- Email: [email protected]
Low socioeconomic status may increase the risk of central obesity in incoming university students in Taiwan
Obesity Research & Clinical Practice (2014) 8, e212—e219
ORIGINAL ARTICLE
Low socioeconomic status may increase the risk of central obesity in incoming university students in Taiwan Chi-Yuan Chao a, Chi-Chen Shih b, Chi-Jen Wang c, Jin-Shang Wu b,d, Feng-Hwa Lu b,d, Chih-Jen Chang b,d, Yi-Ching Yang b,d,∗ a
Department of Family Medicine, Min-Sheng General Hospital, Ching-Kuo Campus, Taoyuan, Taiwan Department of Family Medicine, National Cheng Kung University Hospital, Tainan, Taiwan c Department of Nursing, College of Medicine, National Cheng Kung University, Tainan, Taiwan d Department of Family Medicine, College of Medicine, National Cheng Kung University, Tainan, Taiwan b
Received 23 December 2011 ; received in revised form 16 July 2012; accepted 26 July 2012
KEYWORDS Socioeconomic status; Central obesity
Summary Background: Obesity is related to social disparity. The objective of the study was to evaluate different indicators of parental SES with the association of central obesity in young adult Taiwanese students. Methods: This study was cross-sectionally designed and a total of 4552 subjects were recruited. Each subject completed a self-administrated questionnaire and received anthropometric and laboratory measurements. The indicators of SES in study subjects included parental education, occupation, household incomes, childhood and current index of social position (ISP), measured according to the modified Hollingshead’s ISP. Central obesity was defined as waist circumference ≥90 cm in men and ≥80 cm in women. Results: The prevalence of central obesity was 10.7% in this study. When compared to subjects with normal waist circumferences, subjects with central obesity were older, had a higher BMI, both systolic and diastolic blood pressure, a higher proportion of male gender, family history of diabetes and hypertension, alcohol consumption habit, and a higher proportion of low current household income, current parental blue collar occupational level, and lower current and childhood parental ISP level. Multivariate analysis showed the current parental household income and ISP were significantly higher indicators of risk of central obesity after adjustment for possible confounding factors. The odds ratios were 1.26 and 1.30, respectively.
∗ Corresponding author at: Department of Family Medicine, College of Medicine, National Cheng Kung University, 138, Sheng-Li Rd, Tainan 70403, Taiwan. Tel.: +886 6 2353535x5188; fax: +886 6 2091433. E-mail address: [email protected] (Y.-C. Yang).
1871-403X/$ — see front matter © 2012 Asian Oceanian Association for the Study of Obesity. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.orcp.2012.07.002
Low socieconomic status and central obesity
e213
Conclusions: Our results showed that low household income and current ISP were independently associated with the risk of central obesity. Therefore, young adults with low SES should be an important target group for prevention and management of central obesity in school health promotion programs. © 2012 Asian Oceanian Association for the Study of Obesity. Published by Elsevier Ltd. All rights reserved.
Obesity is a global health problem with high prevalence in both developed and developing countries [1]. Obesity increases the risk of several leading causes of death and diseases in the world [2]. Furthermore, the highly prevalent rate of obesity is not only in middle-aged adults, but also in children and young adults. Several studies have indicated that diseases and causes of death are unequal according to social status [3,4]. One possible reason attributed to the associated higher risk of morbidity and mortality in people with low SES has been lower participation in physical activity and a higher prevalence of obesity [5,6]. Many studies have reported the association between SES and the risk of general obesity, and poor SES in both childhood and adulthood have been implicated in obesity risk among adults [7—9]. However, to date, few studies have attempted to establish the relative importance of childhood [10] and adult SES affecting weight gain and obesity risk in adulthood [11]. One population cohort study showed that the SES in both childhood and adulthood are independently inversely associated with BMI and weight change, but the associations varied by the SES indicator used [12]. Central obesity might have higher predictive value with regard to health problems than BMI [13], but the majority of studies have used BMI as the general obesity index to investigate the association between obesity and SES. Moreover, in these studies, only the father’s education [14—16] or occupational level [17,18,7] was used to explore the relationship between childhood SES and obesity. Furthermore, different SES indicators might affect one’s health via different pathways at different development stages. Therefore, the use of a single SES index to explain the effect of SES on obesity risk is not enough, thus, considering both parents’ SES instead of only the father’s or the individual’s own SES might more accurately reflecting an unbiased association between SES and obesity. In addition, previous studies have also shown that the association between SES and obesity might vary by the economic development of the countries under investigation [1,19]. To our knowledge, there have been few studies investigating the
relationship between SES and central obesity in rapidly developing societies such as Taiwan. Therefore, the objective of this study was to evaluate the association of different parental SES indicators with central obesity in an incoming university student population in Taiwan.
Methods Subjects The baseline data was collected from an entrance health check-up survey in a university in 2007. All 5550 subjects received a health examination and completed a self-administrated structured questionnaire, which included demographic information, medical history, smoking, alcohol and coffee consumption habits, and physical exercise. After excluding subjects with incomplete data, a total of 4552 subjects were recruited for the study (2145 men, 61.8%; 1325 women, 38.2%). Informed consent was obtained from all of the study participants. Since they only agreed to have their questionnaire data and related examination results analyzed anonymously, any identifying information was kept confidential. The Ethical Committee for Human Research at National Cheng Kung University Hospital approved the study protocol.
Instruments Assessment of lifestyle and other factors Smoking habit was classified as current smokers (defined by at least one pack/month) and nonsmokers. Alcohol consumption was classified as drinkers (defined by at least one drink/week) and non-drinkers. Regular physical exercise was defined according to the recommendation of the American College of Sports Medicine guidelines of at least three times weekly vigorous exercise [20] (intense enough to cause sweating and/or heavy breathing/and/or to increase heart rate to a certain
e214 amount). A positive family history of diabetes and hypertension was defined as at least one of the firstdegree relatives having diabetes or hypertension.
Measurement of parental socioeconomic status The different parental SES of the subjects was collected using a self-reported questionnaire. The parental education level was classified as either high (educational years more than 12 years) or low education level groups. The parental occupation level also was classified as white or blue collar groups. The current household income was defined as the total income of the family per month, and it was further dichotomized into two groups, high (household income/month more than NTD 80,000) and low household income. The current household income cutoff value of NTD 80,000 per month, because the ratio between high and low current household income was about 4:6 (42.9%:57.1%) in our study, it approaches to 1:1. In addition, according the report on the survey of family income and expenditure, 2009 by Directorate-General of Budget, Accounting and Statistics, Executive Yuan, Republic of China, the average household income on 2009 was about NTD 887,605 per year [21], it was close to the cutoff value in our study. Parental past SES (at subjects’ third year in elementary school) and current SES were classified according to a modified Hollingshead’s index of social position (ISP), calculated by the sum of education level times 4 and occupation level times 7. The ISP was originally classified into five levels according to the score: Level I: 11—18, Level II: 19—29, Level III: 30—40, Level IV: 41—51, Level V: 52—55. The ISP level was further classified into three groups, low (I), middle (II and III) and high (IV and V) SES level in our study analysis.
