and sociocultural explanations Natasha J. Howard a,∗, Graeme J. Hugo b, Anne W. Taylor c, David H. Wilson d a
Nutrition Obesity Lifestyle and Environment (NOBLE) Project, Discipline of Geographical and Environmental Studies, The University of Adelaide, Adelaide, SA, Australia b Discipline of Geographical and Environmental Studies, The University of Adelaide, Adelaide, SA, Australia c Population Research and Outcome Studies Unit, SA Department of Health, P.O. Box 287, Rundle Mall, Adelaide, SA, Australia d Discipline of Medicine, The University of Adelaide, Adelaide, SA, Australia Received 18 December 2007 ; received in revised form 12 March 2008; accepted 12 March 2008
KEYWORDS Perceptions; Obesity; Weight status; Social disadvantage; Sociocultural
Summary Objective: To compare self-reported perception of weight with biomedically measured body mass index in different socioeconomic and cultural groups. Method: Of the original North West Adelaide Health (Cohort) Study (n = 4060) 68.5% (n = 2780) underwent a computer assisted telephone interview (CATI) answering additional questions related to their social and health status. The participants were asked ‘‘In terms of your weight, do you consider yourself to be. . . too thin, a little thin, normal weight, a little overweight or very overweight’’. The self-perception of weight was compared to biomedically measured BMI (body mass index). Binary logistic regression was used to compare those participants who were obese (BMI ≥ 30) with the self-perceived weight status of ‘a little overweight’. The outcome measures included the Socioeconomic Indexes for Areas Index of Relative Socioeconomic Disadvantage (SEIFA IRSD), country of birth and household income. Results: Of those that were underestimating their obese weight status, 41.5% were male and 32.2% female. The highest misclassification was for those who considered their weight to be ‘a little overweight’, with 59.6% biomedically measured with a BMI of over 30. The odds of being biomedically measured obese (BMI ≥ 30) were compared to those who considered themselves to be ‘a little overweight’. Those that misreported their weight status and
Abbreviations: CATI, computer assisted telephone interview; BMI, body mass index; NWAHS, North West Adelaide Health (Cohort) Study; SEIFA IRSD, Socioeconomic Indexes for Areas Index of Relative Socioeconomic Disadvantage; WHO, World Health Organisation. ∗ Corresponding author. Tel.: +61 8 8303 4815; fax: +61 8 8303 3498. E-mail address: [email protected] (N.J. Howard). 1871-403X/$ — see front matter. Crown Copyright © 2008 Published by Elsevier Ltd on behalf of Asian Oceanian Association for the Study of Obesity. All rights reserved.
doi:10.1016/j.orcp.2008.03.003
126
N.J. Howard et al. were biomedically obese, were more likely to be living in the lowest quintile of disadvantage, have a household income of less then $20,000 or be born in Eastern or Western Europe. Conclusion: There are psychosocial, sociocultural and social environmental influences related to the perception of weight status. Future research will need to understand the processes whereby people are not aware they have a weight problem. Crown Copyright © 2008 Published by Elsevier Ltd on behalf of Asian Oceanian Association for the Study of Obesity. All rights reserved.
Introduction The prevalence of obesity has steadily increased over time within both the developed and developing world [1], with obesity being described as most prevalent among those who are socioeconomically and locationally disadvantaged [2]. The causes, prevention and outcomes associated with obesity are complex and incompletely understood, and there is recognition from leading obesity experts of the need for more interdisciplinary research into the condition [3,4]. One such area needing further investigation is that of the psychosocial influences on, and self-perceptions of, body weight and health. To date, much of the research on self-perception of weight has focused on parents’ perceptions of their children’s weight. Recent findings in the United States report that nearly two-thirds of mothers did not recognise that their children were overweight [5] and another study in Australia found a high-proportion of parents did not express concern of their children’s weight [6]. Studies on weight perception or appropriateness in the United States have shown that there is a misclassification of weight status for adults by medical standards [7]. A majority of research focuses on gender and age influences, although there have been some studies investigating the ethnic disparities associated with weight status in adults [8]. Research into the awareness of risks among rural Australians found that there is a significant difference between those that considered themselves overweight and the number that were classified overweight according to body mass index (BMI) [9]. However, limited studies have been conducted on perception of weight within a general, randomly selected, representative population. Behaviour change research has investigated the challenges faced by health practitioners and researchers. In the transtheoretical model stage of precontemplation, there is no intention by the individual to change behaviour in the foreseeable future, and individuals are unaware or underaware of their problems [10]. There is a need for the issue
of weight gain or excess weight to be recognised if the individual is to be able to proceed on a pathway of achieving healthy weight. There has been research undertaken which explores the lay perspectives on health and wellbeing and discusses some of the different ways in which people perceive the health care system and health issues [11,12]. In gaining an understanding of weight appropriateness an individual will also require a perception of risk, which is the subjective judgement that people make about the characteristics and severity of the risk of a behaviour or activity. The concept of risk is different for public health practitioners and those people that experience the risk. A study of cervical cancer risk explored the embodied risks—–risks identified as characteristics of their body. The study found that the embodied risk is not experienced by the individual until after they are actually diagnosed and this labelling then confronts them with a range of uncertainties they had not previously experienced [13]. The individual will often reconstruct the threat of the perceived risk by denying its personal relevance. These are all important features to consider in addressing healthy weight and helping those who misclassify their weight status to understand the risks associated with excess weight. This paper is a preliminary exploration into the self-perception of weight status, a psychosocial aspect of obesity that is explored through three main themes—–individual and area level socioeconomic disadvantage and sociocultural influences. The aim of the paper is to firstly compare self-perception of weight with the World Health Organisation (WHO) definition for BMI categories. Secondly, to compare and discuss socioeconomic and sociocultural themes related to self-perception of weight using a population-based study.
