Association of Body Mass Index with Educational Level in Iranian Men and Women
Total Page:16
File Type:pdf, Size:1020Kb
European Journal of Clinical Nutrition (2003) 57, 819–823 & 2003 Nature Publishing Group All rights reserved 0954-3007/03 $25.00 www.nature.com/ejcn ORIGINAL COMMUNICATION Association of body mass index with educational level in Iranian men and women M Maddah1*, MR Eshraghian2, A Djazayery3 and R Mirdamadi4 1Department of Human Nutrition, School of Public Health, Guilan University of Medical Sciences, Rasht, Iran; 2Department of Human Nutrition and Biochemistry; 3Department of Epidemiology and Biostatistics, School of Public Health, Tehran University of Medical Sciences, Tehran, Iran; and 4Department of Obstetrics and Midwifery, School of Medicine, Tarbyat Modarress University, Tehran, Iran Objective: Investigation of the relationship between educational level, body mass index (BMI), waist-to-hip ratio (WHR), physical activity and parity in a group of Iranian men and women living in Tehran. Design: A cross-sectional study in a group of Iranian men and women. The subjects were classified into two educational levels: low education (r12 y schooling) and high education (412 y schooling); and BMI, WHR, physical activity and parity (in women) were compared in two groups in men and women, separately. Setting: Metabolic Unit of Tehran University of Medical Sciences in Tehran. Subjects: Three hundred and fifteen men aged 33.1 (22–46) and 403 women aged 27.9 (22–45). Results: After controlling for age and smoking, women with a higher level of education showed a significantly lower mean BMI than less educated women (24.8 7 4.2 vs 28.3 7 4.9, Po0.01), while more educated men had a higher mean BMI than less educated men (28.4 7 4.3 vs 26.7 7 4.5). In multiple regression analysis, physical activity in leisure time in men and years of education in women were the only determinants of BMI. After controlling for BMI, WHR was not related to the level of education in either men or women. Conclusion: The present data indicated an educational difference in BMI for the study population. In Iranian women, like the women in developed countries, the level of education was negatively related to BMI, while in men the association was positive. Sponsorship: This work was financially supported by the Institute of Public Health, Tehran University of Medical Sciences. European Journal of Clinical Nutrition (2003) 57, 819–823. doi:10.1038/sj.ejcn.1601615 Keywords: BMI; WHR; education; social class; Iran; developing countries 1. Introduction since Islamic revolution in 1979. Islamic rules are reflected in Cross-sectional studies have shown that BMI differs by many social aspects of life in Iran and they might interact educational levels. While a consistent inverse association with obesity-related behaviors in the population. Iranian between education and body mass index (BMI) was found women, for example, are dressed in uniform loose garments especially in women in developed countries (Garn et al, to cover their hair and body in public so that men do not see 1989; Tavani et al, 1994; Rosmond & Bjorntorp, 1999; their hair and body shape. They are also less engaged in Stam-Moraga et al, 1999; Van Lenthe et al, 2000), there is outdoor leisure activities for religious reasons. Cosmetic- and less published information from developing countries. body-shape-related fashionable advertisements and media The Islamic Republic of Iran is a non-Arabic Islamic images have also been banned since Islamic revolution in country and has been governed by Islamic-based regulations 1979. Although there are no data to suggest what Iranian men and women might think about ideal body weight, there is *Correspondence: M Maddah, Building 97-1, St 97 Golsar 41649 Rasht, Iran. less social pressure for conforming to an ideal body image, E-mail: [email protected] especially for women, in Iran. Considering that the present Guarantor: M Maddah social context in Iran is very different from those of Western Contributors F :MM concept of work and writing of the paper. countries, we examined whether body weight and fat MREFhelped in data analysis. AD and RMFcollected data. Received 26 February 2002; revised 3 June 2002; revised 14 July 2002; distribution are related to educational level in Iranian men accepted 13 August 2002 and women. BMI and educational level in Iranians M Maddah et al 820 Subjects and methods variables was examined using ANCOVA controlling for age The participants in the study were 315 men (age: 22–46 y; and smoking. Multiple regression analyses were also per- BMI: 18.5–33.6 kg/m2) and 403 women (age: 22–45 y; BMI: formed with BMI as the dependent variable and education 18.1–35.3 kg/m2. (WHO standard for BMI: underweight: and other measured factors as independent variables. Values BMIo18.5 kg/m2; normal: BMI 18.5–24.9 kg/m2; overweight: were given as the mean 7 standard deviation. A P-value of BMI 25.0–29.9 kg/m2; obese: BMI30.0 kg/m2 (World Health less than 0.05 was