Procedure All the anthropometric measurements, blood pressure (BP) and blood sampling were carried out by well-trained nurses. Body weight (to the nearest 0.1 kg) and height (to the nearest 0.1 cm) were measured using a certified machine (DETECTO, Webb City, MO, USA). Waist circumference (WC) was measured from the midway between the lower rib margin and the iliac crest at the end of normal expiration with the subjects standing (to the nearest 0.1 cm). Body mass index (BMI) was calculated as weight in kilogram divided by the square of height in meter. Two readings of BP (systolic and diastolic BP) were measured with subjects in a sitting
C.-Y. Chao et al. position with a blood pressure monitor (DINAMAP@ Pro Series 100, CRITIKON Company L.L.C. Assembled in Mexico) after at least 10 min of rest. Central obesity has been defined as WC 90 cm or more in men and 80 cm or more in women according to World Health Organization (WHO) Asian-Pacific diagnostic criteria. All subjects received blood sampling after a 10-h overnight fast and the tests including liver function, fasting glucose level, and lipid profile were done.
Data analysis All statistical analyses were performed by using the 15th version of SPSS software. The subjects were classified into two groups by central obesity. Clinical characteristics in the study were presented as mean ± SD or percent. Chi-square tests were used to compare the categorical variables between the two groups. In the logistical regression model, the relationship between central obesity and different parental SES was explored after adjustment for age, gender, smoking, alcohol consumption habits, physical exercise, and family history of diabetes. The odds ratio and 95% confidence intervals (CIs) for predictors were derived for each regression model. Statistical significance was defined as a p value less than 0.05.
Results Table 1 showed the clinical characteristics of the 4552 subjects according to the status of central obesity. The prevalence of central obesity was 10.7%, and there were significant differences in gender, age, body mass index, waist circumference, systolic and diastolic BP, cigarette smoking habit, and family history of diabetes and hypertension between normal and central obesity groups. Compared to the normal group, the central obesity group was older, had a higher BMI, both systolic and diastolic BP, a higher proportion of male gender, family history of diabetes and hypertension, alcohol consumption habit. Table 2 shows the comparisons of family SES characteristics between the normal and central obesity groups. It can been seen that there were significant differences in current household income, current parental occupation level, and current and childhood parental ISP level. Compared to the normal group, the central obesity group had a higher proportion of low current household income, current parental blue collar occupational level, and lower current and childhood parental ISP level.
Low socieconomic status and central obesity Table 1
e215
The comparison of clinical characteristics between the subjects according to the status of central obesity.a
Variables
Total (N = 4552)
Normal (n = 4063)
Central obesity (n = 489)
p-Value
Male sex (%) Age (years) Obesity Body Mass Index (BMI: kg/m2 ) Overweight (BMI ≥24) Obesity (BMI ≥27) Waist circumference (cm) Cigarette smoking (%) Alcohol consumption (%) Physical exercise (≥3 times/week)(%) Family history of diabetes (%) Family history of hypertension (%)
67.1 21.9 (4.6)
65.2 21.7 (4.3)
82.8 23.6 (6.4)
<0.0001*** <0.0001***
22.2 (3.6) 15.7 9.3 74.8 (10.6) 4.7 2.7 22.1 9.5 22.5
21.4 (2.6) 14.5 1.9 72.4 (7.9) 4.2 2.6 22.1 8.9 21.4
28.9 (3.5) 25.8 70.6 94.9 (8.0) 9.0 4.1 22.3 14.5 31.5
<0.0001*** <0.0001*** <0.0001*** <0.0001*** <0.0001*** 0.065 0.924 <0.0001*** <0.0001***
a * ** ***
Data are presented as mean ± SD or percent. p < 0.05. p < 0.01. p < 0.001.
The odd ratios and 95% CIs for the different SES indicators and other independent variables for predicting central obesity from six multivariate logistic regression models are shown in Table 3. The current parental household income and ISP were associated with a significantly higher risk of central obesity, even after adjustment for age, sex, cigarette smoking, alcohol consumption, physical exercise, and a family history of diabetes and hypertension. The odds ratio was 1.26 (95% CIs: 1.03—1.54) and 1.30 (95% CIs: 1.04—1.64) respectively.
Discussion Our study showed that household income and current ISP were associated with central obesity in an incoming university incoming student population, after adjustment for the potential confounding factors of lifestyle and family history of diabetes and hypertension. Thus, lower household income and current ISP independently increased the risk of central obesity in a young adult population in Taiwan.
Table 2 The comparison of different socioeconomic status between the subjects according to the status of central obesity.a Variables
Total (N = 4552)
Normal (n = 4063)
Central obesity (n = 489)
p-Value
Parental education: ≤12 years (%) Current household income: ≤80,000 NTD (%) Current parental occupation: blue collar (%) Parental occupation in childhood: blue collar (%) Current parental ISP I (%) II—III (%) IV—V (%) Parental ISP in childhood I (%) II—III (%) IV—V (%)
17.2 57.1 47.8 44.9
16.9 56.4 47.2 44.4
19.6 63.4 53.2 48.9
0.141 0.003** 0.013* 0.062
10.6 60.2 29.2
10.2 59.9 29.9
13.3 63.0 23.7
0.006**
9.7 60.5 29.6
8.5 60.3 30.2
12.7 62.2 25.2
0.017*
NTD, new Taiwan dollar; ISP, index of social position. a Data are presented as percent. * p < 0.05. ** p < 0.01. *** p < 0.001.
e216
Table 3
The multiple logistic regressions for the associations of different SES and central obesity.a
Variables
Model 1
Model 2
Model 3
Model 4
Model 5
Model 6
Parental education (<12 vs ≥12 years) Household income (<80,000 vs ≥80,000 NTD/month) Current parental occupation (blue vs. white collar) Childhood parental occupation (blue vs. white collar) Current ISP I vs. IV—V II—III vs. IV—V Childhood ISP I vs. IV—V II—III vs. IV—V Age (≥20 vs. <20 years) Gender (male vs. female) Cigarette smoking (yes vs. no) Alcohol consumption (yes vs. no) Physical exercise (yes vs. no) Family history of hypertension (yes vs. no) Family history of diabetes (yes vs. no)
0.98 (0.76—1.25)b
—
—
—
—
—
—
1.26 (1.03—1.54)
—
—
—
—
*
1.14 (0.94—1.39) 1.10 (0.91—1.34)
1.30 (1.04—1.64)* 1.24 (0.88—1.73)
1.37 (1.11—1.68)** 2.55 (1.98—3.27)*** 1.59 (1.11—2.29)*
1.33 (1.08—1.64)** 2.53 (1.97—3.24)*** 1.60 (1.11—2.30)*
1.34 (1.09—1.64)** 2.53 (1.97—3.24)*** 1.58 (1.10—2.27)**
1.35 (1.10—1.66)** 2.53 (1.97—3.25)*** 1.57 (1.10—2.62)*
1.34 (1.09—1.65)** 2.53 (1.97—3.25)*** 1.59 (1.10—2028)*
1.21 1.24 1.34 2.53 1.58
(0.97—1.52) (0.88—1.74) (1.09—1065)** (1.97—3.25)*** (1.10—2.27)*
1.17 (0.71—1.95)
1.21 (0.73—2.01)
1.19 (0.72—1.98)
1.19 (0.72—1.98)
1.19 (0.72—1.98)
1.20 (0.72—1.99)
0.96 (0.76—1.21) 1.63 (1.32—2.02)***
0.97 (0.77—1.23) 1.63 (1.32—2.02)***
0.97 (0.77—1.22) 1.63 (1.32—2.02)***
0.97 (0.76—1.22) 1.64 (1.32—2.02)***
0.97 (0.77—1.23) 1.64 (1.33—2.03)***
0.97 (0.77—1.23) 1.64 (1.33—2.03)***
1.53 (1.16—2.04)**
1.50 (1.13—1.99)**
1.51 (1.14—2.01)**
1.53 (1.15—2.02)**
1.52 (1.14—2.02)**
1.52 (1014—2.02)**
C.-Y. Chao et al.
