Measuring adolescents’ weight socioeconomic gradient using parental socioeconomic position

Measuring adolescents’ weight socioeconomic gradient using parental socioeconomic position Abstract Background There is an evidence of social inequalities in weight status in adolescence but the diversity of family socioeconomic status (SES) indicators can lead to discrepant findings. We aimed to identify how combination of family SES indicators can help measuring weight socioeconomic gradient (WSG) among adolescents. Methods Cross-sectional data from 2113 adolescents (13–18 years old) of the PRALIMAP-INÈS trial were used. Multiple SES indicators and assessment of weight status including body mass index (BMI), waist circumference (WC), waist-to-height ratio (WHtR) and self-perception of overweight were used. We used principal component analysis (PCA) followed by structural equation models to identify SES dimensions. A dimension normalized score was calculated ranging from 1 to 10 (a high score corresponding to high SES). Linear regression models (linear trend test) were used to assess the WSG. Results Three SES dimensions were identified: (i) ‘Family social status’, (ii) ‘Family education level’ and (iii) ‘Family income level’. BMI was significantly lower in highly advantaged compared with highly less advantaged [−1.64 (−2.39; −0.89) for family social dimension, −0.86 (−1.37; −0.36) for family education level and −2.35 (−3.65; −1.05) for family income level]. Similar results were observed for all weight indicators excepted for self-perception of overweight status. Socially less advantaged adolescents perceived themselves less fat than they were. Conclusion Although WSG was evident in adolescence, association between SES and weight status differed according to objective or perceived weight indicators. The proposed SES dimension can be applied in other field and future studies are needed to confirm our findings. Introduction Over the last decades, the prevalence of overweight and obesity among adolescents has increased worldwide, both in developed and developing countries, but at an uneven pace.1 In France, the prevalence of overweight and obesity was estimated at 17% including 4% of obesity among adolescents.2 Being overweight in adolescence increases the likelihood of being obese in adulthood with several undesirable health consequences.3,4 Weight status can be assessed by using various anthropometric measurements that are built from combinations of weight, height and waist circumference (WC). The most commonly used measurements are body mass index (BMI) and WC. Other indicators include waist-to-height ratio (WHtR), a measure of the distribution of body fat. In addition to the diversity of weight indicators, defining weight status in adolescence is difficult because of growth during this period. Several cut-offs defined by gender and age account for physiological variations and use national and international growth curves: the International Obesity Task Force (IOTF) thresholds for BMI5 and high WC according to McCarthy cut-off values.6 Social inequalities in health are defined as differences in health according to the socioeconomic status (SES) that are unacceptable and potentially avoidable but, in addition, are also considered unfair.7,8 Social inequalities in health is characterized by differences in indicators of mortality and morbidity according to the SES.9 France is one of the European countries with the highest level of social inequalities in health.10,11 Social inequalities in health are found, not just at the threshold of poverty, but at every level of SES hierarchy and is called social gradient (SG) in health.12,13 The SG in health can be defined as health gradual differences found between all rungs of socioeconomic strata (each socioeconomic strata has higher levels of mortality and morbidity than the one above it).12 Several studies have shown a potential SG in adolescent weight status characterized by a variation in weight depending on adolescent SES.11–13 However, adolescents’ SES is a multifactorial concept which can integrate both adolescents’ perception of their own SES and parental objective SES indicators (education level, socio-professional category, income level, etc.). According to the French Public Health Council, adolescents tend to perceive their SES as equivalent to that of their parents.14 The large number of social indicators available may lead to discrepant results according to the indicator which is used. Each of them may provide different information, but their combination may help to better estimate of adolescents’ SES. Therefore, we aimed to identify how combination of family SES indicators can help measuring weight socioeconomic gradient (WSG) among adolescents. Methods PRALIMAP-INÈS trial The PRALIMAP-INÈS trial is a mixed, prospective, multicentre trial that aimed to evaluate the effectiveness of a school-based intervention to address social inequalities in adolescent overweight. It included overweight and obese adolescents from 13 to 18 years old who were attending state-run middle (grade 9) or high (grade 10) schools. A total of 35 schools (21 high schools and 14 middle schools) participated. Adolescents were recruited during an inclusion session (T0) organized in each school over three waves: 2012–13, 2013–14 and 2014–15 academic years. Adolescents were first measured (weight, height and WC) and when eligible (overweight: BMI greater than the IOTF age and gender specific overweight threshold5 for age and gender and/or WC greater than the McCarthy age and gender specific cut-off;15 close to overweight: BMI greater than the IOTF age and gender specific overweight threshold5 −1 kg/m2), they were invited to complete questionnaires and have a medical visit. If the overweight was confirmed, the physician proposed them to be included in the PRALIMAP-INÈS care program. More details on the PRALIMAP-INÈS trial protocol are given elsewhere.16 The trial was approved by the French consultative committee for treatment of information in health research (no. 12.299), the French National Commission for Data Protection and Liberties (no. 912 372) and the French Persons Protection Committee (no. 2012/15) and was registered at ClinicalTrials.gov (NCT01 688 453). Study sample This study sample included eligible (overweight or close to overweight) adolescents who completed the PRALIMAP-INÈS inclusion questionnaires. Among the 8735 students with available anthropometric measurements, 2283 were eligible for the PRALIMAP-INÈS trial: adolescents between 13 and 18 years old, with BMI greater than the IOTF overweight threshold5 reduced by 1 kg/m2 for age and gender and/or with WC greater than the McCarthy cut-off value for age and gender.15 A total of 2113 eligible adolescents completed the questionnaires and constituted our study sample. Measurements This study used cross-sectional data collected over the three waves of the main study. Anthropometric measures: Weight (kg), height (cm) and WC (cm) were measured twice by a school and a clinical research nurses and used to calculate BMI (weight in kg divided by the square of height in m2) and WHtR (WC in cm divided by height in cm).The weight excess was