Hostname: page-component-7c8c6479df-94d59 Total loading time: 0 Render date: 2024-03-27T04:17:38.326Z Has data issue: false hasContentIssue false

Sociodemographic factors and risk-taking behaviour during adolescence and obesity among more than 40 000 Danes

Published online by Cambridge University Press:  17 October 2012

Pernille Frederiksen
Affiliation:
Virus, Lifestyle and Genes, Danish Cancer Society Research Center, Strandboulevarden 49, DK-2100 Copenhagen, Denmark
Kirsten E Jensen
Affiliation:
Virus, Lifestyle and Genes, Danish Cancer Society Research Center, Strandboulevarden 49, DK-2100 Copenhagen, Denmark
Susanne K Kjaer*
Affiliation:
Virus, Lifestyle and Genes, Danish Cancer Society Research Center, Strandboulevarden 49, DK-2100 Copenhagen, Denmark Department of Obstetrics and Gynecology, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark
*
*Corresponding author: Email Susanne@cancer.dk
Rights & Permissions [Opens in a new window]

Abstract

Objective

The prevalence of obesity has increased in the last decades in the Western world. The aim of the present study was to examine the association between risk-taking behaviour in adolescence and adult obesity in men and women. Furthermore, we wished to describe social differences in obesity in Denmark.

Design

Two population-based questionnaire studies (2004–2005 and 2006–2007) were used to collect information on weight and height, sociodemographic factors and factors regarding risk-taking behaviour during adolescence. Data were analysed using multivariate logistic regression.

Setting

Denmark.

Subjects

Individuals aged 18–45 years (men: n 22 827, participation rate 71·0 %; women: n 20 870, participation rate 81·4 %).

Results

The prevalence of overweight and obesity was respectively 37·8 % and 10·6 % in men and 20·1 % and 9·7 % in women. In both sexes, obesity was found to be associated with older age, low level of schooling and living outside the capital centre. In relation to risk-taking behaviour, young age (≤13 years) at first intercourse significantly increased the odds of being obese in adulthood (men: OR = 1·34, 95 % CI 1·04, 1·71; women: OR = 1·66, 95 % CI 1·27, 1·99). In women specifically, young age at start drinking alcohol (≤12 years) was associated with obesity.

Conclusions

Sociodemographic factors, in particular age, level of schooling and area of residence, are associated with obesity in both men and women. Risk-taking behaviour during adolescence seems to cluster in both obese men and obese women, however most convincingly in women.

Type
Epidemiology
Copyright
Copyright © The Authors 2012 

In Western European countries as well as in the USA, the prevalence of obesity has increased during the past four decades in both men and women( Reference Baum and Ruhm 1 Reference Bendixen, Holst and Sørensen 3 ). According to the WHO the rise in national obesity prevalences can be characterized as an obesity epidemic( 4 ); however, some studies have indicated that the epidemic is levelling off in some specific groups from the late 1990s( Reference Sjöberg, Lissner and Albertsson-Wikland 5 , Reference Svendstrup, Knudsen and Jørgensen 6 ). Being a major cause of morbidity and of premature mortality( Reference Field, Coakley and Must 7 , Reference Stevens, Cai and Pamuk 8 ), obesity has substantial medical and social consequences for the individual and economic consequences for society( Reference Colagiuri, Lee and Colagiuri 9 ). Obesity is often defined as excess body fat tissue, but exact measurements of body fat tissue require expensive methods that are difficult to implement in epidemiological studies( Reference Ogden, Yanovski and Carroll 10 ). Consequently, obesity has been redefined as excess body weight and thus BMI, which expresses weight adjusted for height, has become a highly used tool to measure obesity( Reference Ogden, Yanovski and Carroll 10 ).