NTD, new Taiwan dollar. a Dependant variable: central obesity; independent variables: sex, age, cigarette smoking, alcohol consumption, physical exercise, family history of diabetes and hypertension, socioeconomic status (model 1: parental education, model 2: household income, model 3: current parental occupation, model 4: childhood parental occupation, model 5: current ISP, model 6: childhood ISP). b Data are presented as odds ratio (95% confidence intervals). * p < 0.05. ** p < 0.01. *** p < 0.001.
Low socieconomic status and central obesity After adjusting for sex and age, the subjects in our study with lower SES were found to be shorter and heavier than those with higher SES. Our result was consisted with previous study [22] and implied that if we used BMI for evaluate the prevalence of obesity among the low SES group, we might have overestimated the prevalence of obesity, because of the effects of both increased numerator (body weight) and decreased denominator (body height). Therefore, in our study we used the waist circumference as an alternative measures for classifying obesity in order to avoid the effect of height. Previous studies have shown lower childhood parental SES to be associated with adulthood central obesity [7—9], high blood pressure [7,8,23], dyslipidemia [7,8,24], and insulin resistance [7,9]. Goodman showed parental education level and household income were independently inversely related to young adult obesity, while in their study, parental occupation level was not associated with obesity after adjusting for other social and economic factors (members of family, race and gender, etc.) [14]. Leino et al. reported parental occupation level to be associated with birth weight, childhood obesity, smoking and physical inactivity in a study on young adults in Finland. Furthermore, this association was found to persist to the stage of young adult [25]. Our study used s modified Hollingshead’s index of social position, taking into account parental education and occupation levels simultaneously, and it showed the lowest current ISP to be independently associated with the risk of central obesity. Household income might reflect on the ability to buy healthy food, to have good housing, good working conditions, and access to medical care resources [26]. Other variables have had a positive association with central obesity in multivariate analysis, including older age [27,28], male gender [27], smoking habit [28], a habit of regular physical activity [27,28], and a family history of hypertension [29] and diabetes [28,30]. These results were compatible with previous findings. There are several mechanisms used to explain how the parental SES influences the health of adulthood, such as the latency model, the pathway model and psychosocial stress [24,31]. The latency model emphasized that the health status of the fetus and the subjects during early childhood (such as low birth weight, short status, smaller head circumference and rapid body weight gain during childhood) are associated with the adulthood cardiovascular disease and metabolic
e217 abnormality because these conditions are related to poor maternal nutrition and household poverty [32,33]. The current hypothesis of the developmental origin of health and diseases (DOHaD) has raised the similar concept of the latency model. This new model of causality and of the mechanisms involved in the emergence and development of chronic diseases [34]. Poverty is associated with an unhealthy or insufficient diet, an unhealthy living environment, poor health care quality, crowded residential environments, and low SES [24,25,35]; these factors might influence the uterine environment of a fetus and early childhood behavior, thus, increasing exposure to risk factors and resulting in unhealthy lifestyles and diseases during adulthood. The pathway model suggests that early stage SES might influence adult health because the social and biological risk factors related to health will accumulate during different life stages [31]. Chris et al. showed that seven different social and biological risk factors during different stage of human development (birth weight, maternal health condition during childhood, divorce of parents, body height of adult, education level, smoking habit, and autonomy or control in the workplace) have a different magnitude of association with SES at birth, and thus subjects with low SES are more likely to have more risk factors [31]. The psychosocial stress model suggested that poverty during childhood might result in psychosocial stressors related to mood, social identity, and dignity. The chronic psychosocial stress associated with poverty will in turn, according to this theory, result in cardiovascular disease and metabolic dysregulation in adulthood through the hypothalamus-pituitary-adrenal axis [14,15,36]. Therefore, childhood parental SES might influence adult health through different mechanisms, during different stages of development. Our study showed that the lower the childhood SES, the higher the adulthood obesity risk, but the results did not reach statistical significance. Ebrahim et al. showed that the effect of the childhood SES as an explanation of the adulthood metabolic abnormality to be attenuated after adjustment for adulthood SES, because its effect on health might decrease through the effect of current economic or social factors [8], and the allostatic load model could explain this effect. Allostatic load means that the subject continually activates the physical factors that accommodate for a chronically stressful environment, and this in turn results in cardiovascular problems and metabolic dysregulation of immunity, thus influencing the state of a person’s health [14].
e218
C.-Y. Chao et al.
Limitations
Conclusions
Although our findings showed a negative association between SES and central obesity, a number of limitations to these analyses must be considered. First, our study design is cross-sectional; therefore, a temporal relation between SES and central obesity cannot be examined. Second, we do not have information regarding other relevant household factors that could confound the observed association, such as information on family size, which could be a potential limitation. Third, the DOHaD theory is one of the important mechanisms to explain the effect between low SES and central obesity, however, the data of the birth weight or weights of the mother were not collected in our study. Besides, a study examined the effects of low birth weight on the components of insulin resistance syndrome in children, the results showed low birth weight might predict the risk of the insulin resistance syndrome and its progression over age in childhood [37]. However, the data of birth weight and HOMA-IR are not collected in our study, therefore we cannot prove similar findings in this study. We did explore the association between SES and metabolic syndrome (MS) which could be viewed as a surrogate of insulin resistance, but there were no statistic significance between SES and MS (data not shown). This might explain lack of significant effect of insulin resistance between SES to central obesity in our study subjects, though we did not accurately measure this variable. Fourth, Families with children could be enrolled in the university were likely with higher SES in the past several decades in Taiwan. However, in recent decades, multiple systems for universities entrance have been successfully applied and there were also many newly setup universities in Taiwan, thus, the acceptance rate of the university entrance examination have exceeded 90% in recently years. Therefore, the inference of this study result might not suffer from much bias due to this minimal difference in terms of university entrance barriers. Moreover, self-reported parental SES in childhood might suffer from a recall bias, however, this would result in a non-differentiated misclassification of SES which might only attenuate the relationship between parental SES and central obesity. Although our study had the above referenced study limitations, the strengths of our study include the availability of several indicators of SES and inclusion of potentially confounding variables in the multivariate analysis. Our result also showed parental SES and central obesity had a close relationship even in this highly homogenous population.
In conclusion, our results showed that the low SES is associated with the risk of central obesity; therefore, students with lower SES might be an important target group for prevention and management of obesity in school health promotion programs.
Conflict of interest All the authors declare no conflict of interest.