determined by overweight/obesity prevalence according to the IOTF cut-off,5 high WC according to McCarthy,6 high WHtR defined as WHtR ≥ 0.517,18 and adolescents self-perception of their weight status using the Sorensen silhouettes.19 Socioeconomic measures: Age, gender, school type, family status (living with no, one or two parents), perception of family financial level, scholarship holder and parents (father and mother) education level, social and professional classes and employment status. These data were collected both from the adolescents (self-reported) and the Academic Board of Education. A specific questionnaire on measuring a score of financial affluence of the family (FAS: family affluence scale) was administrated.20,21 The score ranges from 0 (less affluent) to 9 (highest affluence). All these variables were ordinal and reconstructed in ascending order from the less advantaged to the more advantaged. Statistical analysis Socioeconomic and anthropometric characteristics were described as mean (SD) for continuous variables or N (%) for categorical variables. Statistically significant and linear association (social gradient) between each anthropometric and each SES variables was tested using bivariate linear or logistic regression. Significant SES variables were selected for principal component analysis (PCA) followed by structural equation modelling (SEM) to identify adolescents’ SES dimensions. The PCA was used for dimensions identification and the SEM for confirmation. The PCA used the varimax rotation (orthogonal transformation) method and dimension with eigen-value ≥1 were retained.22,23 The SEM was used for confirmation (goodness of fit) and scoring of the identified dimensions. The root mean square error approximation (RMSEA, acceptable fit if <0.08), the comparative fit index (CFI, acceptable fit if ≥0.90) and the normed fit index (NFI, acceptable fit if ≥0.90) were examined to evaluated the fit of the models.24 For each confirmed SES dimension, a normalized score was calculated using SEM path coefficients. To test the internal validity of the identified SES dimensions, the same analysis (PCA followed by SEM) was repeated in a random 50% sample of the initial study sample. We repeated a sub-sample analysis (PCA and SEM) with a random 50% sample of the initial sample. The agreement of SES dimensions was tested with the Cramer V statistic (good agreement corresponding to Cramer V ≥ 0.80).25,26 For each dimension, a normalized score was calculated ranging from 1 (lowest SES) to 10 (highest SES). Each score was then categorized into five classes in order to assess the weight social gradient: highly less advantaged (1–2), less advantaged (3–4), intermediate (5–6), advantaged (7–8) and highly advantaged (9–10). Weight social gradient was assessed as a linear association between each dimension categories and weight indicators using linear regression models adjusted for sex and age. For these analyses, social classes were used as categorical variable to measure differences between social classes and ‘Highly less advantaged’ adolescents constituted the reference group. Then the same analyses were repeated using social classes as continuous variable to obtain the P values for linear trend test. Statistical analysis involved use of SAS 9.3 (SAS Inst., Cary, NC, USA). P < 0.05 was considered statistically significant. Results Characteristics of the study sample Socioeconomic and anthropometric characteristics are presented in table 1. Among the 2113 adolescents, the mean age was 15.3 ± 0.7 years, 56.2% were girls, 48% attending general high schools and the mean affluence FAS score was 5.8 ± 1.7. The BMI was 25.6 ± 3.9 kg/m2, the WC was 85.4 ± 10.6 cm, 16.3% were obese according to the IOTF classification and 80.5% had a high WC according to the McCarthy classification. Table 1 Baseline socioeconomic and anthropometric characteristics of adolescents N % Socioeconomic Age (year), mean (SD) 2113 15.3 (0.7) Gender (girls) 1188 56.2 School type     Vocational high school 720 34.1     General and technological high school 1013 48.0     Middle schools 378 17.9     Scholarship holder 507 29.2 Parental place of birth     Both outside France 156 7.9     One in France 144 7.3     Both in France 1668 84.8  Perception of family financial level     Not at all 28 1.3     Slightly 151 7.2     Moderately 919 43.6     Rather affluent 1012 48.0 Father education level (ISCEDa)     < Level 3 1556 25.1     ≥ Level 3 444 22.2 Mother education level (ISCEDa)     < Level 3 1364 20.9     ≥ Level 3 712 34.3 Father employment status     Job search 101 5.3     Others 151 7.9     Employed 1668 86.9 Mother employment status     Job search 179 8.8     Others 391 19.1     Employed 1475 72.1 Father social and professional class (ISCO-08b)     Other people unemployed 243 11.6     Elementary occupations 125 6.0     Plant and machine operators, and assemblers 776 37.1     Craft and related trades workers 101 4.8     Skilled agricultural, forestry and fishery workers 42 2.0     Service and sales workers 164 7.8     Clerical support workers 182 8.7     Technicians and associate professionals 162 7.7     Professionals 183 8.8     Managers 113 5.4 Mother social and professional class (ISCO-08b)     Other people unemployed 466 22.5     Elementary occupations 121 5.8     Plant and machine operators, and assemblers 322 15.5     Craft and related trades workers 25 1.2     Skilled agricultural, forestry and fishery workers 25 1.2     Service and sales workers 352 17.0     Clerical support workers 299 14.4     Technicians and associate professionals 85 4.1     Professionals 346 16.7     Managers 32 1.5 Family affluence scale score, mean (SD) 2113 5.8 (1.7) Anthropometric measurements     BMI (kg/m2), mean (SD) 2113 25.6 (3.9)     BMI Z-score, mean (SD) 2113 1.4 (0.8)     Obesity (IOTFc classification) 344 16.3     WC (cm), mean (SD) 2112 85.4 (10.6)     High WCd (McCarthy classification) 1701 80.5     WHtR, mean (SD) 2112 0.5 (0.1)     High WHtR 1019 48.2     Overweight self-perception 704 37.3 N % Socioeconomic Age (year), mean (SD) 2113 15.3 (0.7) Gender (girls) 1188 56.2 School type     Vocational high school 720 34.1     General and technological high school 1013 48.0     Middle schools 378 17.9     Scholarship holder 507 29.2 Parental place of birth     Both outside France 156 7.9     One in France 144 7.3     Both in France 1668 84.8  Perception of family financial level     Not at all 28 1.3     Slightly 151 7.2     Moderately 919 43.6     Rather affluent 1012 48.0 Father education level (ISCEDa)     < Level 3 1556 25.1     ≥ Level 3 444 22.2 Mother education level (ISCEDa)     < Level 3 1364 20.9     ≥ Level 3 712 34.3 Father employment status     Job search 101 5.3     Others 151 7.9     Employed 1668 86.9 Mother employment status     Job search 179 8.8     Others 391 19.1     Employed 1475 72.1 Father social and professional class (ISCO-08b)     Other people unemployed 243 11.6     Elementary occupations 125 6.0     Plant and machine operators, and assemblers 776 37.1     Craft and related trades workers 101 4.8     Skilled agricultural, forestry and fishery workers 42 2.0     Service and sales workers 164 7.8     Clerical support workers 182 