Methods The North West Adelaide Health Study (NWAHS) is a biomedical cohort study of a representative population sample of adults living in the north west region
Socioeconomic and sociocultural explanations for perception of weight
127
Figure 1 Flowchart of North West Adelaide Health Study timeline 2000—2008.
of Adelaide [14]. The NWAH Study is spatially referenced data (at the individual point level) for 99% of the original cohort offering a unique opportunity to research sociodemographic and biomedical data within the north west metropolitan area of Adelaide. Further information on the methodology and study population of the North West Adelaide Health Study has been previously published [14]. The biomedical findings have been compared to other metropolitan regions or country areas in South Australia and limited biases were reported [15]. The NWAHS data were weighted to the Australian Bureau of Statistics 1999 Estimated Residential Population by the north west region, age group, sex and probability of selection in the household to provide estimates for the population. Of the total initial eligible sample (n = 4060), 90.1% (n = 3566) provided information for the second stage of the study (Stage 2, 2004—2006) and 81.0% (n = 3206) attended the clinic. Fig. 1 is a flowchart of the NWAHS cohort data collection 2000—2008. The participants’ self-perceived weight status was collected during a follow up of the NWAHS cohort
during August and September 2007 (n = 2382), a response rate of 74.3%. The participants answered the following survey question: ‘‘In terms of your weight, do you consider yourself to be. . .too thin, a little thin, normal weight, a little overweight or very overweight?’’ The objective weight status was based on measured BMI that was calculated using height in centimetres (measured to the nearest 0.5 cm using a stadiometer) and weight in kilograms (measured to the nearest 0.1 kg in light clothing and without shoes using a standard digital scales). The analyses covered three focus areas, individual socioeconomic status (household income), area level disadvantage index (SEIFA IRSD), and sociocultural influences (country of birth). Area level socioeconomic status was measured using the Index of Relative Social Disadvantage a component of the Socioeconomic Indexes for Areas, (SEIFA IRSD) as compiled by the Australian Bureau of Statistics (ABS) at the Collector District Level (around 200 or 300 households) [16]. Questions on the household income were asked in a self-reported paper based questionnaire sent out to the participants
N.J. Howard et al.
0.0 0.5 (n = 4) 3.2 (n = 29) 34.3 (n = 236) 0.0 18.3 (n = 138) 62.1 (n = 561) 59.6 (n = 410)
100.0 100.0 100.0 100.0
(n = 37) (n = 753) (n = 904) (n = 688)
before each clinic appointment. Each participant was asked about their country of birth during the original recruitment for the study at Time 1. The null hypothesis to be tested is that there is no relationship between three measures of socioeconomic status and self-perceived weight status. Self-perceived weight status was compared with the biomedically measured WHO BMI classifications by crosstabulation and Chi-square testing. The overestimation, agreement and underestimation of self-perceived weight status was analysed by gender. Binary logistic regression was used to compare those participants with a BMI ≥ 30 with the self-perceived weight status of ‘a little overweight’ (n = 1109). The outcome measures included the SEIFA IRSD, country of birth and household income.
Results
WHO BMI (World Health Organisation Body Mass Index).
Underweight (BMI < 18.50) Normal (BMI 18.50—24.99) Overweight (BMI 25.00—29.99) Obese (BMI ≥ 30)
(n = 26) (n = 112) (n = 12) (n = 2)
(n = 11) (n = 449) (n = 302) (n = 40) 29.7 66.3 33.4 5.8 70.3 14.9 1.3 0.3
Very overweight (%) A little overweight (%) Normal weight (%) Too thin/a little thin (%)
Self-perception of weight WHO BMI classifications
Table 1
Comparison of self-perception of body weight and biomedically measured WHO BMI classifications
Total (%)
128
The prevalence of measured obesity within the North West Adelaide Health Study at Time 1 (2000—2003) was 27.0%. Table 1 is a comparison of self-perception of body weight and biomedically measured WHO BMI classifications. Of those that were biomedically measured as obese (BMI ≥ 30), 59.6% (n = 410) perceived their weight to be ‘a little overweight’ and 5.8% (n = 40) perceived they were of ‘normal’ weight. Of those people that were overweight, 33.4% (n = 302) actually thought that they were of ‘normal’ weight status. In addition, 29.7% (n = 37) of people who were underweight thought they were of ‘normal’ weight, and of those that were biomedically measured with a BMI between 18.00 and 24.99 (or normal), 18.3% thought that they were ‘a little overweight’ and 0.5% thought that they were ‘very overweight’.
Table 2 Underestimation, agreement and overestimation of self-perceived weight status by gender Self-perception of weight status agreement status with BMI n
%
Males Underestimated Agreement Overestimated
493 637 58
41.5 53.6 4.9
Females Underestimated Agreement Overestimated
384 684 125
32.2 57.3 10.5
Socioeconomic and sociocultural explanations for perception of weight Table 3 Association with self-perceived ‘a little overweight’ and obesity (BMI ≥ 30) ‘A little overweight’
Obesity (BMI ≥ 30) OR (95% CI)
SEIFA (IRSD) Highest quintile High quintile Middle quintile Low quintile Lowest quintile Country of birth Australia UK/Ireland Eastern or Western Europe Asia and other Household income Greater then $60,000 $40.001 to $60,000 $20,001 to $40,000 Up to $20,000
P
1.00 0.89 0.77 1.11 1.67
(0.46—1.73) (0.51—1.16) (0.76—1.61) (1.14—2.46)
0.739 0.205 0.583 0.008
1.00 1.17 (0.83—1.64) 1.72 (1.07—2.75)
0.369 0.024
0.68 (0.27—1.75)
0.428
1.00 1.32 (0.95—1.85)
0.099
1.35 (0.96—1.89)
0.085
1.82 (1.27—2.61)
0.001
SEIFA IRSD (Socioeconomic Indexes for Areas Index of Relative Socioeconomic Disadvantage).
Table 2 explores by gender those people that underestimated, overestimated or had agreement with their self-perceived weight status and body mass index classification. Underestimation of selfperceived weight status was found to be higher for males (41.5%) compared to females (32.2%). There were also found to be 10.5% of females overestimating their weight status, that is having the self-perception of their weight status being heavier when compared with their biomedical classifications of BMI. Table 3 describes obesity by those factors associated with self-perceived weight status ‘a little overweight’. Those who were biomedically obese but self-perceived themselves ‘a little overweight’ were more likely to be from the lowest quintile of SEIFA (OR 1.67), born in Eastern or Western Europe (OR = 1.72) and have an annual household income of less than $20,000 (OR = 1.82).