References [1] Wang Y, Monteiro C, Popkin BM. Trends of obesity and underweight in older children and adolescents in the United States, Brazil, China, and Russia. Am J Clin Nutr 2002;75:971—7. [2] World Health Organization. Obesity: preventing and managing the global epidemic. Geneva: WHO; 2000 [WHO Technical Report Series No. 894]. [3] Mackenbach JP, Stirbu I, Roskam AJR, Schaap MM, Menvielle G, Leinsalu M, et al. Socioeconomic inequalities in health in 22 European countries. N Engl J Med 2008;358:2468—81. [4] Marmot M. Social determinants of health inequalities. Lancet 2005;365:1099—104. [5] Weinsier RL, Hunter GR, Heini AF, Goran MI, Sell SM. The etiology of obesity: relative contribution of metabolic factors, diet, and physical activity. Am J Med 1998;105:145—50. [6] Robertson AR, Brunner E, Shilham A. Food is a political issue. In: Marmot MWR, editor. Social determinants of health. Oxford: Oxford University Press; 1999. p. 179—208. [7] Kivimaki M, Smith GD, Juonala M, Ferrie JE, KeltikangasJarvinen L, Elovainio M, et al. Socioeconomic position in childhood and adult cardiovascular risk factors, vascular structure, and function: cardiovascular risk in young Finns study. Heart 2006;92:474—80. [8] Ebrahim S, Montaner D, Lawlor DA. Clustering of risk factors and social class in childhood and adulthood in British women’s heart and health study: cross sectional analysis. Br Med J 2004;328:861—5. [9] Lawlor DA, Ebrahim S, Smith GD. Socioeconomic position in childhood and adulthood and insulin resistance: cross sectional survey using data from British women’s heart and health study. Br Med J 2002;325:805—10. [10] Parson TJ, Power C, Logan S, Summerbell CD. Childhood predictors of adult obesity: a systemic review. Int J Obes 1999;23(Suppl. 8):S1—107. [11] Ball K, Crawford D. Socioeconomic status and weight change in adults: a review. Soc Sci Med 2005;60:1987—2010. [12] Ball K, Mishra GD. Whose socioeconomic status influences a woman’s obesity risk: her mother’s, her father’s, or her own? Int J Epidemiol 2006;35:131—8. [13] Pischon T, Boeing H, Hoffmann K, Bergmann M, Schulze MB, Overvad K, et al. General and abdominal adiposity and risk of death in Europe. N Engl J Med 2008;359:2105—20. [14] Goodman E. The role of socioeconomic status gradients in explaining differences in US adolescents’ health. Am J Public Health 1999;89:1522—8. [15] Pulkki L, Keltikangas-Jarvinen L, Ravaja N, Viikari J. Childrearing attitudes and cardiovascular risk among children:
Low socieconomic status and central obesity
[16]
[17]
[18]
[19] [20] [21]
[22]
[23]
[24]
[25]
[26]
moderating influence of parental socioeconomic status. Prev Med 2003;36:55—63. Potter BK, Speechley KN, Gutmanis IA, Campbell MK, Koval JJ, Manuel D. A comparison of measures of socioeconomic status for adolescents in a Canadian national health survey. Chronic Dis Can 2005;26:80—9. Marmot M, Shipley M, Brunner E, Hemingway H. Relative contribution of early life and adult socioeconomic factors to adult morbidity in the Whitehall II study. J Epidemiol Community Health 2001;55:301—7. Lucove JC, Kaufman JS, James SA. Association between adult and childhood socioeconomic status and prevalence of the metabolic syndrome in African Americans: the Pitt County study. Am J Public Health 2007;97:234—6. Sobal J, Stunkard AJ. Socioeconomic status and obesity: a review of literature. Psychol Bull 1989;105:260—75. ACSM. ACSM’s guideline for exercise testing and prescription. 7th ed. USA: Lippincott Williams and Wilkins; 2006. Report on the survey of family income and expenditure. Republic of China: Directorate-General of Budget, Accounting and Statistics, Executive Yuan; 2009. Kaluski DN, Keinan-Boker L, Stern F, Green MS, Leventhal A, Goldsmith R, et al. BMI may overestimate the prevalence of obesity among women of lower socioeconomic status. Obesity 2007;15:1808—15. Chen E, Matthews KA, Boyce WT. Socioeconomic differences in children’s health: how and why do these relationship change with age? Psychol Bull 2002;128:295—329. Lawlor DA, Smith GD, Ebrahim S. Association between childhood socioeconomic status and coronary heart disease risk among postmenopausal women: findings from the British Women’s Heart and Health Study. Am J Public Health 2004;94:1386—92. Leino M, Porkka KVK, Raitakari O, Laitinem S, Taimela S, Viikari JSA. Influence of parental occupation on coronary heart disease risk factors in children: the Cardiovascular Risk in Young Finns Study. Int J Epidemiol 1996;25: 1189—95. Liberations P, Link BG, Kelsey JL. The measurement of social class in epidemiology. Epidemiol Rev 1988;10:87— 121.
e219 [27] Carnethon MR, Loria CM, Hill JO, Sidney S, Savage PJ, Liu K. Risk factors for the metabolic syndrome: the Coronary Artery Risk Development in Young Adults (CARDIA) Study, 1985—2001. Diabetes Care 2004;27:2707—15. [28] Park YW, Zhu S, Palaniappan L, Heshka S, Carnethon MR, Heymsfield SB. The metabolic syndrome: prevalence and associated risk factor findings in the US population from the Third National Health and Nutrition Examination Survey, 1988—1994. Arch Intern Med 2003;163:427—36. [29] Chen W, Xu J, Srinivasan SR, Berenson GS, Li S. Metabolic syndrome variables at low levels in childhood are beneficially associated with adulthood cardiovascular risk. The Bogalusa Heart Study. Diabetes Care 2005;28:138—43. [30] Ozaki R, Qiao Q, Wong GWK, Chan MHM, So WY, Tong PCY, et al. Overweight, family history of diabetes and attending schools of lower academic grading are independent predictors for metabolic syndrome in Hong Kong Chinese adolescents. Arch Dis Child 2007;92:224—8. [31] Power C, Hertzman C. Social and biological pathways linking early life and adult disease. Br Med Bull 1997;53:210—21. [32] Barker DJP, Osmond C, Simmonds SJ, Wield GA. The relation of small head circumference and thinness at birth to death from cardiovascular disease in adult life. Br Med J 1993;306:422—6. [33] Ekelund U, Ong KK, Linne Y, Neovius M, Brage S, Dunger DB, et al. Association of weight gain in infancy and early childhood with metabolic risk in young adults. J Clin Endocrinol Metab 2007;92:98—103. [34] Silveira PP, Portella AK, Goldani MZ, Barbieri MA. Developmental origin of health and disease (DOHaD). J Pediatr 2007;83:494—504. [35] Marmot M, Wilkinson RG. Psychosocial and material pathways in the relation between income and health: a response to Lynch et al. Br Med J 2001;322:1233—6. [36] Raikkonen K, Keltikangas-Jarvinen L, Hautanen A, Adlercreutz H. Neuroendocrine mechanisms in chronic perceived stress: associated with the metabolic syndrome. Endocrinol Metab 1997;4:247—54. [37] Li C, Johnson MS, Goran MI. Effects of low birth weight on insulin resistance syndrome in Caucasian and AfricanAmerican children. Diabetes Care 2001;24:2035—42.