A social gradient has been found in the prevalence of obesity: in most Western countries, individuals with lowest social position have the highest prevalence of obesity( Reference Molarius, Seidell and Sans 11 Reference Due, Damsgaard and Rasmussen 14 ). Obesity is found to be associated with different measures of sociodemographic and socio-economic status, such as low level of schooling( Reference Molarius, Seidell and Sans 11 , Reference Gortmaker, Must and Perrin 12 , Reference Due, Damsgaard and Rasmussen 14 , Reference Duvigneaud, Wijndaele and Matton 15 ), living in rural areas( Reference Kuskowska-Wolk and Bergström 16 Reference Halkjær and Sørensen 18 ), low household income( Reference Gortmaker, Must and Perrin 12 , Reference Due, Damsgaard and Rasmussen 14 ) and unemployment( Reference Due, Damsgaard and Rasmussen 14 , Reference Ekholm, Davidsen and Hesse 19 ). The influence of social factors on obesity has been found to differ between sexes; more social factors have been found to be associated with obesity in women than in men and the associations between obesity and social factors have been found to be stronger and more consistent throughout the literature in women than in men( Reference Lathi-Koski, Vartiainen and Männistö 2 , Reference Sjöberg, Lissner and Albertsson-Wikland 5 , Reference Molarius, Seidell and Sans 11 , Reference Gortmaker, Must and Perrin 12 , Reference Sundquist and Johansson 20 ). In addition, sex differences in the historical development of obesity epidemics have also been seen. One study found that men had a steadily increasing prevalence of obesity in the 1970s and through up to the 2000s, whereas the prevalence of obesity in women remained stable until the 1990s after which an increase occurred similar to that in men( Reference Heitmann, Strøger and Mikkelsen 21 ). Another study focusing on the years 1999–2004 showed significant increases in obesity in men but not in women( Reference Ogden, Carroll and Curtin 22 ). These findings suggest that it is highly relevant to examine men and women separately when measuring the prevalence of obesity and in identifying the influence of social factors on obesity.

As obesity in adulthood has been found to be associated with social position both in adulthood and in childhood( Reference Parsons, Power and Logan 23 ), it could be hypothesized that other characteristics early in life such as risk-taking behaviour may also be associated with obesity in adulthood. This hypothesis may be supported by previous findings showing that teens from families characterized by low social position not only have higher risk of obesity in adulthood( Reference Parsons, Power and Logan 23 ), but also engage in risk-related behaviour (e.g. young age at first intercourse and sexual activity without protection) at a younger age than teens from families with high social position( Reference Forste and Heaton 24 ). However, no studies focusing on early risk-taking behaviour in relation to risk of obesity in adulthood have to our knowledge been published so far. Thus, the aim of the present study was to examine the distribution of overweight and obesity in relation to sociodemographic factors (age, level of schooling, area of residence and marital status) among men and women in two large, population-based, cross-sectional studies from Denmark. Furthermore, we wanted to examine the potential association between obesity in adulthood and sociodemographic characteristics in adulthood and variables reflecting risk-taking behaviour in adolescence (young age at initiation of smoking, young age at initiation of alcohol consumption and young age at first sexual intercourse) in a multivariate analysis including both sociodemographic factors and signs of risk-taking behaviour.

Methods

Study population

The present analysis is based on data from two previously described population-based studies concerning lifestyle habits among men and women( Reference Nielsen, Munk and Liaw 25 , Reference Kjaer, Tran and Sparen 26 ). Both studies were approved by the Danish Data Protection Agency. In Denmark all residents are assigned a unique personal identification number (PIN) which comprises information on date of birth and sex. The PIN is registered in the computerized Danish Civil Registration System. In brief, random samples of Danish men and women aged 18–45 years were drawn from the Civil Registration System. From November 2006 to July 2007 and from November 2004 to July 2005 respectively, 33 000 men and 28 000 women living in Denmark were invited to participate in the study. Individuals who had moved, emigrated or died before contact and those who could not speak Danish were ineligible for the study (n men 487, n women 728). In addition, no contact was established or participation was actively denied (by telephone, email or letter) by 9434 men and 5073 women. In total, 23 079 men (response rate 71·0 %) and 22 199 women (response rate 81·4 %) were included in the study. Subsequently, we excluded twelve men and twenty-six women because of discrepancies between their PIN and self-reported year of birth; and 238 men and 1301 women were excluded due to missing answers to the core questions concerning weight, height, marital status or level of schooling. Finally, two men and two women were excluded as outliers as they had BMI lower than 15·0 kg/m2 or higher than 69·0 kg/m2, leaving 22 827 men and 20 870 women available for analysis. Ethical approval was not required for the secondary data analyses reported here.