considered as the level of significance. Organization, 1997)). The subjects were selected from those Analyses were performed using the Statistical Package for who either attended the Metabolic Unit of Tehran University Social Science (SPSS 10.01 for Windows, SPSS Incs Head- of Medical Sciences in Tehran for a routine health check or quarters, Chicago, USA). who were acting as control subjects for a sex steroid study. Fifty-nine percent of the participants were originally from different parts of Iran who were living in Tehran. The Results subjects were all married and had no metabolic illness, Characteristics of the subjects are given in Table 1. Results of judged by blood and urine analyses, and none were on diet the analysis of covariance showed that after controlling for or medicine. The subjects gave written consent for participa- age and smoking, women of higher educational level had tion in this study. lower mean BMI and WHR than less educated women. In contrast, more educated men had a significantly higher mean BMI and WHR than less educated men (Table 2). After Measures controlling for the differences in BMI however, WHR was not The subjects were interviewed to collect information about related to the level of education in either men or women. their years of schooling, marital status, parity, smoking habit Multiple regression analyses were performed with BMI as and physical activity. the dependent variable and years of education, age, work, Body weight was measured to the nearest 0.1 kg using a sport and leisure time activity scores as the independent balanced beam scale wearing light clothing without shoes; variables in the models. Leisure time activity in men, and height was measured to the nearest 0.5 cm under the same years of education in women remained a significant conditions. Waist circumference was taken to the midpoint predictor of BMI when other factors described above were between the ribs and iliac crests, and hip circumference was taken into account (Table 3). measured at the level of greater posterior protuberance of the The interaction between parity, years of education, and buttocks. Waist-to-hip ratio was calculated as an index of BMI in women is shown in Figure 1. Women with the least central obesity (Kooy & Seidell, 1993). parity and the highest level of education were at the lowest The public education system in Iran consists of three risk of being overweight. levels: 5 y of primary education, 3 y of middle education and 4 y of higher education in high school. Students can then enter universities through a competitive entrance exam. In Discussion the present study, educational levels were expressed as actual We aimed to study the relationship between education, BMI, years of schooling. Data analyses included two educational WHR, physical activity and parity in a group of Iranian men groups: high school or less (1–12 y), and beyond high school (412 y). The two educational groups can be representative of Table 1 Characteristics of the subjects high and low social class in Iran. Physical activity was measured using a 16-item exercise Men n=315 Women n=403 questionnaire (Baecke et al, 1982). There was a piloting of the questionnaire in a group of the study population and no Age (y) 33.1 7 5.2 27.9 7 8.7 modification was required except for translation to the local BMI (kg/m2) 27.2 7 4.5 27.4 7 5.0 7 7 language (Persian). Six of the items were related to leisure WHR 0.93 0.08 0.87 0.08 Education (y) 9.14 7 4.34 10.3 7 4.2 time physical activity. Physical activity at work, leisure time and sport were scored according to frequencies defined on a Physical activity per week basis, ranging from 0 to 5. Then, the sums of all 7 7 activity products were computed. The test reliability of the Leisure index 2.18 0.59 0.36 0.07 Work index 2.66 7 0.92 2.84 7 0.62 indices of physical activity, measured with an interval of Sport index 0.41 7 0.31 0.11 7 0.09 1 month, was 0.85, 0.88 and 0.76 for work index, sport index and leisure time index, respectively. Smoking status Non-smoker 209(66.1%) 413(99.6%) Current and ex-smokera 107(33.9%) 2(0.44%) Statistics Parity F 0.85 7 1.36 Separate analyses were performed for men and women. The relationship between educational levels and the dependent aEx-smokers were defined as those who quitted smoking for at least two years. European Journal of Clinical Nutrition BMI and educational level in Iranians M Maddah et al 821 Table 2 Comparison of measured variables in two levels of education in men and women adjusted for age and smoking status (mean 7 s.d.) Men Women More educated Less educated More educated Less educated n=86 n=229 n=109 n=294 BMI (kg/m2) 28.4 7 4.3** 26.7 7 4.5 24.8 7 4.2** 28.3 7 4.9 WHR 0.94 7 0.06* 0.93 7 0.09 0.83 7 0.06** 0.88 7 0.08 Leisure index 2.24 7 0.59** 2.17 7 0.51 0.91 7 0.64** 0.61 7 0.52 Work index 2.30 7 0.69** 2.74 7 0.94 2.13 7 0.81 2.21 7 0.98 Sport index 0.91 7 0.41** 0.75 7 0.32 0.42 7 0.35* 0.39 7 0.20 Parity FF0.19 7 0.59** 0.59 7 1.02 *Po0.05; **Po0.01.