Available online at www.sciencedirect.com
ORIGINAL ARTICLE
Low socioeconomic status may increase the risk of central obesity in incoming university students in Taiwan Chi-Yuan Chao a, Chi-Chen Shih b, Chi-Jen Wang c, Jin-Shang Wu b,d, Feng-Hwa Lu b,d, Chih-Jen Chang b,d, Yi-Ching Yang b,d,∗ a
Department of Family Medicine, Min-Sheng General Hospital, Ching-Kuo Campus, Taoyuan, Taiwan Department of Family Medicine, National Cheng Kung University Hospital, Tainan, Taiwan c Department of Nursing, College of Medicine, National Cheng Kung University, Tainan, Taiwan d Department of Family Medicine, College of Medicine, National Cheng Kung University, Tainan, Taiwan b
Received 23 December 2011 ; received in revised form 16 July 2012; accepted 26 July 2012
KEYWORDS Socioeconomic status; Central obesity
Summary Background: Obesity is related to social disparity. The objective of the study was to evaluate different indicators of parental SES with the association of central obesity in young adult Taiwanese students. Methods: This study was cross-sectionally designed and a total of 4552 subjects were recruited. Each subject completed a self-administrated questionnaire and received anthropometric and laboratory measurements. The indicators of SES in study subjects included parental education, occupation, household incomes, childhood and current index of social position (ISP), measured according to the modified Hollingshead’s ISP. Central obesity was defined as waist circumference ≥90 cm in men and ≥80 cm in women. Results: The prevalence of central obesity was 10.7% in this study. When compared to subjects with normal waist circumferences, subjects with central obesity were older, had a higher BMI, both systolic and diastolic blood pressure, a higher proportion of male gender, family history of diabetes and hypertension, alcohol consumption habit, and a higher proportion of low current household income, current parental blue collar occupational level, and lower current and childhood parental ISP level. Multivariate analysis showed the current parental household income and ISP were significantly higher indicators of risk of central obesity after adjustment for possible confounding factors. The odds ratios were 1.26 and 1.30, respectively.
∗ Corresponding author at: Department of Family Medicine, College of Medicine, National Cheng Kung University, 138, Sheng-Li Rd, Tainan 70403, Taiwan. Tel.: +886 6 2353535x5188; fax: +886 6 2091433. E-mail address: [email protected] (Y.-C. Yang).
1871-403X/$ — see front matter © 2012 Asian Oceanian Association for the Study of Obesity. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.orcp.2012.07.002
Low socieconomic status and central obesity
e213
Conclusions: Our results showed that low household income and current ISP were independently associated with the risk of central obesity. Therefore, young adults with low SES should be an important target group for prevention and management of central obesity in school health promotion programs. © 2012 Asian Oceanian Association for the Study of Obesity. Published by Elsevier Ltd. All rights reserved.
Obesity is a global health problem with high prevalence in both developed and developing countries [1]. Obesity increases the risk of several leading causes of death and diseases in the world [2]. Furthermore, the highly prevalent rate of obesity is not only in middle-aged adults, but also in children and young adults. Several studies have indicated that diseases and causes of death are unequal according to social status [3,4]. One possible reason attributed to the associated higher risk of morbidity and mortality in people with low SES has been lower participation in physical activity and a higher prevalence of obesity [5,6]. Many studies have reported the association between SES and the risk of general obesity, and poor SES in both childhood and adulthood have been implicated in obesity risk among adults [7—9]. However, to date, few studies have attempted to establish the relative importance of childhood [10] and adult SES affecting weight gain and obesity risk in adulthood [11]. One population cohort study showed that the SES in both childhood and adulthood are independently inversely associated with BMI and weight change, but the associations varied by the SES indicator used [12]. Central obesity might have higher predictive value with regard to health problems than BMI [13], but the majority of studies have used BMI as the general obesity index to investigate the association between obesity and SES. Moreover, in these studies, only the father’s education [14—16] or occupational level [17,18,7] was used to explore the relationship between childhood SES and obesity. Furthermore, different SES indicators might affect one’s health via different pathways at different development stages. Therefore, the use of a single SES index to explain the effect of SES on obesity risk is not enough, thus, considering both parents’ SES instead of only the father’s or the individual’s own SES might more accurately reflecting an unbiased association between SES and obesity. In addition, previous studies have also shown that the association between SES and obesity might vary by the economic development of the countries under investigation [1,19]. To our knowledge, there have been few studies investigating the
relationship between SES and central obesity in rapidly developing societies such as Taiwan. Therefore, the objective of this study was to evaluate the association of different parental SES indicators with central obesity in an incoming university student population in Taiwan.
Methods Subjects The baseline data was collected from an entrance health check-up survey in a university in 2007. All 5550 subjects received a health examination and completed a self-administrated structured questionnaire, which included demographic information, medical history, smoking, alcohol and coffee consumption habits, and physical exercise. After excluding subjects with incomplete data, a total of 4552 subjects were recruited for the study (2145 men, 61.8%; 1325 women, 38.2%). Informed consent was obtained from all of the study participants. Since they only agreed to have their questionnaire data and related examination results analyzed anonymously, any identifying information was kept confidential. The Ethical Committee for Human Research at National Cheng Kung University Hospital approved the study protocol.
Instruments Assessment of lifestyle and other factors Smoking habit was classified as current smokers (defined by at least one pack/month) and nonsmokers. Alcohol consumption was classified as drinkers (defined by at least one drink/week) and non-drinkers. Regular physical exercise was defined according to the recommendation of the American College of Sports Medicine guidelines of at least three times weekly vigorous exercise [20] (intense enough to cause sweating and/or heavy breathing/and/or to increase heart rate to a certain
e214 amount). A positive family history of diabetes and hypertension was defined as at least one of the firstdegree relatives having diabetes or hypertension.
Measurement of parental socioeconomic status The different parental SES of the subjects was collected using a self-reported questionnaire. The parental education level was classified as either high (educational years more than 12 years) or low education level groups. The parental occupation level also was classified as white or blue collar groups. The current household income was defined as the total income of the family per month, and it was further dichotomized into two groups, high (household income/month more than NTD 80,000) and low household income. The current household income cutoff value of NTD 80,000 per month, because the ratio between high and low current household income was about 4:6 (42.9%:57.1%) in our study, it approaches to 1:1. In addition, according the report on the survey of family income and expenditure, 2009 by Directorate-General of Budget, Accounting and Statistics, Executive Yuan, Republic of China, the average household income on 2009 was about NTD 887,605 per year [21], it was close to the cutoff value in our study. Parental past SES (at subjects’ third year in elementary school) and current SES were classified according to a modified Hollingshead’s index of social position (ISP), calculated by the sum of education level times 4 and occupation level times 7. The ISP was originally classified into five levels according to the score: Level I: 11—18, Level II: 19—29, Level III: 30—40, Level IV: 41—51, Level V: 52—55. The ISP level was further classified into three groups, low (I), middle (II and III) and high (IV and V) SES level in our study analysis.