Discussion Previous research in South Australia has shown increasing rates of overweight and obesity in
129
adults and children [17—19]. In this particular population there was a high percentage of people in this study who were misclassifying their weight status. Around two-thirds (65.7%) of those that were biomedically obese were misclassifying their weight status by not reporting their current weight to be ‘very overweight’. The obesity epidemic confronts us with emerging issues across the social gradient. The relationship of obesity with socioeconomic status is complex with both those that are highly educated and of a higher income experiencing increasing prevalence as well as a lower prevalence of the condition [20]. Those within the lowest quintile of disadvantage and low household income levels were found to be more likely to interpret their weight status as ‘a little overweight’ when they really were biomedically obese. The social and physical environmental influences and their relationship with an individual’s health status is yet to be fully determined, although research has suggested that individuals living in disadvantaged areas are at greater risk of obesity regardless of their individual level of disadvantage [21,22]. This research has shown that it is those that are locationally and socioeconomically disadvantaged who are more likely to underreport their weight. This may mean an individual’s temporal awareness of the ‘risk’ of chronic disease further into their future, and the association with social environments in which these individuals exist, is yet to be fully determined [23]. People born in Eastern or Western Europe were more likely to misinterpret their weight status to be a ‘little overweight’ when they were actually biomedically obese when compared to those that were Australian born. Within this population there have been limited studies exploring the perception of weight amongst different cultural groups. In the United States there have been a number of studies exploring the racial discrepancies between black, hispanic and white populations [24,25]. There are sociocultural factors that drive the standards of desirable body weight within cultures, which in turn drive the lifestyles that people lead. As the population mean of body mass index shifts further along an obesity continuum there is a need to determine whether these social and cultural norms and the perceptions of body image or weight are changing simultaneously. Of those people that were found to misreport their obesity status the distribution was varied along the BMI continuum, with 14.3% (n = 40) of those participants having a BMI > 35 or classified with severe obesity (BMI ≥ 35).1 Those that are mis1
Howard, NJ. Further analyses of data, 2007.
130 classifying their weight status are actually biomedically measured as being at high risk of a chronic condition and are therefore not just slightly misclassifying [26]. In addition, there were a high percentage of people that were misreporting their weight in the other direction, that is over reporting their weight status. These findings draw attention to issues such as body image, self-esteem and increasing eating disorders among our population concurrently with the rising obesity prevalence [27]. More investigations are needed to look at the complexities in the distribution of weight, for example waist to hip ratio, and other definitions related to the perception of weight status. It was the aim of this research to use a commonly defined measurement for obesity, the BMI. The analysis is undertaken with caution as BMI is an arbitrary measure for defining obesity in our population and an individual may not necessarily report incorrectly if other measurements of body fat distribution are taken into account. Numerous studies have reported that BMI is not the most appropriate measure in determining risk of chronic disease, rather other body fat and distribution measures provide more reliable estimates [28,29]. These discrepancies are apparent in this research with 0.3% self-reporting their obesity status as ‘too thin or a little thin’ when actually they have a BMI greater then 30. Further investigations highlighted that those misreporting in this case were males with a normal waist to hip ratio.2 This research draws upon Giddens’ Structuration Theory which places an individual within a broader context and accounts for the multilevel nature of simultaneous individual and group level interaction [30]. There is a two way process between individuals and the social environments in which they live, work and play. Giddens writes that ‘‘It is important to grasp how history is made through the active involvements and struggles of human beings, and yet at the same time both forms those human beings and produces outcomes which they neither intend nor forsee [31].’’ More research is needed to explore these processes of the individual and social environmental interaction in relation to healthy weight. There are a broad range of other sociocultural and psychosocial factors that could be explored in relation to the self-perception of weight status. These include the influence of other sociocultural themes such as lifecourse indicators such as mother’s and father’s country of birth, and influence of an individual’s life trajectory and social
2
Howard, NJ. Further analyses of data, 2007.
N.J. Howard et al. context which shape the contemporary social environment, both of which contribute to the cultural landscapes of consumption and behaviour. The research presented in this paper presents challenges for policy and intervention. The transposing of this research into healthy weight messages at the population level appears to be a difficult task if people do not perceive their weight to be a problem. An intervention is not likely to be effective if it is based on the assumption that people recognize their weight to be a problem when in actual fact they do not consider this to be the case. More recognition of the perception of risk and lay perspectives of obesity are needed to understand the complexity of obesity within our population. The strength in the NWAHS lies in the representative nature, the large random sample, and the relatively high response rate. In addition, the NWAHS has strength in the standardised clinical procedures with stringent measurement standards and regular calibration of scales and height measures to provide reliable BMI data. Self-reported height and weight has commonly been used to determine BMI at a population level. There are recognised problems associated with self-reported height and weight, in that height is generally over-reported and weight under-reported [32]. The self-report data have previously been shown to underestimate the extent of obesity within this population [33]. Future research would need to consider the different measures of obesity and cut off points to look at biomedically assessed weight and perceived body image. As well as investigating how to target interventions towards healthy weight when people are not aware they have a problem. The present paper has been able to make only some preliminary observations on one main theme area of the socio-spatial context, that is the perceptions of weight status. There is a need to gain an understanding of the processes whereby people are not aware that they have a problem when it comes to their weight. Future research of the cohort will explore the multilevel nature of these relationships and how the spatial landscape has the ability to influence individual level outcomes such as social reproduction of lifestyles, behaviour and social disadvantage.
Acknowledgements The authors would like to express their appreciation to the Population Research and Outcome Studies Unit, South Australian Department of Health for providing access to data. We would especially like to acknowledge the work of the North West Ade-
Socioeconomic and sociocultural explanations for perception of weight laide Health Study Team and the participants that have given up their valuable time. Natasha Howard is funded under a scholarship with the Nutrition Obesity Lifestyle and Environment (NOBLE) project which is an Australian Research Council (ARC) Linkage Project (LP0455737) ‘‘Obesity, Health, Social Disadvantage and Environment’’.