Data collection

Identical data collection methods were used for the male and the female surveys. All potential participants were appointed a unique study number to guarantee confidentiality and received an invitation letter and a self-administered questionnaire along with a stamped and addressed envelope. Alternatively to returning the questionnaire by postal mail, the participants had the possibility of answering an identical web-based questionnaire. Individuals who did not respond within four weeks received a reminder. For those who still did not respond, telephone interviews were attempted comprising the same questions as the self-administered questionnaire. The questionnaire contained questions about sociodemographic factors, smoking history, alcohol consumption and sexual history.

Self-reported body weight and body height were used to calculate individual BMI values using the standard formula provided by the WHO: body weight (kg)/(body height (m))2. To categorize the BMI values, we used standard reference values also provided by the WHO( 4 ): underweight was defined as BMI < 18·5 kg/m2, normal weight as BMI = 18·5–24·9 kg/m2, overweight as BMI =25·0–29·9 kg/m2 and obesity was defined as BMI ≥ 30·0 kg/m2. Age was categorized in five-year age groups. In Denmark, nine years of schooling is mandatory, the tenth year of schooling is voluntary and eleven or more years of schooling indicate high school/gymnasium. Thus we categorized level of schooling accordingly into three categories: low (≤9 years of schooling), middle (10 years of schooling) and high (≥11 years of schooling). The variable area of residence contained seven categories: capital centre, northern capital areas, southern capital areas, large provincial city areas, small provincial city areas, rural areas and peripheral rural areas( Reference Faber, Nielsen and Kjaer 27 ). This categorization made it possible to identify possible differences between urban and rural areas in Denmark and to distinguish between urban areas of the capital, i.e. the northern municipalities in the capital area are characterized by a high proportion of individuals with high social position whereas the southern municipalities have a high proportion of individuals with low social position( Reference Glümer, Hilding-Nørkjær and Jensen 28 ). The variable marital status consisted of two categories: married/cohabiting and not cohabiting, whereas the three variables reflecting different levels of risk-taking behaviour (age at initiation of smoking, age at initiation of alcohol consumption and age at first sexual intercourse) were divided into four or five categories enabling us to identify differences in age at initiation of risk-taking behaviour.

Statistical analysis

Initially, we described the sex-specific BMI distribution as well as assessed the proportions of men and women who were normal weight, overweight or obese in relation to four sociodemographic factors: age, level of schooling, area of residence and marital status. Subsequently, we examined the association between obesity and sociodemographic factors and risk-taking behaviour using multivariate logistic regression by estimating odds ratios and corresponding 95 % confidence intervals. In the logistic regression analysis individuals with normal weight served as the comparison group for obese individuals to have a clear distinction between the two body weight groups. We show age-adjusted odds ratios and odds ratios where all variables were mutually adjusted. For statistical analysis, the SAS/STAT statistical software package version 8·2 was used.

Results

BMI distribution

The sex-specific self-reported BMI distribution is shown in Fig. 1. The distribution of BMI in the male study population peaked at 22·0–24·0 kg/m2, with a median BMI of 24·9 kg/m2 (25th–75th percentile: 22·9–27·4 kg/m2; data not shown), whereas in the female study population the highest proportions of women had a BMI in the range 20·0–22·0 kg/m2 with a median BMI of 22·8 kg/m2 (25th–75th percentile: 20·8–25·7 kg/m2; data not shown). Furthermore, the male study population had a higher proportion of overweight individuals than the female study population.

Fig. 1 Distribution of BMI in the study population: men (n 22 827; - - ● - -) and women (n 20 870; - - ○ - -) aged 18–45 years, Denmark, 2004–2005 (women) and 2006–2007 (men)

Prevalence of overweight and obesity in relation to sociodemographic factors

Tables 1 and 2 show the distribution of normal weight, overweight and obesity according to age, level of schooling, area of residence and marital status for men and women, respectively. The prevalence of overweight was significantly higher in men (37·8 %, 95 % CI 37·2, 38·4 %) than in women (20·1 %, 95 % CI 19·5, 20·6 %), whereas the prevalence of obesity was similar (men: 10·6 %, 95 % CI 10·3, 11·1 %; women: 9·7 %, 95 % CI 9·3, 10·1 %). In both men and women the largest proportion of obesity was found among the oldest study participants (40–45-year-olds) (men: 13.4 %; women: 11·8 %), in individuals with low level of schooling (men: 15·5 %; women: 16·4 %), in individuals living in rural areas (men: 14·7 %; women: 15·0 %) and in married/cohabiting men and women (men: 11·4 %; women: 10·4 %).