8.7     Technicians and associate professionals 162 7.7     Professionals 183 8.8     Managers 113 5.4 Mother social and professional class (ISCO-08b)     Other people unemployed 466 22.5     Elementary occupations 121 5.8     Plant and machine operators, and assemblers 322 15.5     Craft and related trades workers 25 1.2     Skilled agricultural, forestry and fishery workers 25 1.2     Service and sales workers 352 17.0     Clerical support workers 299 14.4     Technicians and associate professionals 85 4.1     Professionals 346 16.7     Managers 32 1.5 Family affluence scale score, mean (SD) 2113 5.8 (1.7) Anthropometric measurements     BMI (kg/m2), mean (SD) 2113 25.6 (3.9)     BMI Z-score, mean (SD) 2113 1.4 (0.8)     Obesity (IOTFc classification) 344 16.3     WC (cm), mean (SD) 2112 85.4 (10.6)     High WCd (McCarthy classification) 1701 80.5     WHtR, mean (SD) 2112 0.5 (0.1)     High WHtR 1019 48.2     Overweight self-perception 704 37.3 Data are N (%) unless indicated. BMI, body mass index; WC, waist circumference; WHtR, waist to height ratio; a International Standard Classification of Education; b International Standard Classification of Occupations—08; c International Obesity Task Force; d Waist Circumference. Table 1 Baseline socioeconomic and anthropometric characteristics of adolescents N % Socioeconomic Age (year), mean (SD) 2113 15.3 (0.7) Gender (girls) 1188 56.2 School type     Vocational high school 720 34.1     General and technological high school 1013 48.0     Middle schools 378 17.9     Scholarship holder 507 29.2 Parental place of birth     Both outside France 156 7.9     One in France 144 7.3     Both in France 1668 84.8  Perception of family financial level     Not at all 28 1.3     Slightly 151 7.2     Moderately 919 43.6     Rather affluent 1012 48.0 Father education level (ISCEDa)     < Level 3 1556 25.1     ≥ Level 3 444 22.2 Mother education level (ISCEDa)     < Level 3 1364 20.9     ≥ Level 3 712 34.3 Father employment status     Job search 101 5.3     Others 151 7.9     Employed 1668 86.9 Mother employment status     Job search 179 8.8     Others 391 19.1     Employed 1475 72.1 Father social and professional class (ISCO-08b)     Other people unemployed 243 11.6     Elementary occupations 125 6.0     Plant and machine operators, and assemblers 776 37.1     Craft and related trades workers 101 4.8     Skilled agricultural, forestry and fishery workers 42 2.0     Service and sales workers 164 7.8     Clerical support workers 182 8.7     Technicians and associate professionals 162 7.7     Professionals 183 8.8     Managers 113 5.4 Mother social and professional class (ISCO-08b)     Other people unemployed 466 22.5     Elementary occupations 121 5.8     Plant and machine operators, and assemblers 322 15.5     Craft and related trades workers 25 1.2     Skilled agricultural, forestry and fishery workers 25 1.2     Service and sales workers 352 17.0     Clerical support workers 299 14.4     Technicians and associate professionals 85 4.1     Professionals 346 16.7     Managers 32 1.5 Family affluence scale score, mean (SD) 2113 5.8 (1.7) Anthropometric measurements     BMI (kg/m2), mean (SD) 2113 25.6 (3.9)     BMI Z-score, mean (SD) 2113 1.4 (0.8)     Obesity (IOTFc classification) 344 16.3     WC (cm), mean (SD) 2112 85.4 (10.6)     High WCd (McCarthy classification) 1701 80.5     WHtR, mean (SD) 2112 0.5 (0.1)     High WHtR 1019 48.2     Overweight self-perception 704 37.3 N % Socioeconomic Age (year), mean (SD) 2113 15.3 (0.7) Gender (girls) 1188 56.2 School type     Vocational high school 720 34.1     General and technological high school 1013 48.0     Middle schools 378 17.9     Scholarship holder 507 29.2 Parental place of birth     Both outside France 156 7.9     One in France 144 7.3     Both in France 1668 84.8  Perception of family financial level     Not at all 28 1.3     Slightly 151 7.2     Moderately 919 43.6     Rather affluent 1012 48.0 Father education level (ISCEDa)     < Level 3 1556 25.1     ≥ Level 3 444 22.2 Mother education level (ISCEDa)     < Level 3 1364 20.9     ≥ Level 3 712 34.3 Father employment status     Job search 101 5.3     Others 151 7.9     Employed 1668 86.9 Mother employment status     Job search 179 8.8     Others 391 19.1     Employed 1475 72.1 Father social and professional class (ISCO-08b)     Other people unemployed 243 11.6     Elementary occupations 125 6.0     Plant and machine operators, and assemblers 776 37.1     Craft and related trades workers 101 4.8     Skilled agricultural, forestry and fishery workers 42 2.0     Service and sales workers 164 7.8     Clerical support workers 182 8.7     Technicians and associate professionals 162 7.7     Professionals 183 8.8     Managers 113 5.4 Mother social and professional class (ISCO-08b)     Other people unemployed 466 22.5     Elementary occupations 121 5.8     Plant and machine operators, and assemblers 322 15.5     Craft and related trades workers 25 1.2     Skilled agricultural, forestry and fishery workers 25 1.2     Service and sales workers 352 17.0     Clerical support workers 299 14.4     Technicians and associate professionals 85 4.1     Professionals 346 16.7     Managers 32 1.5 Family affluence scale score, mean (SD) 2113 5.8 (1.7) Anthropometric measurements     BMI (kg/m2), mean (SD) 2113 25.6 (3.9)     BMI Z-score, mean (SD) 2113 1.4 (0.8)     Obesity (IOTFc classification) 344 16.3     WC (cm), mean (SD) 2112 85.4 (10.6)     High WCd (McCarthy classification) 1701 80.5     WHtR, mean (SD) 2112 0.5 (0.1)     High WHtR 1019 48.2     Overweight self-perception 704 37.3 Data are N (%) unless indicated. BMI, body mass index; WC, waist circumference; WHtR, waist to height ratio; a International Standard Classification of Education; b International Standard Classification of Occupations—08; c International Obesity Task Force; d Waist Circumference. Identification of socioeconomic dimensions The following social indicators were linearly and significantly associated with anthropometric measurements and were selected for PCA: FAS score, scholarship holder, perception of family financial level, parents’ (mother and father) education level, social and professional category and employment status. PCA identified three dimensions labelled as followed: (i) ‘Family social status’ (including scholarship holder, mother employment status and father and mother social and professional category), (ii) ‘Family education level’ (including education level of each parent) and (iii) ‘Family income level’ (including FAS score, perception of family financial level and father’s employment status) (table 2). SEM confirmed an acceptable fit of the three dimensions (RMSEA = 0.08 (95% CI 0.07–0.09), CFI = 0.91 and NFI = 0.90) and allowed a calculation of each dimension score using path coefficient: Table 2 Principal component analysis of family socioeconomic dimensions Dimension 1 Dimension 2 Dimension 3 Family social status Family education level Family income level Parents’ education level     Father 0.06 0.88 0.08     Mother 0.19 0.83 0.07 Parents’ employment status     Father −0.01 0.02 0.64     Mother 0.73 −0.02 0.17 Parents’ social and professional classes (ISCOa classification)     Father 0.57 0.30 