References [1] World Health Organisation. Obesity and overweight. Geneva: World Health Organisation; 2005. [2] Ball K, Crawford D. Socioeconomic status and weight change in adults: a review. Soc Sci Med 2005;60(9):1987—2010. [3] Swinburn B, Egger G. The runaway weight gain train: too many accelerators, not enough brakes. BMJ 2004;329 (7468):736—9. [4] EurActive Network. Scientists call for more research into obesity. EurActive Network: Brief News 1, 2006. [5] Hackie M, Bowles CL. Maternal perception of their overweight children. Public Health Nurs 2007;24(6):538—46. [6] Wake M, Salmon L, Waters E, Wright M, Hesketh K. Parent reported health status of overweight and obese Australian primary school children: a cross-sectorial population survey. Int J Obes Relat Metab Disord 2002;26(5):717—24. [7] Chang VW, Christakis NA. Self-perception of weight appropriateness in the United States. Am J Prev Med 2003;24(4):332—9. [8] Olvera N, Suminski R, Power TG. Intergenerational perceptions of body image in hispanics: role of BMI, gender, and acculturation. Obes Res 2005;13(11):1970—9. [9] Coulson FR, Ypinazar VA, Margolis SA. Awareness of risks of overweight among rural Australians. Rural Remote Health 2006;6:514 [online]. [10] Prochaska JO, Norcross JC. Stages of change. Psychotherapy 2001;38(4):443—8. [11] Macintyre S, McKay L, Ellaway A. Lay concepts of the relative importance of different influences on health; are there major socio-demographic variations? Health Educ Res 2006;21(5):731—9. [12] Popay J, Thomas C, Williams G, Bennett S, Gatrell A, Bostock L. A proper place to live: health inequalities, agency and the normative dimensions of space. Soc Sci Med 2003;57(1):55—69. [13] Kavanagh AM, Broom DH. Embodied risk: my body, myself? Soc Sci Med 1998;46(3):437—44. [14] Grant JF, Chittleborough CR, Taylor AW, Dal Grande E, Wilson DH, Phillips PJ, et al. The North West Adelaide Health Study: detailed methods and baseline segmentation of a cohort along a chronic disease continuum. Epidemiol Perspect Innov 2005;3(4). [15] Taylor A, Dal Grande E, Gill T, Chittleborough CR, Wilson DH, Adams RJ, et al. Do people with risky behaviours participate in biomedical cohort studies? BMC Public Health 2006;6(1):11.
131
[16] Australian Bureau of Statistics. Socio-economic indexes for areas. In: Trewin D, editor. Information paper: census of population and housing. Canberra: Australian Bureau of Statistics (ABS); 2001. p. 1—29. [17] Gill T, Appleton S, Dal Grande E, Taylor A, Carter P, Wilson D. A weighty issue: the status of overweight and obesity in South Australian adults from 1991 to 2001. Adelaide, Department of Human Services; 2004. p. 1—71. [18] Dal Grande E, Gill T, Taylor AW, Chittleborough CR. Obesity in South Australia adults—–prevalence, projections and generational assessment over 13 years. Aust N Z J Public Health 2005;29(4):343—8. [19] Hugo G, Franzon J. Child obesity in South Australia: some initial findings. Food Culture Soc Int J Multidisciplinary Res 2006;9:299—316. [20] McLaren L. Socioeconomic status and obesity. Epidemiol Rev 2007;29(1):29—48. [21] Berkman LF, Kawachi I. Social Epidemiology. New York: Oxford University Press; 2000. [22] King T, Kavanagh AM, Jolley D, Turrell G, Crawford D. Weight and place: a multilevel cross-sectional survey of area-level social disadvantage and overweight/obesity in Australia. Int J Obes Relat Metab Disord 2005;30(2):281—7. [23] Leccardi C. About future and planning. future matters: futures known, created and minded. In: New biographical constructions in the ‘‘Risk Society’’? Cardiff University; 2006 [September 4—6]. [24] Paeratakul S, White MA, Williamson DA, Ryan DH, Bray GA. Sex, race/ethnicity, socioeconomic status, and BMI in relation to self-perception of overweight. Obesity Res 2002;10(5):345—50. [25] Neff LJ, Sargent RG, McKeown RE, Jackson KL, Valois RF. Black—White differences in body size perceptions and weight management practices among adolescent females. J Adolesc Health 1997;20(6):459—65. [26] Haffner SM. Relationship of metabolic risk factors and development of cardiovascular disease and diabetes. Obesity 2006;14(suppl. 3):121S—7S. [27] Mission Australia. National Survey of Young Australians 2007 Key and Emerging Issues. Mission Australia. December 2007. [28] Dalton M, Cameron AJ, Zimmet PZ, Shaw JE, Jolley D, Dunstan DW, et al. Waist circumference, waist—hip ratio and body mass index and their correlation with cardiovascular disease risk factors in Australian adults. J Intern Med 2003;254:555—63. [29] Sangi H, Mueller WH. Which measure of body fat distribution is best for epidemiologic research among adolescents? Am J Epidemiol 1991;133(9):870—83. [30] Gatrell AC. Geographies of Health: An Introduction. United Kingdom: Blackwell Publishers; 2002. [31] Giddens A. Sociology: A Brief but Critical Introduction. London & Basingstoke: MacMillan Press Ltd.; 1982. p. 166. [32] Australian Bureau of Statistics. How Australians Measure Up. Canberra: ABS; 1999. [33] Taylor AW, Dal Grande E, Gill T, Chittleborough CR, Wilson DH, Adams RJ, et al. How valid are self-reported height and weight? A comparison between CATI self-report and clinic measurements using a large cohort study. Aust N Z J Public Health 2006;30(3):238—46.