Table 1 Prevalence of normal weight, overweight and obesity according to sociodemographic factors among men (n 22 827) aged 18–45 years, Denmark, 2006–2007

*Of the total male study population 191 men were underweight.

Table 2 Prevalence of normal weight, overweight and obesity according to sociodemographic factors among women (n 20 870) aged 18–45 years, Denmark, 2004–2005

*Of the total female study population 866 women were underweight.

Factors associated with obesity

Table 3 displays age-adjusted and mutually adjusted odds ratios for the associations between obesity, early risk-taking behaviour and sociodemographic factors in men and women. In both men and women, obesity was more likely among those who reported a younger age at first sexual intercourse (e.g. ≤13 years, men: OR = 1·34, 95 % CI 1·04, 1·71; women: OR = 1·66, 95 % CI 1·27, 2·19) and in women who initiated alcohol consumption at a young age (≤12 years; OR = 1·52, 95 % CI 1·05, 2·20). In addition, young age at initiation of smoking (≤13 years) tended to be associated with obesity, although the associations did not reach statistical significance in the mutually adjusted analysis. Furthermore, men and women who had never smoked, and men and women who did not consume alcohol, were more likely to be obese. With regard to the sociodemographic factors, there was an overall tendency that the risk of obesity increased with increasing age, decreasing level of schooling and living in areas outside the capital centre even after mutual adjustment and adjustment for the behavioural factors. In terms of marital status we found in the mutually adjusted analysis that men who were married/cohabiting had a statistically significantly increased risk of obesity compared with men who were not cohabiting (OR = 1·15, 95 % CI 1·03, 1·30), whereas for women marital status was not associated with obesity in the mutually adjusted analysis.

Table 3 Associations between obesity and risk-taking behaviour and sociodemographic factors among men (n 12 960)* and women (n 14 250)* aged 18–45 years, Denmark, 2004–2005 (women) and 2006–2007 (men)

*Study participants with missing values on marital status, age at initiation of smoking, age at initiation of alcohol consumption or age at first sexual intercourse were excluded from the analyses (1045 men/1567 women).

†Adjusted for age.

‡Mutually adjusted.

§Confidence interval does not include 1·00.

Discussion

In the current population-based study of more than 43 000 Danish men and women aged 18–45 years we found that nearly 40 % of the men and 20 % of the women were overweight, while ∼10 % of both men and women were obese. This is in line with findings from other studies in Nordic countries( Reference Lathi-Koski, Vartiainen and Männistö 2 , Reference Ekholm, Davidsen and Hesse 19 , Reference Reas, Nygård and Svensson 29 ). A higher prevalence of overweight in men than in women has been a common finding in most European populations( 30 ), and the overall sex-specific distributions of BMI in our study showed that the BMI distribution in the male study population was skewed to the right compared with the female study population. This pattern resembles the British male and female BMI distribution curves most recently presented by the National Obesity Observatory in England( 31 ).

The sex-specific differences in the prevalence of overweight found in many studies can theoretically be caused by a misclassification of muscularly built men as being overweight instead of being normal weight. The misclassification takes place because of the difference in weight of muscle tissue and body fat tissue and because BMI measures excessive body weight and is not an optimal measure of body fat tissue( Reference Deurenberg, Andreoli and Borg 32 ). However, a study conducted in five European populations found that BMI is a valid overall predictor of the body fat mass percentage when used at a group level( Reference Deurenberg, Andreoli and Borg 32 ). In combination with BMI waist circumference measurement could be a useful measure of abdominal fat( Reference Pischon, Boeing and Hoffmann 33 ); however, this measurement was not available in our study.