0.06     Mother 0.86 0.14 0.00 Family affluence scale in 5 classes 0.31 0.13 0.61 Scholarship holder 0.51 0.06 0.44 Perception of family financial level 0.07 0.04 0.68 Dimension 1 Dimension 2 Dimension 3 Family social status Family education level Family income level Parents’ education level     Father 0.06 0.88 0.08     Mother 0.19 0.83 0.07 Parents’ employment status     Father −0.01 0.02 0.64     Mother 0.73 −0.02 0.17 Parents’ social and professional classes (ISCOa classification)     Father 0.57 0.30 0.06     Mother 0.86 0.14 0.00 Family affluence scale in 5 classes 0.31 0.13 0.61 Scholarship holder 0.51 0.06 0.44 Perception of family financial level 0.07 0.04 0.68 a International Standard Classification of Occupations. Data are factor scores. Table 2 Principal component analysis of family socioeconomic dimensions Dimension 1 Dimension 2 Dimension 3 Family social status Family education level Family income level Parents’ education level     Father 0.06 0.88 0.08     Mother 0.19 0.83 0.07 Parents’ employment status     Father −0.01 0.02 0.64     Mother 0.73 −0.02 0.17 Parents’ social and professional classes (ISCOa classification)     Father 0.57 0.30 0.06     Mother 0.86 0.14 0.00 Family affluence scale in 5 classes 0.31 0.13 0.61 Scholarship holder 0.51 0.06 0.44 Perception of family financial level 0.07 0.04 0.68 Dimension 1 Dimension 2 Dimension 3 Family social status Family education level Family income level Parents’ education level     Father 0.06 0.88 0.08     Mother 0.19 0.83 0.07 Parents’ employment status     Father −0.01 0.02 0.64     Mother 0.73 −0.02 0.17 Parents’ social and professional classes (ISCOa classification)     Father 0.57 0.30 0.06     Mother 0.86 0.14 0.00 Family affluence scale in 5 classes 0.31 0.13 0.61 Scholarship holder 0.51 0.06 0.44 Perception of family financial level 0.07 0.04 0.68 a International Standard Classification of Occupations. Data are factor scores. ‘Family social status’ = +0.4988*(Scholarship holder) + 0.5898*(mother’s employment status) + 0.5147*(father’s social and professional category) + 0.7638*(mother’s social and professional category) ‘Family education level’ = +0.6378*(father’s education level) + 0.8635*(mother’s education level) ‘Family income level’ = +0.3936*(perception of family financial level) + 0.6840*(FAS) + s0.2727*(father’s employment status). Each score was them normalized ranging from 0 to 10 (the higher social class). The mean score of the dimensions were 5.2 ± 2.4 for ‘Family social status’, 4.7 ± 3.4 for ‘Family education level’ and 7.1 ± 2.0 for ‘Family income level’. For internal validated, the same analyses in a random 50% subsample led to similar dimensions and scoring with a good agreement (Cramer’s V statistic ranged from 0.77 to1.0). Social gradient analysis We found a significant social gradient in weight status for each identified SES dimension (table 3). Form all the three dimensions, the higher the score, the lower the weight status (BMI, BMI Z-score, WC and WHtR) with significant linear trend test. For the ‘Family income level’ dimension, c with highly less advantaged, highly advantaged adolescents had lower weight: −2.35 (−3.65; −1.05) kg/m2 for BMI, −0.45 (−0.71; −0.19) for BMI Z-score, −5.5 (−9.01; −2.02) cm for WC, −3.9 (−5.99; −1.81) for WHtR. Similar but slightly less significant results were found for ‘Family social status’ and ‘Family education level’ dimensions. The figure 1 shows the association between measured or perceived weight excess and socioeconomic dimensions. For measured weight excess indicators, there was a significant social gradient. Proportion of obesity (IOTF), high WHtR and high WC (McCarthy) significantly increased with the decrease SES. In contrast to measured weight excess, no social gradient was observed with the perceived weight status. Socially less advantaged adolescents did not perceived themselves more overweight than socially advantaged adolescents. Table 3 Association of family socioeconomic dimensions and adolescent weight measurements BMI BMI Z-score WC WHtR ratio (*100) β (95% CI)# P β (95% CI)# P β (95% CI)# P β (95% CI)# P Dimension 1: family social status     Highly less advantaged (17.2%) 0.0 0.0 0.0 0.0 –  Less advantaged (23.9%) −0.44 0.14 −0.07 0.22 −0.48 0.56 −0.54 0.26 (−1.02; 0.15) (−0.19; 0.04) (−2.08; 1.12) (−1.47; 0.39)  Intermediate (22.8%) −0.73 0.02 −0.14 0.02 −1.31 0.10 −1.55 0.001 (−1.33; −0.13) (−0.26; −0.02) (−2.89; 0.27) (−2.5; −0.6)  Advantaged (26.3%) −0.67 0.02 −0.12 0.04 −1.59 0.05 −1.51 0.001 (−1.25; −0.09) (−0.24; −0.01) (−3.22; 0.03) (−2.43; −0.59)  Highly advantaged (9.8%) −1.64 <0.0001 −0.33 <0.0001 −3.06 0.003 −2.53 <0.0001 (−2.39; −0.89) (−0.48; −0.18) (−5.1; −1.02) (−3.71; −1.34) P value for linear trend 0.0001 <0.0001 0.003 <0.0001 Dimension 2: Family education level     Highly less advantaged (37.5%) 0.0 0.0 0.0 0.0 –  Less advantaged (10.5%) 0.11 0.71 −0.01 0.92 0.58 0.48 0.28 0.56 (−0.48; 0.7) (−0.12; 0.11) (−1.04; 2.2) (−0.66; 1.22)  Intermediate (22.8%) −0.17 0.46 −0.03 0.46 −0.33 0.60 −0.56 0.13 (−0.62; 0.28) (−0.12; 0.06) (−1.56; 0.9) (−1.27; 0.16)  Advantaged (13.3%) −0.61 0.03 −0.12 0.03 −1.58 0.04 −1.21 0.006 (−1.15; −0.07) (−0.23; −0.01) (−3.06; −0.1) (−2.07; −0.35)  Highly advantaged (15.9%) −0.86 0.0008 −0.15 0.004 −2.42 0.0007 −2.08 <0.0001 (−1.37; −0.36) (−0.25; −0.05) (−3.8; −1.03) (−2.89; −1.27) P value for linear trend 0.0003 0.001 0.0003 <0.0001 Dimension 3: Family income level     Highly less advantaged (1.9%) 0.0 0.0 0.0 0.0 –  Less advantaged (14.4%) −1.32 0.049 −0.25 0.05 −3.22 0.07 −2.09 0.05 (−2.63; −0.01) (−0.51; 0.005) (−6.74; 0.3) (−4.20; 0.02)  Intermediate (16.8%) −1.50 0.02 −0.29 0.03 −3.79 0.03 −2.60 0.02 (−2.81; −0.2) (−0.55; −0.03) (−7.29; −0.3) (−4.69; −0.50)  Advantaged (49.1%) −1.90 0.003 −0.37 0.003 −4.94 0.004 −3.40 0.001 (−3.16; −0.64) (−0.62; −0.13) (−8.32; −1.55) (−5.43; −1.37)  Highly advantaged (17.8%) −2.35 0.0004 −0.45 0.0006 −5.52 0.002 −3.90 0.0003 (−3.65; −1.05) (−0.71; −0.19) (−9.01; −2.02) (−5.99; −1.81) P value for linear trend <0.0001 <0.0001 <0.0001 <0.0001 BMI BMI Z-score WC WHtR ratio (*100) β (95% CI)# P β (95% CI)# P β (95% CI)# P β (95% CI)# P Dimension 1: family social status     Highly less advantaged (17.2%) 0.0 0.0 0.0 0.0 –  Less advantaged (23.9%) −0.44 0.14 −0.07 0.22 −0.48 0.56 −0.54 0.26 (−1.02; 0.15) (−0.19; 0.04) (−2.08; 1.12) (−1.47; 0.39)  Intermediate (22.8%) −0.73 0.02 −0.14 0.02 −1.31 0.10 −1.55 0.001 (−1.33; −0.13) (−0.26; −0.02) (−2.89; 0.27) (−2.5; −0.6)  Advantaged (26.3%) −0.67 0.02 −0.12 0.04 −1.59 0.05 −1.51 0.001 (−1.25; −0.09) (−0.24; −0.01) (−3.22; 0.03) (−2.43; −0.59)  Highly advantaged (9.8%) −1.64 <0.0001 −0.33 <0.0001 −3.06 0.003 −2.53 <0.0001 (−2.39; −0.89) (−0.48; −0.18) (−5.1; −1.02) (−3.71; −1.34) P value for linear trend 0.0001 <0.0001 0.003 <0.0001 Dimension 2: Family education level     Highly less advantaged (37.5%) 0.0 0.0 0.0 0.0 –  Less advantaged (10.5%) 0.11 0.71 −0.01 0.92 0.58 0.48 0.28 0.56 (−0.48; 0.7) (−0.12; 0.11) (−1.04; 2.2) (−0.66; 1.22)  Intermediate (22.8%) −0.17 0.46 −0.03 0.46 −0.33 0.60 −0.56 0.13 (−0.62; 0.28) (−0.12; 0.06) (−1.56; 0.9) (−1.27; 0.16)  Advantaged (13.3%) −0.61 0.03 −0.12 0.03 −1.58 0.04 −1.21 0.006 (−1.15; −0.07) (−0.23; −0.01) (−3.06; −0.1) (−2.07; −0.35)  Highly advantaged (15.9%) −0.86 0.0008 −0.15 0.004 −2.42 0.0007 −2.08 <0.0001 (−1.37; −0.36) (−0.25; −0.05) (−3.8; −1.03) (−2.89; −1.27) P value for linear trend 0.0003 0.001 0.0003 <0.0001 Dimension 3: Family income level     Highly less advantaged (1.9%) 0.0 0.0 0.0 0.0 –  Less advantaged (14.4%) −1.32 0.049 −0.25 0.05 −3.22 0.07 −2.09 0.05 (−2.63; −0.01) (−0.51; 0.005) (−6.74; 0.3) (−4.20; 0.02)  Intermediate (16.8%) −1.50 0.02 −0.29 0.03 −3.79 0.03 −2.60 0.02 (−2.81; −0.2) (−0.55; −0.03) (−7.29; −0.3) (−4.69; −0.50)  Advantaged (49.1%) −1.90 0.003 −0.37 0.003 −4.94 0.004 −3.40 0.001 (−3.16; −0.64) (−0.62; −0.13) (−8.32; −1.55) (−5.43; −1.37)  Highly advantaged (17.8%) −2.35 0.0004 −0.45 0.0006 −5.52 0.002 −3.90 0.0003 (−3.65; −1.05) (−0.71; −0.19) (−9.01; −2.02) (−5.99; −1.81) P value for linear trend <0.0001 <0.0001 <0.0001 <0.0001 BMI, body mass index; WC, waist circumference; WHtR, waist to height ratio; # Beta coefficient of linear regression (95% confidence interval) adjusted for age and gender. Table 3 Association of family socioeconomic dimensions and adolescent weight measurements BMI BMI Z-score WC WHtR ratio (*100) β (95% CI)# P β (95% CI)# P β (95% CI)# P β (95% CI)# P Dimension 1: family social status     Highly less advantaged (17.2%) 0.0 0.0 0.0 0.0 –  Less advantaged (23.9%) −0.44 0.14 −0.07 0.22 −0.48 0.56 −0.54 0.26 (−1.02; 0.15) (−0.19; 0.04) (−2.08; 1.12) (−1.47; 0.39)  Intermediate (22.8%) −0.73 0.02 −0.14 0.02 −1.31 0.10 −1.55 0.001 (−1.33; −0.13) (−0.26; −0.02) (−2.89; 0.27) (−2.5; −0.6)  Advantaged (26.3%) −0.67 0.02 −0.12 0.04 −1.59 0.05 −1.51 0.001 (−1.25; −0.09) (−0.24; −0.01) (−3.22; 0.03) (−2.43; −0.59)  Highly advantaged (9.8%) −1.64 <0.0001 −0.33 <0.0001 −3.06 0.003 −2.53 <0.0001 (−2.39; −0.89) (−0.48; −0.18) (−5.1; −1.02) (−3.71; −1.34) P value for linear trend 0.0001 <0.0001 0.003 <0.0001 Dimension 2: Family education level     Highly less advantaged (37.5%) 0.0 0.0 0.0 0.0 –  Less advantaged (10.5%) 0.11 0.71 −0.01 0.92 0.58 0.48 0.28 0.56 (−0.48; 0.7) (−0.12; 0.11) (−1.04; 2.2) (−0.66; 1.22)  Intermediate (22.8%) −0.17 0.46 −0.03 0.46 −0.33 0.60 −0.56 0.13 (−0.62; 0.28) (−0.12; 0.06) (−1.56; 0.9) (−1.27; 0.16)  Advantaged (13.3%) −0.61 0.03 −0.12 0.03 −1.58 0.04 −1.21 0.006 (−1.15; −0.07) (−0.23; −0.01) (−3.06; −0.1) (−2.07; −0.35)  Highly advantaged (15.9%) −0.86 0.0008 −0.15 0.004 −2.42 0.0007 −2.08 <0.0001 (−1.37; −0.36) (−0.25; −0.05) (−3.8; −1.03) (−2.89; −1.27) P value for linear trend 0.0003 0.001 0.0003 <0.0001 Dimension 3: Family income level     Highly less advantaged (1.9%) 0.0 0.0 0.0 0.0 –  Less advantaged (14.4%) −1.32 0.049 −0.25 0.05 −3.22 0.07 −2.09 0.05 (−2.63; −0.01) (−0.51; 0.005) (−6.74; 0.3) (−4.20; 0.02)  Intermediate (16.8%) −1.50 0.02 −0.29 0.03 −3.79 0.03 −2.60 0.02 (−2.81; −0.2) (−0.55; −0.03) (−7.29; −0.3) (−4.69; −0.50)  Advantaged (49.1%) −1.90 0.003 −0.37 0.003 −4.94 0.004 −3.40 0.001 (−3.16; −0.64) (−0.62; −0.13) (−8.32; −1.55) (−5.43; −1.37)  Highly advantaged (17.8%) −2.35 0.0004 −0.45 0.0006 −5.52 0.002 −3.90 0.0003 (−3.65; −1.05) (−0.71; −0.19) (−9.01; −2.02) (−5.99; −1.81) P value for linear trend <0.0001 <0.0001 <0.0001 <0.0001 BMI BMI Z-score WC WHtR ratio (*100) β (95% CI)# P β (95% CI)# P β (95% CI)# P β (95% CI)# P Dimension 1: family social status     Highly less advantaged (17.2%) 0.0 0.0 0.0 0.0 –  Less advantaged (23.9%) −0.44 0.14 −0.07 0.22 −0.48 0.56 −0.54 0.26 (−1.02; 0.15) (−0.19; 0.04) (−2.08; 1.12) (−1.47; 0.39)  Intermediate (22.8%) −0.73 0.02 −0.14 0.02 −1.31 0.10 −1.55 0.001 (−1.33; −0.13) (−0.26; −0.02) (−2.89; 0.27) (−2.5; −0.6)  Advantaged (26.3%) −0.67 0.02 −0.12 0.04 −1.59 0.05 −1.51 0.001 (−1.25; −0.09) (−0.24; −0.01) (−3.22; 0.03) (−2.43; −0.59)  Highly advantaged (9.8%) −1.64 <0.0001 −0.33 <0.0001 −3.06 0.003 −2.53 <0.0001 (−2.39; −0.89) (−0.48; −0.18) (−5.1; −1.02) (−3.71; −1.34) P value for linear trend 0.0001 <0.0001 0.003 <0.0001 Dimension 2: Family education level     Highly less advantaged (37.5%) 0.0 0.0 0.0 0.0 –  Less advantaged (10.5%) 0.11 0.71 −0.01 0.92 0.58 0.48 0.28 0.56 (−0.48; 0.7) (−0.12; 0.11) (−1.04; 2.2) (−0.66; 1.22)  Intermediate (22.8%) −0.17 0.46 −0.03 0.46 −0.33 0.60 −0.56 0.13 (−0.62; 0.28) (−0.12; 0.06) (−1.56; 0.9) (−1.27; 0.16)  Advantaged (13.3%) −0.61 0.03 −0.12 0.03 −1.58 0.04 −1.21 0.006 (−1.15; −0.07) (−0.23; −0.01) (−3.06; −0.1) (−2.07; −0.35)  Highly advantaged (15.9%) −0.86 0.0008 −0.15 0.004 −2.42 0.0007 −2.08 <0.0001 (−1.37; −0.36) (−0.25; −0.05) (−3.8; −1.03) (−2.89; −1.27) P value for linear trend 0.0003 0.001 0.0003 <0.0001 Dimension 3: Family income level     Highly less advantaged (1.9%) 0.0 0.0 0.0 0.0 –  Less advantaged (14.4%) −1.32 0.049 −0.25 0.05 −3.22 0.07 −2.09 0.05 (−2.63; −0.01) (−0.51; 0.005) (−6.74; 0.3) (−4.20; 0.02)  Intermediate (16.8%) −1.50 0.02 −0.29 0.03 −3.79 0.03 −2.60 0.02 (−2.81; −0.2) (−0.55; −0.03) (−7.29; −0.3) (−4.69; −0.50)  Advantaged (49.1%) −1.90 0.003 −0.37 0.003 −4.94 0.004 −3.40 0.001 (−3.16; −0.64) (−0.62; −0.13) (−8.32; −1.55) (−5.43; −1.37)  Highly advantaged (17.8%) −2.35 0.0004 −0.45 0.0006 −5.52 0.002 −3.90 0.0003 (−3.65; −1.05) (−0.71; −0.19) (−9.01; −2.02) (−5.99; −1.81) P value for linear trend <0.0001 <0.0001 <0.0001 <0.0001 BMI, body mass index; WC, waist circumference; WHtR, waist to height ratio; # Beta coefficient of linear regression (95% confidence interval) adjusted for age and gender. Figure 1 View largeDownload slide Associations between socioeconomic dimensions and objective or perceived weight excess Figure 1 View largeDownload slide Associations between socioeconomic dimensions and objective or perceived weight excess Discussion Main results This population-based study demonstrates that an adequate combination of parental socioeconomic position is an effective way of measuring WSG in adolescence. The study also highlighted two interesting points. First, mother’s employment status was found to be a component of ‘Dimension 1: Family social status’ while father’s employment status was in the ‘Dimension 3: Family income level’. Consequently, in study including adolescents, father and mother employment status may not be substitutable social indicators. Second, the WSG was significant for objective measures of weight status while it was not for subjective measures. Adolescents of socially less advantaged classes perceived themselves less overweight than they are. Identified SES dimensions The combinations of parental SES indicators yielded three clear and consistent family-related dimensions: ‘Family social status’, ‘Family education level’ and ‘Family income level’. This result confirmed that SES is a composite indicator which combines education level, social status and income level and with a different weight for each SES among each dimension. It should be noted that being employed reflected different SES component of adolescent depending on whether it was the status of mother or father. Studies often used the head of household SES (education level or employment status) as the SES of adolescent which can result in different findings depending on which parent is consider as head of household.27,28 Moreover, each component of identified SES dimensions had a specific