Available online at www.sciencedirect.com
Our perception of weight: Socioeconomic and sociocultural explanations Natasha J. Howard a,∗, Graeme J. Hugo b, Anne W. Taylor c, David H. Wilson d a
Nutrition Obesity Lifestyle and Environment (NOBLE) Project, Discipline of Geographical and Environmental Studies, The University of Adelaide, Adelaide, SA, Australia b Discipline of Geographical and Environmental Studies, The University of Adelaide, Adelaide, SA, Australia c Population Research and Outcome Studies Unit, SA Department of Health, P.O. Box 287, Rundle Mall, Adelaide, SA, Australia d Discipline of Medicine, The University of Adelaide, Adelaide, SA, Australia Received 18 December 2007 ; received in revised form 12 March 2008; accepted 12 March 2008
KEYWORDS Perceptions; Obesity; Weight status; Social disadvantage; Sociocultural
Summary Objective: To compare self-reported perception of weight with biomedically measured body mass index in different socioeconomic and cultural groups. Method: Of the original North West Adelaide Health (Cohort) Study (n = 4060) 68.5% (n = 2780) underwent a computer assisted telephone interview (CATI) answering additional questions related to their social and health status. The participants were asked ‘‘In terms of your weight, do you consider yourself to be. . . too thin, a little thin, normal weight, a little overweight or very overweight’’. The self-perception of weight was compared to biomedically measured BMI (body mass index). Binary logistic regression was used to compare those participants who were obese (BMI ≥ 30) with the self-perceived weight status of ‘a little overweight’. The outcome measures included the Socioeconomic Indexes for Areas Index of Relative Socioeconomic Disadvantage (SEIFA IRSD), country of birth and household income. Results: Of those that were underestimating their obese weight status, 41.5% were male and 32.2% female. The highest misclassification was for those who considered their weight to be ‘a little overweight’, with 59.6% biomedically measured with a BMI of over 30. The odds of being biomedically measured obese (BMI ≥ 30) were compared to those who considered themselves to be ‘a little overweight’. Those that misreported their weight status and
Abbreviations: CATI, computer assisted telephone interview; BMI, body mass index; NWAHS, North West Adelaide Health (Cohort) Study; SEIFA IRSD, Socioeconomic Indexes for Areas Index of Relative Socioeconomic Disadvantage; WHO, World Health Organisation. ∗ Corresponding author. Tel.: +61 8 8303 4815; fax: +61 8 8303 3498. E-mail address: [email protected] (N.J. Howard). 1871-403X/$ — see front matter. Crown Copyright © 2008 Published by Elsevier Ltd on behalf of Asian Oceanian Association for the Study of Obesity. All rights reserved.
doi:10.1016/j.orcp.2008.03.003
126
N.J. Howard et al. were biomedically obese, were more likely to be living in the lowest quintile of disadvantage, have a household income of less then $20,000 or be born in Eastern or Western Europe. Conclusion: There are psychosocial, sociocultural and social environmental influences related to the perception of weight status. Future research will need to understand the processes whereby people are not aware they have a weight problem. Crown Copyright © 2008 Published by Elsevier Ltd on behalf of Asian Oceanian Association for the Study of Obesity. All rights reserved.
Introduction The prevalence of obesity has steadily increased over time within both the developed and developing world [1], with obesity being described as most prevalent among those who are socioeconomically and locationally disadvantaged [2]. The causes, prevention and outcomes associated with obesity are complex and incompletely understood, and there is recognition from leading obesity experts of the need for more interdisciplinary research into the condition [3,4]. One such area needing further investigation is that of the psychosocial influences on, and self-perceptions of, body weight and health. To date, much of the research on self-perception of weight has focused on parents’ perceptions of their children’s weight. Recent findings in the United States report that nearly two-thirds of mothers did not recognise that their children were overweight [5] and another study in Australia found a high-proportion of parents did not express concern of their children’s weight [6]. Studies on weight perception or appropriateness in the United States have shown that there is a misclassification of weight status for adults by medical standards [7]. A majority of research focuses on gender and age influences, although there have been some studies investigating the ethnic disparities associated with weight status in adults [8]. Research into the awareness of risks among rural Australians found that there is a significant difference between those that considered themselves overweight and the number that were classified overweight according to body mass index (BMI) [9]. However, limited studies have been conducted on perception of weight within a general, randomly selected, representative population. Behaviour change research has investigated the challenges faced by health practitioners and researchers. In the transtheoretical model stage of precontemplation, there is no intention by the individual to change behaviour in the foreseeable future, and individuals are unaware or underaware of their problems [10]. There is a need for the issue
of weight gain or excess weight to be recognised if the individual is to be able to proceed on a pathway of achieving healthy weight. There has been research undertaken which explores the lay perspectives on health and wellbeing and discusses some of the different ways in which people perceive the health care system and health issues [11,12]. In gaining an understanding of weight appropriateness an individual will also require a perception of risk, which is the subjective judgement that people make about the characteristics and severity of the risk of a behaviour or activity. The concept of risk is different for public health practitioners and those people that experience the risk. A study of cervical cancer risk explored the embodied risks—–risks identified as characteristics of their body. The study found that the embodied risk is not experienced by the individual until after they are actually diagnosed and this labelling then confronts them with a range of uncertainties they had not previously experienced [13]. The individual will often reconstruct the threat of the perceived risk by denying its personal relevance. These are all important features to consider in addressing healthy weight and helping those who misclassify their weight status to understand the risks associated with excess weight. This paper is a preliminary exploration into the self-perception of weight status, a psychosocial aspect of obesity that is explored through three main themes—–individual and area level socioeconomic disadvantage and sociocultural influences. The aim of the paper is to firstly compare self-perception of weight with the World Health Organisation (WHO) definition for BMI categories. Secondly, to compare and discuss socioeconomic and sociocultural themes related to self-perception of weight using a population-based study.