In agreement with findings from other large surveys, the prevalence of obesity in our study increased with increasing age in both men and women( Reference Duvigneaud, Wijndaele and Matton 15 , Reference Ekholm, Davidsen and Hesse 19 ). Our finding of an association between low level of schooling and higher risk of obesity is similar to findings in several other studies in economically developed countries( Reference Molarius, Seidell and Sans 11 Reference Sarlio-Lähteenkorva and Lahelma 13 , Reference Duvigneaud, Wijndaele and Matton 15 , Reference Sundquist and Johansson 20 ). This may reflect differences in lifestyle such as less physical activity during leisure time and poorer dietary habits in individuals with a lower education( Reference Groth, Fagt and Stockmarr 34 ), and furthermore, differences in perceptions of healthy body weight or importance of focusing on health may play a role( Reference Jeffery and French 35 ). By contrast, in less-developed countries, primarily in Central and Eastern Europe, low level of schooling is found to be associated with low risk of obesity( Reference Molarius, Seidell and Sans 11 , Reference Sarlio-Lähteenkorva and Lahelma 13 ), which may reflect limited access to food and necessity for hard physical labour in individuals of low social position( Reference Ogden, Yanovski and Carroll 10 ).

Our findings of a higher likelihood of obesity in men and women living in rural areas or areas outside the capital centre are also consistent with findings from other studies( Reference Kuskowska-Wolk and Bergström 16 Reference Halkjær and Sørensen 18 , Reference Glümer, Hilding-Nørkjær and Jensen 28 ). A study from Finland examining causes of BMI differences in relation to urbanization found that a clustering of individuals with a similar social status and an equal BMI status could be explained by both social selection (individuals of similar social position and with similar body composition choose to live in the same areas) and social causation (individuals who live in the same areas are affected by the local culture, e.g. in eating patterns, habits of physical activity, preferred body image)( Reference Jokela, Kivimäki and Elovainio 17 ). Furthermore, we found that the risk of obesity in the southern capital areas was similar to the risk observed in the more rural areas. This might be explained by a higher density of individuals with low social position in southern capital areas( Reference Glümer, Hilding-Nørkjær and Jensen 28 ) and indicates that area of residence is correlated with other factors important for obesity.

In relation to marital status in the mutually adjusted analysis we found that married/cohabiting men were slightly more likely to be obese than men who are not cohabiting, whereas we did not find any association in women. Studies have reported varying results with regard to marital status( Reference Due, Damsgaard and Rasmussen 14 , Reference Ekholm, Davidsen and Hesse 19 , Reference Sundquist and Johansson 20 , Reference Tavani, Negri and La Vecchia 36 Reference Rissannen, Heliövaara and Knekt 38 ); one study reported results consistent with ours( Reference Sjobal, Rauschenbach and Frongillo 37 ) and a study from Sweden found that entering marriage or moving in with a partner often results in weight gains( Reference Sundquist and Johansson 20 ).

In our study we used young age at first intercourse, young age at start smoking and young age at start drinking to measure signs of risk-taking behaviour in adolescence. This is not an exhaustive measure of risk-taking behaviour, but it gives an indication of a certain behavioural pattern in adolescence. In both sexes we found an association between obesity and young age at initiation of smoking and young age at first intercourse. In relation to alcohol consumption, about a 50 % higher likelihood of obesity was found in women who were young when initiating alcohol consumption but not in men. It has been a consistent finding in other studies that socially related factors are more consistent and stronger contributors in women's risk of obesity than in men's( Reference Lathi-Koski, Vartiainen and Männistö 2 , Reference Molarius, Seidell and Sans 11 , Reference Gortmaker, Must and Perrin 12 , Reference Due, Damsgaard and Rasmussen 14 , Reference Sundquist and Johansson 20 , Reference Tavani, Negri and La Vecchia 36 ). Our overall findings of positive associations between early risk-taking behaviour and obesity after adjusting for sociodemographic factors indicate that behavioural characteristics in adolescents may influence obesity in adulthood. In both men and women we found that never smoking and never drinking alcohol also were associated with obesity. In relation to smoking, these findings are largely in line with those from other studies, which found that non-smokers have higher risk of obesity compared with individuals who are current smokers( Reference Duvigneaud, Wijndaele and Matton 15 , Reference Tavani, Negri and La Vecchia 36 ). In relation to alcohol, a similar pattern was found in a Belgian study in which it was reported that individuals with a moderate alcohol intake have a reduced risk of obesity compared with those who never drink alcohol( Reference Duvigneaud, Wijndaele and Matton 15 ). In contrast, other studies did not find an association between alcohol consumption and obesity( Reference Tavani, Negri and La Vecchia 36 ) and a study from Finland only found an association between obesity and alcohol consumption in men but not in women( Reference Rissannen, Heliövaara and Knekt 38 ). We cannot explain these findings between no alcohol consumption and obesity.