Procedure All the anthropometric measurements, blood pressure (BP) and blood sampling were carried out by well-trained nurses. Body weight (to the nearest 0.1 kg) and height (to the nearest 0.1 cm) were measured using a certified machine (DETECTO, Webb City, MO, USA). Waist circumference (WC) was measured from the midway between the lower rib margin and the iliac crest at the end of normal expiration with the subjects standing (to the nearest 0.1 cm). Body mass index (BMI) was calculated as weight in kilogram divided by the square of height in meter. Two readings of BP (systolic and diastolic BP) were measured with subjects in a sitting
C.-Y. Chao et al. position with a blood pressure monitor (DINAMAP@ Pro Series 100, CRITIKON Company L.L.C. Assembled in Mexico) after at least 10 min of rest. Central obesity has been defined as WC 90 cm or more in men and 80 cm or more in women according to World Health Organization (WHO) Asian-Pacific diagnostic criteria. All subjects received blood sampling after a 10-h overnight fast and the tests including liver function, fasting glucose level, and lipid profile were done.
Data analysis All statistical analyses were performed by using the 15th version of SPSS software. The subjects were classified into two groups by central obesity. Clinical characteristics in the study were presented as mean ± SD or percent. Chi-square tests were used to compare the categorical variables between the two groups. In the logistical regression model, the relationship between central obesity and different parental SES was explored after adjustment for age, gender, smoking, alcohol consumption habits, physical exercise, and family history of diabetes. The odds ratio and 95% confidence intervals (CIs) for predictors were derived for each regression model. Statistical significance was defined as a p value less than 0.05.
Results Table 1 showed the clinical characteristics of the 4552 subjects according to the status of central obesity. The prevalence of central obesity was 10.7%, and there were significant differences in gender, age, body mass index, waist circumference, systolic and diastolic BP, cigarette smoking habit, and family history of diabetes and hypertension between normal and central obesity groups. Compared to the normal group, the central obesity group was older, had a higher BMI, both systolic and diastolic BP, a higher proportion of male gender, family history of diabetes and hypertension, alcohol consumption habit. Table 2 shows the comparisons of family SES characteristics between the normal and central obesity groups. It can been seen that there were significant differences in current household income, current parental occupation level, and current and childhood parental ISP level. Compared to the normal group, the central obesity group had a higher proportion of low current household income, current parental blue collar occupational level, and lower current and childhood parental ISP level.
Low socieconomic status and central obesity Table 1
e215
The comparison of clinical characteristics between the subjects according to the status of central obesity.a
Variables
Total (N = 4552)
Normal (n = 4063)
Central obesity (n = 489)
p-Value
Male sex (%) Age (years) Obesity Body Mass Index (BMI: kg/m2 ) Overweight (BMI ≥24) Obesity (BMI ≥27) Waist circumference (cm) Cigarette smoking (%) Alcohol consumption (%) Physical exercise (≥3 times/week)(%) Family history of diabetes (%) Family history of hypertension (%)
67.1 21.9 (4.6)
65.2 21.7 (4.3)
82.8 23.6 (6.4)
<0.0001*** <0.0001***
22.2 (3.6) 15.7 9.3 74.8 (10.6) 4.7 2.7 22.1 9.5 22.5
21.4 (2.6) 14.5 1.9 72.4 (7.9) 4.2 2.6 22.1 8.9 21.4
28.9 (3.5) 25.8 70.6 94.9 (8.0) 9.0 4.1 22.3 14.5 31.5
<0.0001*** <0.0001*** <0.0001*** <0.0001*** <0.0001*** 0.065 0.924 <0.0001*** <0.0001***
a * ** ***
Data are presented as mean ± SD or percent. p < 0.05. p < 0.01. p < 0.001.
The odd ratios and 95% CIs for the different SES indicators and other independent variables for predicting central obesity from six multivariate logistic regression models are shown in Table 3. The current parental household income and ISP were associated with a significantly higher risk of central obesity, even after adjustment for age, sex, cigarette smoking, alcohol consumption, physical exercise, and a family history of diabetes and hypertension. The odds ratio was 1.26 (95% CIs: 1.03—1.54) and 1.30 (95% CIs: 1.04—1.64) respectively.
Discussion Our study showed that household income and current ISP were associated with central obesity in an incoming university incoming student population, after adjustment for the potential confounding factors of lifestyle and family history of diabetes and hypertension. Thus, lower household income and current ISP independently increased the risk of central obesity in a young adult population in Taiwan.
Table 2 The comparison of different socioeconomic status between the subjects according to the status of central obesity.a Variables
Total (N = 4552)
Normal (n = 4063)
Central obesity (n = 489)
p-Value
Parental education: ≤12 years (%) Current household income: ≤80,000 NTD (%) Current parental occupation: blue collar (%) Parental occupation in childhood: blue collar (%) Current parental ISP I (%) II—III (%) IV—V (%) Parental ISP in childhood I (%) II—III (%) IV—V (%)
17.2 57.1 47.8 44.9
16.9 56.4 47.2 44.4
19.6 63.4 53.2 48.9
0.141 0.003** 0.013* 0.062
10.6 60.2 29.2
10.2 59.9 29.9
13.3 63.0 23.7
0.006**
9.7 60.5 29.6
8.5 60.3 30.2
12.7 62.2 25.2
0.017*
NTD, new Taiwan dollar; ISP, index of social position. a Data are presented as percent. * p < 0.05. ** p < 0.01. *** p < 0.001.
e216
Table 3
The multiple logistic regressions for the associations of different SES and central obesity.a
Variables
Model 1
Model 2
Model 3
Model 4
Model 5
Model 6
Parental education (<12 vs ≥12 years) Household income (<80,000 vs ≥80,000 NTD/month) Current parental occupation (blue vs. white collar) Childhood parental occupation (blue vs. white collar) Current ISP I vs. IV—V II—III vs. IV—V Childhood ISP I vs. IV—V II—III vs. IV—V Age (≥20 vs. <20 years) Gender (male vs. female) Cigarette smoking (yes vs. no) Alcohol consumption (yes vs. no) Physical exercise (yes vs. no) Family history of hypertension (yes vs. no) Family history of diabetes (yes vs. no)
0.98 (0.76—1.25)b
—
—
—
—
—
—
1.26 (1.03—1.54)
—
—
—
—
*
1.14 (0.94—1.39) 1.10 (0.91—1.34)
1.30 (1.04—1.64)* 1.24 (0.88—1.73)
1.37 (1.11—1.68)** 2.55 (1.98—3.27)*** 1.59 (1.11—2.29)*
1.33 (1.08—1.64)** 2.53 (1.97—3.24)*** 1.60 (1.11—2.30)*
1.34 (1.09—1.64)** 2.53 (1.97—3.24)*** 1.58 (1.10—2.27)**
1.35 (1.10—1.66)** 2.53 (1.97—3.25)*** 1.57 (1.10—2.62)*
1.34 (1.09—1.65)** 2.53 (1.97—3.25)*** 1.59 (1.10—2028)*
1.21 1.24 1.34 2.53 1.58
(0.97—1.52) (0.88—1.74) (1.09—1065)** (1.97—3.25)*** (1.10—2.27)*
1.17 (0.71—1.95)
1.21 (0.73—2.01)
1.19 (0.72—1.98)
1.19 (0.72—1.98)
1.19 (0.72—1.98)
1.20 (0.72—1.99)
0.96 (0.76—1.21) 1.63 (1.32—2.02)***
0.97 (0.77—1.23) 1.63 (1.32—2.02)***
0.97 (0.77—1.22) 1.63 (1.32—2.02)***
0.97 (0.76—1.22) 1.64 (1.32—2.02)***
0.97 (0.77—1.23) 1.64 (1.33—2.03)***
0.97 (0.77—1.23) 1.64 (1.33—2.03)***
1.53 (1.16—2.04)**
1.50 (1.13—1.99)**
1.51 (1.14—2.01)**
1.53 (1.15—2.02)**
1.52 (1.14—2.02)**
1.52 (1014—2.02)**
C.-Y. Chao et al.