Methods The North West Adelaide Health Study (NWAHS) is a biomedical cohort study of a representative population sample of adults living in the north west region
Socioeconomic and sociocultural explanations for perception of weight
127
Figure 1 Flowchart of North West Adelaide Health Study timeline 2000—2008.
of Adelaide [14]. The NWAH Study is spatially referenced data (at the individual point level) for 99% of the original cohort offering a unique opportunity to research sociodemographic and biomedical data within the north west metropolitan area of Adelaide. Further information on the methodology and study population of the North West Adelaide Health Study has been previously published [14]. The biomedical findings have been compared to other metropolitan regions or country areas in South Australia and limited biases were reported [15]. The NWAHS data were weighted to the Australian Bureau of Statistics 1999 Estimated Residential Population by the north west region, age group, sex and probability of selection in the household to provide estimates for the population. Of the total initial eligible sample (n = 4060), 90.1% (n = 3566) provided information for the second stage of the study (Stage 2, 2004—2006) and 81.0% (n = 3206) attended the clinic. Fig. 1 is a flowchart of the NWAHS cohort data collection 2000—2008. The participants’ self-perceived weight status was collected during a follow up of the NWAHS cohort
during August and September 2007 (n = 2382), a response rate of 74.3%. The participants answered the following survey question: ‘‘In terms of your weight, do you consider yourself to be. . .too thin, a little thin, normal weight, a little overweight or very overweight?’’ The objective weight status was based on measured BMI that was calculated using height in centimetres (measured to the nearest 0.5 cm using a stadiometer) and weight in kilograms (measured to the nearest 0.1 kg in light clothing and without shoes using a standard digital scales). The analyses covered three focus areas, individual socioeconomic status (household income), area level disadvantage index (SEIFA IRSD), and sociocultural influences (country of birth). Area level socioeconomic status was measured using the Index of Relative Social Disadvantage a component of the Socioeconomic Indexes for Areas, (SEIFA IRSD) as compiled by the Australian Bureau of Statistics (ABS) at the Collector District Level (around 200 or 300 households) [16]. Questions on the household income were asked in a self-reported paper based questionnaire sent out to the participants
N.J. Howard et al.
0.0 0.5 (n = 4) 3.2 (n = 29) 34.3 (n = 236) 0.0 18.3 (n = 138) 62.1 (n = 561) 59.6 (n = 410)
100.0 100.0 100.0 100.0
(n = 37) (n = 753) (n = 904) (n = 688)
before each clinic appointment. Each participant was asked about their country of birth during the original recruitment for the study at Time 1. The null hypothesis to be tested is that there is no relationship between three measures of socioeconomic status and self-perceived weight status. Self-perceived weight status was compared with the biomedically measured WHO BMI classifications by crosstabulation and Chi-square testing. The overestimation, agreement and underestimation of self-perceived weight status was analysed by gender. Binary logistic regression was used to compare those participants with a BMI ≥ 30 with the self-perceived weight status of ‘a little overweight’ (n = 1109). The outcome measures included the SEIFA IRSD, country of birth and household income.
Results
WHO BMI (World Health Organisation Body Mass Index).
Underweight (BMI < 18.50) Normal (BMI 18.50—24.99) Overweight (BMI 25.00—29.99) Obese (BMI ≥ 30)
(n = 26) (n = 112) (n = 12) (n = 2)
(n = 11) (n = 449) (n = 302) (n = 40) 29.7 66.3 33.4 5.8 70.3 14.9 1.3 0.3
Very overweight (%) A little overweight (%) Normal weight (%) Too thin/a little thin (%)
Self-perception of weight WHO BMI classifications
Table 1
Comparison of self-perception of body weight and biomedically measured WHO BMI classifications
Total (%)
128
The prevalence of measured obesity within the North West Adelaide Health Study at Time 1 (2000—2003) was 27.0%. Table 1 is a comparison of self-perception of body weight and biomedically measured WHO BMI classifications. Of those that were biomedically measured as obese (BMI ≥ 30), 59.6% (n = 410) perceived their weight to be ‘a little overweight’ and 5.8% (n = 40) perceived they were of ‘normal’ weight. Of those people that were overweight, 33.4% (n = 302) actually thought that they were of ‘normal’ weight status. In addition, 29.7% (n = 37) of people who were underweight thought they were of ‘normal’ weight, and of those that were biomedically measured with a BMI between 18.00 and 24.99 (or normal), 18.3% thought that they were ‘a little overweight’ and 0.5% thought that they were ‘very overweight’.
Table 2 Underestimation, agreement and overestimation of self-perceived weight status by gender Self-perception of weight status agreement status with BMI n
%
Males Underestimated Agreement Overestimated
493 637 58
41.5 53.6 4.9
Females Underestimated Agreement Overestimated
384 684 125
32.2 57.3 10.5
Socioeconomic and sociocultural explanations for perception of weight Table 3 Association with self-perceived ‘a little overweight’ and obesity (BMI ≥ 30) ‘A little overweight’
Obesity (BMI ≥ 30) OR (95% CI)
SEIFA (IRSD) Highest quintile High quintile Middle quintile Low quintile Lowest quintile Country of birth Australia UK/Ireland Eastern or Western Europe Asia and other Household income Greater then $60,000 $40.001 to $60,000 $20,001 to $40,000 Up to $20,000
P
1.00 0.89 0.77 1.11 1.67
(0.46—1.73) (0.51—1.16) (0.76—1.61) (1.14—2.46)
0.739 0.205 0.583 0.008
1.00 1.17 (0.83—1.64) 1.72 (1.07—2.75)
0.369 0.024
0.68 (0.27—1.75)
0.428
1.00 1.32 (0.95—1.85)
0.099
1.35 (0.96—1.89)
0.085
1.82 (1.27—2.61)
0.001
SEIFA IRSD (Socioeconomic Indexes for Areas Index of Relative Socioeconomic Disadvantage).
Table 2 explores by gender those people that underestimated, overestimated or had agreement with their self-perceived weight status and body mass index classification. Underestimation of selfperceived weight status was found to be higher for males (41.5%) compared to females (32.2%). There were also found to be 10.5% of females overestimating their weight status, that is having the self-perception of their weight status being heavier when compared with their biomedical classifications of BMI. Table 3 describes obesity by those factors associated with self-perceived weight status ‘a little overweight’. Those who were biomedically obese but self-perceived themselves ‘a little overweight’ were more likely to be from the lowest quintile of SEIFA (OR 1.67), born in Eastern or Western Europe (OR = 1.72) and have an annual household income of less than $20,000 (OR = 1.82).