The strengths of our study include the random sampling of study participants and the high response rates, both increasing the generalizability of the study results to the general population of men and women being 18–45 years old and living in Denmark. Furthermore, the large study populations imply a greater statistical strength. The study also has some potential limitations. In spite of high response rates we cannot rule out the possibility that selection bias has occurred due to lack of participation. The fact that we use self-reported data to estimate BMI may also be a limitation; some studies have found a tendency of underestimation of BMI in both men and women( Reference Madrigal, Sánches-Villegas and Martínez-Gonzsález 39 , Reference Hill and Roberts 40 ). However, other studies have found that BMI is a useful measurement of obesity in large populations due to its accuracy and high cost-effectiveness when used on a population scale( Reference Deurenberg, Andreoli and Borg 32 , Reference Snijder, van Dam and Visser 41 ). Finally, some studies have found that under-reporting of BMI is associated with obesity, young age and low educational level( Reference Sonne-Holm, Sørensen and Jensen 42 , Reference Boström and Diderichsen 43 ), explaining why our results regarding the social differences in prevalence of obesity potentially could be underestimated.

Conclusions

We found strong associations between obesity and increasing age, decreasing level of schooling and area of residence in both men and women. Furthermore, initiation of sexual activity (intercourse), tobacco smoking and alcohol drinking at an early age were associated with obesity, the associations being strongest in women. On the basis of our findings of some differences between men and women in the risk of obesity, further research into the sex-specific characteristics in social background, lifestyle and health behaviour will be relevant.

Acknowledgements

Sources of funding: This research received no specific grant from any funding agency in the public, commercial or not-for-profit sector. Conflicts of interest: The authors do not have any conflict of interest to declare. Authors’ contribution: P.F. contributed to the conception and design of the study, did the data analysis, discussed core ideas, drafted the article and prepared the final manuscript. K.E.J. contributed to the conception and design of the study, discussed core ideas and commented on drafts. S.K.K. contributed to the conception and design of the study, led the project, discussed core ideas and commented on drafts. All authors read and approved the final manuscript.