NTD, new Taiwan dollar. a Dependant variable: central obesity; independent variables: sex, age, cigarette smoking, alcohol consumption, physical exercise, family history of diabetes and hypertension, socioeconomic status (model 1: parental education, model 2: household income, model 3: current parental occupation, model 4: childhood parental occupation, model 5: current ISP, model 6: childhood ISP). b Data are presented as odds ratio (95% confidence intervals). * p < 0.05. ** p < 0.01. *** p < 0.001.
Low socieconomic status and central obesity After adjusting for sex and age, the subjects in our study with lower SES were found to be shorter and heavier than those with higher SES. Our result was consisted with previous study [22] and implied that if we used BMI for evaluate the prevalence of obesity among the low SES group, we might have overestimated the prevalence of obesity, because of the effects of both increased numerator (body weight) and decreased denominator (body height). Therefore, in our study we used the waist circumference as an alternative measures for classifying obesity in order to avoid the effect of height. Previous studies have shown lower childhood parental SES to be associated with adulthood central obesity [7—9], high blood pressure [7,8,23], dyslipidemia [7,8,24], and insulin resistance [7,9]. Goodman showed parental education level and household income were independently inversely related to young adult obesity, while in their study, parental occupation level was not associated with obesity after adjusting for other social and economic factors (members of family, race and gender, etc.) [14]. Leino et al. reported parental occupation level to be associated with birth weight, childhood obesity, smoking and physical inactivity in a study on young adults in Finland. Furthermore, this association was found to persist to the stage of young adult [25]. Our study used s modified Hollingshead’s index of social position, taking into account parental education and occupation levels simultaneously, and it showed the lowest current ISP to be independently associated with the risk of central obesity. Household income might reflect on the ability to buy healthy food, to have good housing, good working conditions, and access to medical care resources [26]. Other variables have had a positive association with central obesity in multivariate analysis, including older age [27,28], male gender [27], smoking habit [28], a habit of regular physical activity [27,28], and a family history of hypertension [29] and diabetes [28,30]. These results were compatible with previous findings. There are several mechanisms used to explain how the parental SES influences the health of adulthood, such as the latency model, the pathway model and psychosocial stress [24,31]. The latency model emphasized that the health status of the fetus and the subjects during early childhood (such as low birth weight, short status, smaller head circumference and rapid body weight gain during childhood) are associated with the adulthood cardiovascular disease and metabolic
e217 abnormality because these conditions are related to poor maternal nutrition and household poverty [32,33]. The current hypothesis of the developmental origin of health and diseases (DOHaD) has raised the similar concept of the latency model. This new model of causality and of the mechanisms involved in the emergence and development of chronic diseases [34]. Poverty is associated with an unhealthy or insufficient diet, an unhealthy living environment, poor health care quality, crowded residential environments, and low SES [24,25,35]; these factors might influence the uterine environment of a fetus and early childhood behavior, thus, increasing exposure to risk factors and resulting in unhealthy lifestyles and diseases during adulthood. The pathway model suggests that early stage SES might influence adult health because the social and biological risk factors related to health will accumulate during different life stages [31]. Chris et al. showed that seven different social and biological risk factors during different stage of human development (birth weight, maternal health condition during childhood, divorce of parents, body height of adult, education level, smoking habit, and autonomy or control in the workplace) have a different magnitude of association with SES at birth, and thus subjects with low SES are more likely to have more risk factors [31]. The psychosocial stress model suggested that poverty during childhood might result in psychosocial stressors related to mood, social identity, and dignity. The chronic psychosocial stress associated with poverty will in turn, according to this theory, result in cardiovascular disease and metabolic dysregulation in adulthood through the hypothalamus-pituitary-adrenal axis [14,15,36]. Therefore, childhood parental SES might influence adult health through different mechanisms, during different stages of development. Our study showed that the lower the childhood SES, the higher the adulthood obesity risk, but the results did not reach statistical significance. Ebrahim et al. showed that the effect of the childhood SES as an explanation of the adulthood metabolic abnormality to be attenuated after adjustment for adulthood SES, because its effect on health might decrease through the effect of current economic or social factors [8], and the allostatic load model could explain this effect. Allostatic load means that the subject continually activates the physical factors that accommodate for a chronically stressful environment, and this in turn results in cardiovascular problems and metabolic dysregulation of immunity, thus influencing the state of a person’s health [14].
e218
C.-Y. Chao et al.
Limitations
Conclusions
Although our findings showed a negative association between SES and central obesity, a number of limitations to these analyses must be considered. First, our study design is cross-sectional; therefore, a temporal relation between SES and central obesity cannot be examined. Second, we do not have information regarding other relevant household factors that could confound the observed association, such as information on family size, which could be a potential limitation. Third, the DOHaD theory is one of the important mechanisms to explain the effect between low SES and central obesity, however, the data of the birth weight or weights of the mother were not collected in our study. Besides, a study examined the effects of low birth weight on the components of insulin resistance syndrome in children, the results showed low birth weight might predict the risk of the insulin resistance syndrome and its progression over age in childhood [37]. However, the data of birth weight and HOMA-IR are not collected in our study, therefore we cannot prove similar findings in this study. We did explore the association between SES and metabolic syndrome (MS) which could be viewed as a surrogate of insulin resistance, but there were no statistic significance between SES and MS (data not shown). This might explain lack of significant effect of insulin resistance between SES to central obesity in our study subjects, though we did not accurately measure this variable. Fourth, Families with children could be enrolled in the university were likely with higher SES in the past several decades in Taiwan. However, in recent decades, multiple systems for universities entrance have been successfully applied and there were also many newly setup universities in Taiwan, thus, the acceptance rate of the university entrance examination have exceeded 90% in recently years. Therefore, the inference of this study result might not suffer from much bias due to this minimal difference in terms of university entrance barriers. Moreover, self-reported parental SES in childhood might suffer from a recall bias, however, this would result in a non-differentiated misclassification of SES which might only attenuate the relationship between parental SES and central obesity. Although our study had the above referenced study limitations, the strengths of our study include the availability of several indicators of SES and inclusion of potentially confounding variables in the multivariate analysis. Our result also showed parental SES and central obesity had a close relationship even in this highly homogenous population.
In conclusion, our results showed that the low SES is associated with the risk of central obesity; therefore, students with lower SES might be an important target group for prevention and management of obesity in school health promotion programs.
Conflict of interest All the authors declare no conflict of interest.