Discussion Previous research in South Australia has shown increasing rates of overweight and obesity in
129
adults and children [17—19]. In this particular population there was a high percentage of people in this study who were misclassifying their weight status. Around two-thirds (65.7%) of those that were biomedically obese were misclassifying their weight status by not reporting their current weight to be ‘very overweight’. The obesity epidemic confronts us with emerging issues across the social gradient. The relationship of obesity with socioeconomic status is complex with both those that are highly educated and of a higher income experiencing increasing prevalence as well as a lower prevalence of the condition [20]. Those within the lowest quintile of disadvantage and low household income levels were found to be more likely to interpret their weight status as ‘a little overweight’ when they really were biomedically obese. The social and physical environmental influences and their relationship with an individual’s health status is yet to be fully determined, although research has suggested that individuals living in disadvantaged areas are at greater risk of obesity regardless of their individual level of disadvantage [21,22]. This research has shown that it is those that are locationally and socioeconomically disadvantaged who are more likely to underreport their weight. This may mean an individual’s temporal awareness of the ‘risk’ of chronic disease further into their future, and the association with social environments in which these individuals exist, is yet to be fully determined [23]. People born in Eastern or Western Europe were more likely to misinterpret their weight status to be a ‘little overweight’ when they were actually biomedically obese when compared to those that were Australian born. Within this population there have been limited studies exploring the perception of weight amongst different cultural groups. In the United States there have been a number of studies exploring the racial discrepancies between black, hispanic and white populations [24,25]. There are sociocultural factors that drive the standards of desirable body weight within cultures, which in turn drive the lifestyles that people lead. As the population mean of body mass index shifts further along an obesity continuum there is a need to determine whether these social and cultural norms and the perceptions of body image or weight are changing simultaneously. Of those people that were found to misreport their obesity status the distribution was varied along the BMI continuum, with 14.3% (n = 40) of those participants having a BMI > 35 or classified with severe obesity (BMI ≥ 35).1 Those that are mis1
Howard, NJ. Further analyses of data, 2007.
130 classifying their weight status are actually biomedically measured as being at high risk of a chronic condition and are therefore not just slightly misclassifying [26]. In addition, there were a high percentage of people that were misreporting their weight in the other direction, that is over reporting their weight status. These findings draw attention to issues such as body image, self-esteem and increasing eating disorders among our population concurrently with the rising obesity prevalence [27]. More investigations are needed to look at the complexities in the distribution of weight, for example waist to hip ratio, and other definitions related to the perception of weight status. It was the aim of this research to use a commonly defined measurement for obesity, the BMI. The analysis is undertaken with caution as BMI is an arbitrary measure for defining obesity in our population and an individual may not necessarily report incorrectly if other measurements of body fat distribution are taken into account. Numerous studies have reported that BMI is not the most appropriate measure in determining risk of chronic disease, rather other body fat and distribution measures provide more reliable estimates [28,29]. These discrepancies are apparent in this research with 0.3% self-reporting their obesity status as ‘too thin or a little thin’ when actually they have a BMI greater then 30. Further investigations highlighted that those misreporting in this case were males with a normal waist to hip ratio.2 This research draws upon Giddens’ Structuration Theory which places an individual within a broader context and accounts for the multilevel nature of simultaneous individual and group level interaction [30]. There is a two way process between individuals and the social environments in which they live, work and play. Giddens writes that ‘‘It is important to grasp how history is made through the active involvements and struggles of human beings, and yet at the same time both forms those human beings and produces outcomes which they neither intend nor forsee [31].’’ More research is needed to explore these processes of the individual and social environmental interaction in relation to healthy weight. There are a broad range of other sociocultural and psychosocial factors that could be explored in relation to the self-perception of weight status. These include the influence of other sociocultural themes such as lifecourse indicators such as mother’s and father’s country of birth, and influence of an individual’s life trajectory and social
2
Howard, NJ. Further analyses of data, 2007.
N.J. Howard et al. context which shape the contemporary social environment, both of which contribute to the cultural landscapes of consumption and behaviour. The research presented in this paper presents challenges for policy and intervention. The transposing of this research into healthy weight messages at the population level appears to be a difficult task if people do not perceive their weight to be a problem. An intervention is not likely to be effective if it is based on the assumption that people recognize their weight to be a problem when in actual fact they do not consider this to be the case. More recognition of the perception of risk and lay perspectives of obesity are needed to understand the complexity of obesity within our population. The strength in the NWAHS lies in the representative nature, the large random sample, and the relatively high response rate. In addition, the NWAHS has strength in the standardised clinical procedures with stringent measurement standards and regular calibration of scales and height measures to provide reliable BMI data. Self-reported height and weight has commonly been used to determine BMI at a population level. There are recognised problems associated with self-reported height and weight, in that height is generally over-reported and weight under-reported [32]. The self-report data have previously been shown to underestimate the extent of obesity within this population [33]. Future research would need to consider the different measures of obesity and cut off points to look at biomedically assessed weight and perceived body image. As well as investigating how to target interventions towards healthy weight when people are not aware they have a problem. The present paper has been able to make only some preliminary observations on one main theme area of the socio-spatial context, that is the perceptions of weight status. There is a need to gain an understanding of the processes whereby people are not aware that they have a problem when it comes to their weight. Future research of the cohort will explore the multilevel nature of these relationships and how the spatial landscape has the ability to influence individual level outcomes such as social reproduction of lifestyles, behaviour and social disadvantage.