weight and highlighted the importance of building composite index of SES when investigating socioeconomic. To ensure a great understanding of socioeconomic disparities, studies should consider these components simultaneously. Unfortunately, this studies often focus on one or some of them with debatable combinations methods.28 The result of this study can be the base of creating a composite scoring of adolescents SES using parents SES. Weight socioeconomic gradient Weight status was significantly and inversely associated with the SES of adolescents (for each SES dimension) confirming the presence of WSG as described in numerous studies. Adolescents with low SES were more likely to have high weight status. Despite the fact that overweight or obesity prevalence has remained relatively stable in recent years, it still increasing in low SES classes resulting in the worsening of this WSG.29–32 Worst still, the perceived WSG was not significant while the measured one was. It highlights a potential weight misperception according to the SES. Socially less advantaged adolescents may perceive themselves less corpulent than they were or advantaged adolescents may perceive themselves more corpulent than they were. This studies have pointed the importance of weight misperception among adolescents. Lu et al. showed that only 21% of boys and 36% of girls who are overweight, and 58% of boys and 73% of girls with obesity, accurately perceive their weight status.33 An accurate perception of one’s weight is believed to be necessary to motivate weight loss intention because misperception of weight status is known to be associated with unhealthy weight-related behaviours.34,35 Differential weight misperception according to adolescents’ SES may contribute to maintain or worsen WSG. Public health intervention may also act on reducing weight misperception for more effectiveness. Although the WSG was significant for all the three SES dimensions when using objective weight status, the dimension of the ‘family income level’ seemed to have the strongest association with adolescent weight status and it raises the question of how relevant is each SES dimension in social inequalities in health. The SES dimensions may not equally impacted adolescents weight status. It is well established that health-related behaviours such as physical activity, sedentary behaviour and diet are significantly associated with adolescents’ weight status. It is also known that the affluence of the adolescents’ family can impact their health-behaviours.35 So understanding the causal mechanism of possible mediating effect of the relationship between health-related behaviours and weight status through SES (mediation effect of SES) can help developing more effective public health interventions to reduce WSG in adolescents. Limitations and strengths The main limitation of this study is the cross-sectional design with could not allowed for a causal relationship between weight status and SES. Even if analyzing the relationship between SES and weight status is a relevant issue, in this study we were more interested in how to measure efficiently the WSG but combining SES indicators rather than analyzing the causality relationship between weight status and SES. Other limitations include the use of self-reported questionnaire to assess SES with a potential reporting bias. However, adolescents’ self-reported SES was completed by data collected from the Academic Board of Education with a view to ensure consistency of these data. Despite these limitations, a number of aspects of the study should provide confidence in its findings. This is the first study to describe a socioeconomic gradient in weight status among adolescents by combining parental SES indicators. The identified SES dimensions were consistent to what we expected. Other strengths were the large sample size, multicentre design and valid measure of weight status. Conclusion Beyond the confirmation of significant WSG in adolescence, the result of this study highlights the importance of an adequate combination of parental SES to determine adolescents SES. The results of this study can help to develop multicomponent tool for measuring socioeconomic gradient in adolescence. A difference between objective and perceived weight status according to the SES of the adolescent was also observed. Adolescent misperception of one’s weight status according to their SES may be one of the explaining factors of maintaining or worsening social inequalities in health. Reducing weight misperception especially in socially less advantaged adolescents can help changing health-related behaviour in public health intervention. Longitudinal studies focusing on weight misperception are needed. Acknowledgements Many people worked together selflessly and enthusiastically to make the PRALIMAP-INÈS trial a success. The PRALIMAP-INÈS trial group warmly acknowledges the students and their parents who participated in the measurements and interventions as well as the school professionals (nurses, teachers, administrative staff and headmasters’ staff) who helped recruit students and deliver the interventions. PRALIMAP-INÈS Trial Group: Philip Böhme (PB), Serge Briançon (SB), Rozenn De Lavenne (RDL), Cécile Gailliard (CG), Johanne Langlois (JL), Edith Lecomte (EL), Karine Legrand (KL), Laurent Muller (LM), Abdou Y. Omorou (AO), Céline Pourcher (CP), Marie-Hélène Quinet (MHQ), Laura Saez (LS), Elisabeth Spitz (ES), Brigitte Toussaint (BT). Funding The PRALIMAP-INÈS trial is funded by the French National Cancer institute (Institut National du Cancer N°2011-239). It also received support from public institution (Conseil Régional de Lorraine and Agence Régionale de Santé). The funding or sponsoring agency had and will have no role in all trial steps, design, data collection, analysis, write-up and reports. Conflicts of interest: None declared. Key points Adolescents’ SES is multidimensional and each dimension is composed of different parental socioeconomic indicator. Mother’s and father’s employment status are components of two different dimensions of adolescents’ SES (father’s employment status is a component of family income dimension while mother’s employment status is a component of family social status). Weight social gradient is significant when used objective measures (BMI, BMI Z-score and obesity proportion) but not significant when using adolescent perceived weight status. Socially less advantaged adolescents perceived themselves less overweight than they were. References 1 Lobstein T , Baur L , Uauy R . IASO International Obesity TaskForce . Obesity in children and young people: a crisis in public health . Obes Rev Off J Int Assoc Study Obes 2004 ; 5 : 4 – 104 . Google Scholar Crossref Search ADS 2 Verdot C , Torres M , Salanave B , Deschamps V . Corpulence des enfants et des adultes en France métropolitaine en 2015. Résultats de l’étude Esteban et évolution depuis 2006 . Bull Epidémiol Hebd 2017 ; 13 : 234 – 41 . 