References [1] Wang Y, Monteiro C, Popkin BM. Trends of obesity and underweight in older children and adolescents in the United States, Brazil, China, and Russia. Am J Clin Nutr 2002;75:971—7. [2] World Health Organization. Obesity: preventing and managing the global epidemic. Geneva: WHO; 2000 [WHO Technical Report Series No. 894]. [3] Mackenbach JP, Stirbu I, Roskam AJR, Schaap MM, Menvielle G, Leinsalu M, et al. Socioeconomic inequalities in health in 22 European countries. N Engl J Med 2008;358:2468—81. [4] Marmot M. Social determinants of health inequalities. Lancet 2005;365:1099—104. [5] Weinsier RL, Hunter GR, Heini AF, Goran MI, Sell SM. The etiology of obesity: relative contribution of metabolic factors, diet, and physical activity. Am J Med 1998;105:145—50. [6] Robertson AR, Brunner E, Shilham A. Food is a political issue. In: Marmot MWR, editor. Social determinants of health. Oxford: Oxford University Press; 1999. p. 179—208. [7] Kivimaki M, Smith GD, Juonala M, Ferrie JE, KeltikangasJarvinen L, Elovainio M, et al. Socioeconomic position in childhood and adult cardiovascular risk factors, vascular structure, and function: cardiovascular risk in young Finns study. Heart 2006;92:474—80. [8] Ebrahim S, Montaner D, Lawlor DA. Clustering of risk factors and social class in childhood and adulthood in British women’s heart and health study: cross sectional analysis. Br Med J 2004;328:861—5. [9] Lawlor DA, Ebrahim S, Smith GD. Socioeconomic position in childhood and adulthood and insulin resistance: cross sectional survey using data from British women’s heart and health study. Br Med J 2002;325:805—10. [10] Parson TJ, Power C, Logan S, Summerbell CD. Childhood predictors of adult obesity: a systemic review. Int J Obes 1999;23(Suppl. 8):S1—107. [11] Ball K, Crawford D. Socioeconomic status and weight change in adults: a review. Soc Sci Med 2005;60:1987—2010. [12] Ball K, Mishra GD. Whose socioeconomic status influences a woman’s obesity risk: her mother’s, her father’s, or her own? Int J Epidemiol 2006;35:131—8. [13] Pischon T, Boeing H, Hoffmann K, Bergmann M, Schulze MB, Overvad K, et al. General and abdominal adiposity and risk of death in Europe. N Engl J Med 2008;359:2105—20. [14] Goodman E. The role of socioeconomic status gradients in explaining differences in US adolescents’ health. Am J Public Health 1999;89:1522—8. [15] Pulkki L, Keltikangas-Jarvinen L, Ravaja N, Viikari J. Childrearing attitudes and cardiovascular risk among children:
Low socieconomic status and central obesity
[16]
[17]
[18]
[19] [20] [21]
[22]
[23]
[24]
[25]
[26]
moderating influence of parental socioeconomic status. Prev Med 2003;36:55—63. Potter BK, Speechley KN, Gutmanis IA, Campbell MK, Koval JJ, Manuel D. A comparison of measures of socioeconomic status for adolescents in a Canadian national health survey. Chronic Dis Can 2005;26:80—9. Marmot M, Shipley M, Brunner E, Hemingway H. Relative contribution of early life and adult socioeconomic factors to adult morbidity in the Whitehall II study. J Epidemiol Community Health 2001;55:301—7. Lucove JC, Kaufman JS, James SA. Association between adult and childhood socioeconomic status and prevalence of the metabolic syndrome in African Americans: the Pitt County study. Am J Public Health 2007;97:234—6. Sobal J, Stunkard AJ. Socioeconomic status and obesity: a review of literature. Psychol Bull 1989;105:260—75. ACSM. ACSM’s guideline for exercise testing and prescription. 7th ed. USA: Lippincott Williams and Wilkins; 2006. Report on the survey of family income and expenditure. Republic of China: Directorate-General of Budget, Accounting and Statistics, Executive Yuan; 2009. Kaluski DN, Keinan-Boker L, Stern F, Green MS, Leventhal A, Goldsmith R, et al. BMI may overestimate the prevalence of obesity among women of lower socioeconomic status. Obesity 2007;15:1808—15. Chen E, Matthews KA, Boyce WT. Socioeconomic differences in children’s health: how and why do these relationship change with age? Psychol Bull 2002;128:295—329. Lawlor DA, Smith GD, Ebrahim S. Association between childhood socioeconomic status and coronary heart disease risk among postmenopausal women: findings from the British Women’s Heart and Health Study. Am J Public Health 2004;94:1386—92. Leino M, Porkka KVK, Raitakari O, Laitinem S, Taimela S, Viikari JSA. Influence of parental occupation on coronary heart disease risk factors in children: the Cardiovascular Risk in Young Finns Study. Int J Epidemiol 1996;25: 1189—95. Liberations P, Link BG, Kelsey JL. The measurement of social class in epidemiology. Epidemiol Rev 1988;10:87— 121.
e219 [27] Carnethon MR, Loria CM, Hill JO, Sidney S, Savage PJ, Liu K. Risk factors for the metabolic syndrome: the Coronary Artery Risk Development in Young Adults (CARDIA) Study, 1985—2001. Diabetes Care 2004;27:2707—15. [28] Park YW, Zhu S, Palaniappan L, Heshka S, Carnethon MR, Heymsfield SB. The metabolic syndrome: prevalence and associated risk factor findings in the US population from the Third National Health and Nutrition Examination Survey, 1988—1994. Arch Intern Med 2003;163:427—36. [29] Chen W, Xu J, Srinivasan SR, Berenson GS, Li S. Metabolic syndrome variables at low levels in childhood are beneficially associated with adulthood cardiovascular risk. The Bogalusa Heart Study. Diabetes Care 2005;28:138—43. [30] Ozaki R, Qiao Q, Wong GWK, Chan MHM, So WY, Tong PCY, et al. Overweight, family history of diabetes and attending schools of lower academic grading are independent predictors for metabolic syndrome in Hong Kong Chinese adolescents. Arch Dis Child 2007;92:224—8. [31] Power C, Hertzman C. Social and biological pathways linking early life and adult disease. Br Med Bull 1997;53:210—21. [32] Barker DJP, Osmond C, Simmonds SJ, Wield GA. The relation of small head circumference and thinness at birth to death from cardiovascular disease in adult life. Br Med J 1993;306:422—6. [33] Ekelund U, Ong KK, Linne Y, Neovius M, Brage S, Dunger DB, et al. Association of weight gain in infancy and early childhood with metabolic risk in young adults. J Clin Endocrinol Metab 2007;92:98—103. [34] Silveira PP, Portella AK, Goldani MZ, Barbieri MA. Developmental origin of health and disease (DOHaD). J Pediatr 2007;83:494—504. [35] Marmot M, Wilkinson RG. Psychosocial and material pathways in the relation between income and health: a response to Lynch et al. Br Med J 2001;322:1233—6. [36] Raikkonen K, Keltikangas-Jarvinen L, Hautanen A, Adlercreutz H. Neuroendocrine mechanisms in chronic perceived stress: associated with the metabolic syndrome. Endocrinol Metab 1997;4:247—54. [37] Li C, Johnson MS, Goran MI. Effects of low birth weight on insulin resistance syndrome in Caucasian and AfricanAmerican children. Diabetes Care 2001;24:2035—42.
Available online at www.sciencedirect.com