References

1. Baum, CL & Ruhm, CJ (2009) Age, socioeconomic status and obesity growth. J Health Econ 28, 635648.Google Scholar
2. Lathi-Koski, M, Vartiainen, E, Männistö, S et al. (2000) Age, education and occupation as determinants of trends in body mass index in Finland form 1982 to 1997. Int J Obes Relat Metab Disord 24, 16691676.CrossRefGoogle Scholar
3. Bendixen, H, Holst, C, Sørensen, TIA et al. (2004) A major increase in prevalence of overweight and obesity between 1987 and 2001 among Danish adults. Obes Res 12, 14641472.Google Scholar
4. World Health Organization (2009) Global Database on Body Mass Index. http://www.who.int/bmi/index.jsp?introPage=intro_3.html (accessed May 2010).Google Scholar
5. Sjöberg, A, Lissner, L, Albertsson-Wikland, K et al. (2008) Recent anthropometric trends among Swedish school children: evidence for decreasing prevalence of overweight in girls. Acta Paediatr 97, 118123.Google Scholar
6. Svendstrup, M, Knudsen, NJ, Jørgensen, T et al. (2011) Stagnation in body mass index in Denmark from 1997/1998 to 2004/2005, but with geographical diversity. Dan Med Bull 58, A4344.Google ScholarPubMed
7. Field, AE, Coakley, EH, Must, A et al. (2001) Impact of overweight on the risk and developing common chronic diseases during a 10-year period. Arch Intern Med 161, 15811586.Google Scholar
8. Stevens, J, Cai, J, Pamuk, ER et al. (1998) The effect of age on the association between body mass index and mortality. N Engl J Med 338, 17.Google Scholar
9. Colagiuri, S, Lee, CM, Colagiuri, R et al. (2010) The cost of overweight and obesity in Australia. Med J Aust 192, 260264.CrossRefGoogle ScholarPubMed
10. Ogden, CL, Yanovski, SZ, Carroll, MD et al. (2007) The epidemiology of obesity. Gastroenterology 132, 20872102.Google Scholar
11. Molarius, A, Seidell, JC, Sans, S et al. (2000) Educational level, relative body weight and changes in their association over 10 years: an international perspective from the WHO MONICA Project. Am J Public Health 90, 12601268.Google Scholar
12. Gortmaker, SL, Must, A, Perrin, JM et al. (1993) Social and economic consequences of overweight in adolescence and young adulthood. N Engl J Med 329, 10081012.Google Scholar
13. Sarlio-Lähteenkorva, S & Lahelma, E (1999) The association of body mass index with social and economic disadvantage in women and men. Int J Epidemiol 28, 445449.Google Scholar
14. Due, P, Damsgaard, MT, Rasmussen, M et al. (2009) Socioeconomic position, macroeconomic environment and overweight among adolescents in 35 countries. Int J Obes (Lond) 33, 10841093.Google Scholar
15. Duvigneaud, N, Wijndaele, K, Matton, L et al. (2007) Socio-economic and lifestyle factors associated with overweight in Flemish adult men and women. BMC Public Health 7, 2332.Google Scholar
16. Kuskowska-Wolk, A & Bergström, R (1993) Trends in body mass index and prevalence of obesity in Swedish women 1980–89. J Epidemiol Community Health 47, 195199.Google Scholar
17. Jokela, M, Kivimäki, M, Elovainio, M et al. (2009) Urban/rural differences in body weight: evidence for social selection and causation hypotheses in Finland. Soc Sci Med 68, 867875.Google Scholar
18. Halkjær, J & Sørensen, TIA (2004) Psychological and demographic determinants of regional differences in the prevalence of obesity. J Biosoc Sci 36, 141152.Google Scholar
19. Ekholm, O, Davidsen, M, Hesse, U et al. (2006) Sundhed og Sygelighed i Danmark 2005 og udviklingen siden 1987 (Health and Morbidity in Denmark 2005 and Their Trends Since 1987). København: Statens Institut for Folkesundhed.Google Scholar
20. Sundquist, J & Johansson, S (1998) The influence of socioeconomic status, ethnicity and lifestyle on body mass index in a longitudinal study. Int J Epidemiol 27, 5763.Google Scholar
21. Heitmann, BL, Strøger, U, Mikkelsen, K et al. (2003) Large heterogeneity of the obesity in Danish adults. Public Health Nutr 7, 453460.CrossRefGoogle Scholar
22. Ogden, CL, Carroll, MD, Curtin, LR et al. (2006) Prevalence of overweight and obesity in the United States, 1999–2004. JAMA 295, 15491555.Google Scholar
23. Parsons, TJ, Power, C, Logan, S et al. (1999) Childhood predictors of adult obesity: a systematic review. Int J Obes Relat Metab Disord 23, Suppl. 8, S1S107.Google Scholar
24. Forste, RT & Heaton, TB (1988) Initiation of sexual activity among female adolescents. Youth Soc 19, 250268.Google Scholar