3 Magarey AM , Daniels LA , Boulton TJ , Cockington RA . Predicting obesity in early adulthood from childhood and parental obesity . Int J Obes Relat Metab Disord J Int Assoc Study Obes 2003 ; 27 : 505 – 13 . Google Scholar Crossref Search ADS 4 Dietz WH . Childhood weight affects adult morbidity and mortality . J Nutr 1998 ; 128 : 411S – 4S . Google Scholar Crossref Search ADS PubMed 5 Cole TJ , Bellizzi MC , Flegal KM , Dietz WH . Establishing a standard definition for child overweight and obesity worldwide: international survey . BMJ 2000 ; 320 : 1240 – 3 . Google Scholar Crossref Search ADS PubMed 6 McCarthy HD , Jarrett KV , Crawley HF . The development of waist circumference percentiles in British children aged 5.0-16.9 y . Eur J Clin Nutr 2001 ; 55 : 902 – 7 . Google Scholar Crossref Search ADS PubMed 7 Marmot M . Social justice, epidemiology and health inequalities . Eur J Epidemiol 2017 ; 32 : 537 – 46 . Google Scholar Crossref Search ADS PubMed 8 Preda A , Voigt K . The social determinants of health: why should we care? Am J Bioeth AJOB 2015 ; 15 : 25 – 36 . Google Scholar Crossref Search ADS PubMed 9 Mackenbach JP , Stirbu I , Roskam A-JR , et al. Socioeconomic inequalities in health in 22 European countries . N Engl J Med 2008 ; 358 : 2468 – 81 . Google Scholar Crossref Search ADS PubMed 10 Saurel-Cubizolles M-J , Chastang J-F , Menvielle G , et al. Social inequalities in mortality by cause among men and women in France . J Epidemiol Community Health 2009 ; 63 : 197 – 202 . Google Scholar Crossref Search ADS PubMed 11 Leclerc A , Chastang J-F , Menvielle G , Luce D . Socioeconomic inequalities in premature mortality in France: have they widened in recent decades? . Soc Sci Med 2006 ; 62 : 2035 – 45 . Google Scholar Crossref Search ADS PubMed 12 Marmot M , Ryff CD , Bumpass LL , et al. Social inequalities in health: next questions and converging evidence . Soc Sci Med 1997 ; 44 : 901 – 10 . Google Scholar Crossref Search ADS PubMed 13 Whitehead M , Dahlgren G . Concepts and Principles for Tackling Social Inequities in Health: Levelling up Part 1. World Health Organization, Studies on Social and Economic Determinants of Population Health, 2006 , No. 2. Available at: http://www.enothe.eu/cop/docs/concepts_and_principles.pdf (26 July 2017, date last accessed). 14 Haut Conseil de la Santé Publique. Indicateurs de suivi des inégalités sociales de santé. Haut Conseil de la Santé Publique, Paris, 2013 . Available at: http://www.hcsp.fr/explore.cgi/avisrapportsdomaine? clefr=360 (7 April 2017, date last accessed). 15 Junek W . Developmental Health and the Wealth of Nations: Social, Biological and Educational Dynamics . Can Child Adolesc Psychiatr Rev 2003 ; 12 : 26 – 7 . 16 Legrand K , Lecomte E , Langlois J , et al. Reducing social inequalities in access to overweight and obesity care management for adolescents: the PRALIMAP-INÈS trial protocol and inclusion data analysis . Contemp Clin Trials Commun 2017 ; 7 : 141 – 57 . Google Scholar Crossref Search ADS PubMed 17 Ashwell M . Charts based on body mass index and waist-to-height ratio to assess the health risks of obesity: a review . Open Obes J 2011 ; 3 : 78 – 84 . Google Scholar Crossref Search ADS 18 Ashwell M , Gibson S . Waist-to-height ratio as an indicator of ‘early health risk’: simpler and more predictive than using a ‘matrix’based on BMI and waist circumference . BMJ Open 2016 ; 6 : e010159 . Google Scholar Crossref Search ADS PubMed 19 Sørensen TI , Stunkard AJ , Teasdale TW , Higgins MW . The accuracy of reports of weight: children’s recall of their parents’ weights 15 years earlier . Int J Obes 1983 ; 7 : 115 – 22 . Google Scholar PubMed 20 Currie CE , Elton RA , Todd J , Platt S . Indicators of socioeconomic status for adolescents: the WHO Health Behaviour in School-aged Children Survey . Health Educ Res 1997 ; 12 : 385 – 97 . Google Scholar Crossref Search ADS PubMed 21 Boyce W , Torsheim T , Currie C , Zambon A . The family affluence scale as a measure of national wealth: validation of an adolescent self-report measure . Soc Indic Res 2006 ; 78 : 473 – 87 . Google Scholar Crossref Search ADS 22 Kaiser HF . The varimax criterion for analytic rotation in factor analysis . Psychometrika 1958 ; 23 : 187 – 200 . Google Scholar Crossref Search ADS 23 Kaiser HF . An index of factorial simplicity . Psychometrika 1974 ; 39 : 31 – 6 . Google Scholar Crossref Search ADS 24 Hooper D , Coughlan J , Mullen M . Structural Equation Modelling: Guidelines for Determining Model Fit . Electronic Journal of Business Research Methods 2008 ; 6 : 53 – 60 . 25 Cramér H . Mathematical Methods of Statistics . Princeton : Princeton University Press, Vol. IX , 1999 : 575 . 26 Omorou AY , Coste J , Escalon H , Vuillemin A . Patterns of physical activity and sedentary behaviour in the general population in France: cluster analysis with personal and socioeconomic correlates . J Public Health Oxf Engl 2015 ; 38 : 483 – 92 . Google Scholar Crossref Search ADS 27 Lioret S , Touvier M , Dubuisson C , et al. Trends in child overweight rates and energy intake in France from 1999 to 2007: relationships with socioeconomic status . Obes Silver Spring Md 2009 ; 17 : 1092 – 100 . Google Scholar Crossref Search ADS 28 Salanave B , Peneau S , Rolland-Cachera M-F , et al. Stabilization of overweight prevalence in French children between 2000 and 2007 . Int J Pediatr Obes IJPO Off J Int Assoc Study Obes 2009 ; 4 : 66 – 72 . 29 Chung A , Backholer K , Wong E , et al. Trends in child and adolescent obesity prevalence in economically advanced countries according to socioeconomic position: a systematic review . Obes Rev Off J Int Assoc Study Obes 2016 ; 17 : 276 – 95 . Google Scholar Crossref Search ADS 30 Hardy LL , Mihrshahi S , Gale J , et al. 30-year trends in overweight, obesity and waist-to-height ratio by socioeconomic status in Australian children, 1985 to 2015 . Int J Obes 2005 2017 ; 41 : 76 – 82 . 31 Lu H , Tarasenko YN , Asgari-Majd FC , et al. More overweight adolescents think they are just fine: generational shift in body weight perceptions among adolescents in the U.S . Am J Prev Med 2015 ; 49 : 670 – 7 . Google Scholar Crossref Search ADS PubMed 32 Datar A , Chung PJ . Accuracy of weight perceptions in a nationally representative cohort of US 8th grade adolescents . Acad Pediatr 2016 ; 16 : 267 – 74 . Google Scholar Crossref Search ADS PubMed 33 Fredrickson J , Kremer P , Swinburn B , et al. Weight perception in overweight adolescents: associations with body change intentions, diet and physical activity . J Health Psychol 2015 ; 20 : 774 – 84 . Google Scholar Crossref Search ADS PubMed 34 Storey KE , Forbes LE , Fraser SN , et al. Adolescent weight status and related behavioural factors: web survey of physical activity and nutrition . J Obes 2012 ; 2012 : 1 . Google Scholar Crossref Search ADS 35 Langlois J , Omorou AY , Vuillemin A , et al. Association of socioeconomic, school-related and family factors and physical activity and sedentary behaviour among adolescents: multilevel analysis of the PRALIMAP trial inclusion data . BMC Public Health 2017 ; 17 : 175 . Google Scholar Crossref Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The European Journal of Public Health Oxford University Press

Loading next page...
 
/lp/ou_press/measuring-adolescents-weight-socioeconomic-gradient-using-parental-zf3WxjNmVM
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved.
ISSN
1101-1262
eISSN
1464-360X
D.O.I.