Acknowledgements The authors would like to express their appreciation to the Population Research and Outcome Studies Unit, South Australian Department of Health for providing access to data. We would especially like to acknowledge the work of the North West Ade-
Socioeconomic and sociocultural explanations for perception of weight laide Health Study Team and the participants that have given up their valuable time. Natasha Howard is funded under a scholarship with the Nutrition Obesity Lifestyle and Environment (NOBLE) project which is an Australian Research Council (ARC) Linkage Project (LP0455737) ‘‘Obesity, Health, Social Disadvantage and Environment’’.
References [1] World Health Organisation. Obesity and overweight. Geneva: World Health Organisation; 2005. [2] Ball K, Crawford D. Socioeconomic status and weight change in adults: a review. Soc Sci Med 2005;60(9):1987—2010. [3] Swinburn B, Egger G. The runaway weight gain train: too many accelerators, not enough brakes. BMJ 2004;329 (7468):736—9. [4] EurActive Network. Scientists call for more research into obesity. EurActive Network: Brief News 1, 2006. [5] Hackie M, Bowles CL. Maternal perception of their overweight children. Public Health Nurs 2007;24(6):538—46. [6] Wake M, Salmon L, Waters E, Wright M, Hesketh K. Parent reported health status of overweight and obese Australian primary school children: a cross-sectorial population survey. Int J Obes Relat Metab Disord 2002;26(5):717—24. [7] Chang VW, Christakis NA. Self-perception of weight appropriateness in the United States. Am J Prev Med 2003;24(4):332—9. [8] Olvera N, Suminski R, Power TG. Intergenerational perceptions of body image in hispanics: role of BMI, gender, and acculturation. Obes Res 2005;13(11):1970—9. [9] Coulson FR, Ypinazar VA, Margolis SA. Awareness of risks of overweight among rural Australians. Rural Remote Health 2006;6:514 [online]. [10] Prochaska JO, Norcross JC. Stages of change. Psychotherapy 2001;38(4):443—8. [11] Macintyre S, McKay L, Ellaway A. Lay concepts of the relative importance of different influences on health; are there major socio-demographic variations? Health Educ Res 2006;21(5):731—9. [12] Popay J, Thomas C, Williams G, Bennett S, Gatrell A, Bostock L. A proper place to live: health inequalities, agency and the normative dimensions of space. Soc Sci Med 2003;57(1):55—69. [13] Kavanagh AM, Broom DH. Embodied risk: my body, myself? Soc Sci Med 1998;46(3):437—44. [14] Grant JF, Chittleborough CR, Taylor AW, Dal Grande E, Wilson DH, Phillips PJ, et al. The North West Adelaide Health Study: detailed methods and baseline segmentation of a cohort along a chronic disease continuum. Epidemiol Perspect Innov 2005;3(4). [15] Taylor A, Dal Grande E, Gill T, Chittleborough CR, Wilson DH, Adams RJ, et al. Do people with risky behaviours participate in biomedical cohort studies? BMC Public Health 2006;6(1):11.
131
[16] Australian Bureau of Statistics. Socio-economic indexes for areas. In: Trewin D, editor. Information paper: census of population and housing. Canberra: Australian Bureau of Statistics (ABS); 2001. p. 1—29. [17] Gill T, Appleton S, Dal Grande E, Taylor A, Carter P, Wilson D. A weighty issue: the status of overweight and obesity in South Australian adults from 1991 to 2001. Adelaide, Department of Human Services; 2004. p. 1—71. [18] Dal Grande E, Gill T, Taylor AW, Chittleborough CR. Obesity in South Australia adults—–prevalence, projections and generational assessment over 13 years. Aust N Z J Public Health 2005;29(4):343—8. [19] Hugo G, Franzon J. Child obesity in South Australia: some initial findings. Food Culture Soc Int J Multidisciplinary Res 2006;9:299—316. [20] McLaren L. Socioeconomic status and obesity. Epidemiol Rev 2007;29(1):29—48. [21] Berkman LF, Kawachi I. Social Epidemiology. New York: Oxford University Press; 2000. [22] King T, Kavanagh AM, Jolley D, Turrell G, Crawford D. Weight and place: a multilevel cross-sectional survey of area-level social disadvantage and overweight/obesity in Australia. Int J Obes Relat Metab Disord 2005;30(2):281—7. [23] Leccardi C. About future and planning. future matters: futures known, created and minded. In: New biographical constructions in the ‘‘Risk Society’’? Cardiff University; 2006 [September 4—6]. [24] Paeratakul S, White MA, Williamson DA, Ryan DH, Bray GA. Sex, race/ethnicity, socioeconomic status, and BMI in relation to self-perception of overweight. Obesity Res 2002;10(5):345—50. [25] Neff LJ, Sargent RG, McKeown RE, Jackson KL, Valois RF. Black—White differences in body size perceptions and weight management practices among adolescent females. J Adolesc Health 1997;20(6):459—65. [26] Haffner SM. Relationship of metabolic risk factors and development of cardiovascular disease and diabetes. Obesity 2006;14(suppl. 3):121S—7S. [27] Mission Australia. National Survey of Young Australians 2007 Key and Emerging Issues. Mission Australia. December 2007. [28] Dalton M, Cameron AJ, Zimmet PZ, Shaw JE, Jolley D, Dunstan DW, et al. Waist circumference, waist—hip ratio and body mass index and their correlation with cardiovascular disease risk factors in Australian adults. J Intern Med 2003;254:555—63. [29] Sangi H, Mueller WH. Which measure of body fat distribution is best for epidemiologic research among adolescents? Am J Epidemiol 1991;133(9):870—83. [30] Gatrell AC. Geographies of Health: An Introduction. United Kingdom: Blackwell Publishers; 2002. [31] Giddens A. Sociology: A Brief but Critical Introduction. London & Basingstoke: MacMillan Press Ltd.; 1982. p. 166. [32] Australian Bureau of Statistics. How Australians Measure Up. Canberra: ABS; 1999. [33] Taylor AW, Dal Grande E, Gill T, Chittleborough CR, Wilson DH, Adams RJ, et al. How valid are self-reported height and weight? A comparison between CATI self-report and clinic measurements using a large cohort study. Aust N Z J Public Health 2006;30(3):238—46.
Available online at www.sciencedirect.com