25. Nielsen, A, Munk, C, Liaw, K et al. (2009) Awareness of human papillomavirus in 23,000 Danish men from the general male population. Eur J Cancer Prev 18, 236239.Google Scholar
26. Kjaer, SK, Tran, TN, Sparen, P et al. (2007) The burden of genital warts: a study of nearly 70,000 women from the general female population in the 4 Nordic countries. J Infect Dis 196, 14471454.Google Scholar
27. Faber, MT, Nielsen, A & Kjaer, SK (2010) Rygning blandt 45.000 tilfældigt udvalgte danskere (Smoking among more than 45,000 randomly sampled Danish women and men). Ugeskr Laeger 172, 27562761.Google Scholar
28. Glümer, C, Hilding-Nørkjær, H, Jensen, HN et al. (2008) Sundhedsprofil for region og kommuner 2008 (Health Profile for the Region and Municipalities 2008). Region Hovedstaden: Forskningscenter for Forebyggelse og Sundhed.Google Scholar
29. Reas, LR, Nygård, JF, Svensson, E et al. (2007) Changes in body mass index by age, gender, and socio-economic status among a cohort of Norwegian men and women (1990–2001). BMC Public Health 7, 269275.Google Scholar
30. British Heart Foundation (2010) Statistics. http://www.bhf.org.uk/research/statistics/risk-factors/obesity.aspx (accessed October 2010).Google Scholar
31. National Obesity Observatory (2010) NOO data briefing: Adult Weight. http://www.noo.org.uk/uploads/doc/vid_5463_NOO%20Adult%20Weight%20Data%20Briefing%2029%20March%202010.pdf (accessed October 2010).Google Scholar
32. Deurenberg, P, Andreoli, A, Borg, P et al. (2001) The validity of predicted body fat percentage from body mass index and from impedance samples of five European populations. Eur J Clin Nutr 55, 973979.Google Scholar
33. Pischon, T, Boeing, H, Hoffmann, K et al. (2008) General and abdominal adiposity and risk of death in Europe. N Engl J Med 359, 21052120.Google Scholar
34. Groth, MV, Fagt, S, Stockmarr, A et al. (2001) Social determinants of dietary habits in Denmark. Eur J Clin Nutr 55, 959966.CrossRefGoogle ScholarPubMed
35. Jeffery, RW & French, SA (1996) Socioeconomic status and weight control practices among 20- to 45-year old women. Am J Public Health 86, 10051010.Google Scholar
36. Tavani, A, Negri, E & La Vecchia, C (1994) Determinants of body mass index: a study from Northern Italy. Int J Obes Relat Metab Disord 18, 497502.Google Scholar
37. Sjobal, J, Rauschenbach, B & Frongillo, EA (1992) Marital status, fatness and obesity. Soc Sci Med 35, 915923.Google Scholar
38. Rissannen, AM, Heliövaara, M, Knekt, P et al. (1991) Determinants of weight gain and overweight in adult Finns. Eur J Clin Nutr 45, 419430.Google Scholar
39. Madrigal, H, Sánches-Villegas, A, Martínez-Gonzsález, M et al. (2000) Underestimation of body mass index through perceived body image as compared to self-reported body mass index In the European Union. Public Health 114, 468473.Google Scholar
40. Hill, A & Roberts, J (1998) Body mass index: a comparison between self-reported and measured height and weight. J Public Health Med 20, 206210.Google Scholar
41. Snijder, MB, van Dam, RM, Visser, M et al. (2006) What aspects of body fat are particularly hazardous and how do we measure them? Int J Epidemiol 35, 8392.Google Scholar
42. Sonne-Holm, S, Sørensen, TIA, Jensen, G et al. (1989) Influence of fatness, intelligence, education and socio-demographic factors on response rate in health survey. J Epidemiol Community Health 43, 369374.Google Scholar
43. Boström, G & Diderichsen, F (1997) Socioeconomic differentials in misclassification of height, weight and body mass index based on questionnaire data. Int J Epidemiol 25, 860866.CrossRefGoogle Scholar
Figure 0

Fig. 1 Distribution of BMI in the study population: men (n 22 827; - - ● - -) and women (n 20 870; - - ○ - -) aged 18–45 years, Denmark, 2004–2005 (women) and 2006–2007 (men)

Figure 1

Table 1 Prevalence of normal weight, overweight and obesity according to sociodemographic factors among men (n 22 827) aged 18–45 years, Denmark, 2006–2007

Figure 2

Table 2 Prevalence of normal weight, overweight and obesity according to sociodemographic factors among women (n 20 870) aged 18–45 years, Denmark, 2004–2005

Figure 3

Table 3 Associations between obesity and risk-taking behaviour and sociodemographic factors among men (n 12 960)* and women (n 14 250)* aged 18–45 years, Denmark, 2004–2005 (women) and 2006–2007 (men)