10.1093/eurpub/cky064
Publisher site
See Article on Publisher Site

Abstract

Abstract Background There is an evidence of social inequalities in weight status in adolescence but the diversity of family socioeconomic status (SES) indicators can lead to discrepant findings. We aimed to identify how combination of family SES indicators can help measuring weight socioeconomic gradient (WSG) among adolescents. Methods Cross-sectional data from 2113 adolescents (13–18 years old) of the PRALIMAP-INÈS trial were used. Multiple SES indicators and assessment of weight status including body mass index (BMI), waist circumference (WC), waist-to-height ratio (WHtR) and self-perception of overweight were used. We used principal component analysis (PCA) followed by structural equation models to identify SES dimensions. A dimension normalized score was calculated ranging from 1 to 10 (a high score corresponding to high SES). Linear regression models (linear trend test) were used to assess the WSG. Results Three SES dimensions were identified: (i) ‘Family social status’, (ii) ‘Family education level’ and (iii) ‘Family income level’. BMI was significantly lower in highly advantaged compared with highly less advantaged [−1.64 (−2.39; −0.89) for family social dimension, −0.86 (−1.37; −0.36) for family education level and −2.35 (−3.65; −1.05) for family income level]. Similar results were observed for all weight indicators excepted for self-perception of overweight status. Socially less advantaged adolescents perceived themselves less fat than they were. Conclusion Although WSG was evident in adolescence, association between SES and weight status differed according to objective or perceived weight indicators. The proposed SES dimension can be applied in other field and future studies are needed to confirm our findings. Introduction Over the last decades, the prevalence of overweight and obesity among adolescents has increased worldwide, both in developed and developing countries, but at an uneven pace.1 In France, the prevalence of overweight and obesity was estimated at 17% including 4% of obesity among adolescents.2 Being overweight in adolescence increases the likelihood of being obese in adulthood with several undesirable health consequences.3,4 Weight status can be assessed by using various anthropometric measurements that are built from combinations of weight, height and waist circumference (WC). The most commonly used measurements are body mass index (BMI) and WC. Other indicators include waist-to-height ratio (WHtR), a measure of the distribution of body fat. In addition to the diversity of weight indicators, defining weight status in adolescence is difficult because of growth during this period. Several cut-offs defined by gender and age account for physiological variations and use national and international growth curves: the International Obesity Task Force (IOTF) thresholds for BMI5 and high WC according to McCarthy cut-off values.6 Social inequalities in health are defined as differences in health according to the socioeconomic status (SES) that are unacceptable and potentially avoidable but, in addition, are also considered unfair.7,8 Social inequalities in health is characterized by differences in indicators of mortality and morbidity according to the SES.9 France is one of the European countries with the highest level of social inequalities in health.10,11 Social inequalities in health are found, not just at the threshold of poverty, but at every level of SES hierarchy and is called social gradient (SG) in health.12,13 The SG in health can be defined as health gradual differences found between all rungs of socioeconomic strata (each socioeconomic strata has higher levels of mortality and morbidity than the one above it).12 Several studies have shown a potential SG in adolescent weight status characterized by a variation in weight depending on adolescent SES.11–13 However, adolescents’ SES is a multifactorial concept which can integrate both adolescents’ perception of their own SES and parental objective SES indicators (education level, socio-professional category, income level, etc.). According to the French Public Health Council, adolescents tend to perceive their SES as equivalent to that of their parents.14 The large number of social indicators available may lead to discrepant results according to the indicator which is used. Each of them may provide different information, but their combination may help to better estimate of adolescents’ SES. Therefore, we aimed to identify how combination of family SES indicators can help measuring weight socioeconomic gradient (WSG) among adolescents. Methods PRALIMAP-INÈS trial The PRALIMAP-INÈS trial is a mixed, prospective, multicentre trial that aimed to evaluate the effectiveness of a school-based intervention to address social inequalities in adolescent overweight. It included overweight and obese adolescents from 13 to 18 years old who were attending state-run middle (grade 9) or high (grade 10) schools. A total of 35 schools (21 high schools and 14 middle schools) participated. Adolescents were recruited during an inclusion session (T0) organized in each school over three waves: 2012–13, 2013–14 and 2014–15 academic years. Adolescents were first measured (weight, height and WC) and when eligible (overweight: BMI greater than the IOTF age and gender specific overweight threshold5 for age and gender and/or WC greater than the McCarthy age and gender specific cut-off;15 close to overweight: BMI greater than the IOTF age and gender specific overweight threshold5 −1 kg/m2), they were invited to complete questionnaires and have a medical visit. If the overweight was confirmed, the physician proposed them to be included in the PRALIMAP-INÈS care program. More details on the PRALIMAP-INÈS trial protocol are given elsewhere.16 The trial was approved by the French consultative committee for treatment of information in health research (no. 12.299), the French National Commission for Data Protection and Liberties (no. 912 372) and the French Persons Protection Committee (no. 2012/15) and was registered at ClinicalTrials.gov (NCT01 688 453). Study sample This study sample included eligible (overweight or close to overweight) adolescents who completed the PRALIMAP-INÈS inclusion questionnaires. Among the 8735 students with available anthropometric measurements, 2283 were eligible for the PRALIMAP-INÈS trial: adolescents between 13 and 18 years old, with BMI greater than the IOTF overweight threshold5 reduced by 1 kg/m2 for age and gender and/or with WC greater than the McCarthy cut-off value for age and gender.15 A total of 2113 eligible adolescents completed the questionnaires and constituted our study sample. Measurements This study used cross-sectional data collected over the three waves of the main study. Anthropometric measures: Weight (kg), height (cm) and WC (cm) were measured twice by a school and a clinical research nurses and used to calculate BMI (weight in kg divided by the square of height in m2) and WHtR (WC in cm divided by height in cm).The weight excess was determined by overweight/obesity prevalence according to the IOTF cut-off,5 high WC according to McCarthy,6 high WHtR defined as WHtR ≥ 0.517,18 and adolescents self-perception of their weight status using the Sorensen silhouettes.19 Socioeconomic measures: Age, gender, school type, family status (living with no, one or two parents), perception of family financial level, scholarship holder and parents (father and mother) education level, social and professional classes and employment status. These data were collected both from the adolescents (self-reported) and the Academic Board of Education. A specific questionnaire on measuring a score of financial affluence of the family (FAS: family affluence scale) was administrated.20,21 The score ranges from 0 (less affluent) to 9 (highest affluence). All these variables were ordinal and reconstructed in ascending order from the less advantaged to the more advantaged. Statistical analysis Socioeconomic and anthropometric characteristics were described as mean (SD) for continuous variables or N (%) for categorical variables. Statistically significant and linear association (social gradient) between each anthropometric and each SES variables was tested using bivariate linear or logistic regression. Significant SES variables were selected for principal component analysis (PCA) followed by structural equation modelling (SEM) to identify adolescents’ SES dimensions. The PCA was used for dimensions identification and the SEM for confirmation. The PCA used the varimax rotation (orthogonal transformation) method and dimension with eigen-value ≥1 were retained.22,23 The SEM was used for confirmation (goodness of fit) and scoring of the identified dimensions. The root mean square error approximation (RMSEA, acceptable fit if <0.08), the comparative fit index (CFI, acceptable fit if ≥0.90) and the normed fit index (NFI, acceptable fit if ≥0.90) were examined to evaluated the fit of the models.24 For each confirmed SES dimension, a normalized score was calculated using SEM path coefficients. To test the internal validity of the identified SES dimensions, the same analysis (PCA followed by SEM) was repeated in a random 50% sample of the initial study sample. We repeated a sub-sample analysis (PCA and SEM) with a random 50% sample of the initial sample. The agreement of SES dimensions was tested with the Cramer V statistic (good agreement corresponding to Cramer V ≥ 0.80).25,26 For each dimension, a normalized score was calculated ranging from 1 (lowest SES) to 10 (highest SES). Each score was then categorized into five classes in order to assess the weight social gradient: highly less advantaged (1–2), less advantaged (3–4), intermediate (5–6), advantaged (7–8) and highly advantaged (9–10). Weight social gradient was assessed as a linear association between each dimension categories and weight indicators using linear regression models adjusted for sex and age. For these analyses, social classes were used as categorical variable to measure differences between social classes and ‘Highly less advantaged’ adolescents constituted the reference group. Then the same analyses were repeated using social classes as continuous variable to obtain the P values for linear trend test. Statistical analysis involved use of SAS 9.3 (SAS Inst., Cary, NC, USA). P < 0.05 was considered statistically significant. Results Characteristics of the study sample Socioeconomic and anthropometric characteristics are presented in table 1. Among the 2113 adolescents, the mean age was 15.3 ± 0.7 years, 56.2% were girls, 48% attending general high schools and the mean affluence FAS score was 5.8 ± 1.7. The BMI was 25.6 ± 3.9 kg/m2, the WC was 85.4 ± 10.6 cm, 16.3% were obese according to the IOTF classification and 80.5% had a high WC according to the McCarthy classification. Table 1 Baseline socioeconomic and anthropometric characteristics of adolescents N % Socioeconomic Age (year), mean (SD) 2113 15.3 (0.7) Gender (girls) 1188 56.2 School type     Vocational high school 720 34.1     General and technological high school 1013 48.0     Middle schools 378 17.9     Scholarship holder 507 29.2 Parental place of birth     Both outside France 156 7.9     One in France 144 7.3     Both in France 1668 84.8  Perception of family financial level     Not at all 28 1.3     Slightly 151 7.2     Moderately 919 43.6     Rather affluent 1012 48.0 Father education level (ISCEDa)     < Level 3 1556 25.1     ≥ Level 3 444 22.2 Mother education level (ISCEDa)     < Level 3 1364 20.9     ≥ Level 3 712 34.3 Father employment status     Job search 101 5.3     Others 151 7.9     Employed 1668 86.9 Mother employment status     Job search 179 8.8     Others 391 19.1     Employed 1475 72.1 Father social and professional class (ISCO-08b)     Other people unemployed 243 11.6     Elementary occupations 125 6.0     Plant and machine operators, and assemblers 776 37.1     Craft and related trades workers 101 4.8     Skilled agricultural, forestry and fishery workers 42 2.0     Service and sales workers 164 7.8     Clerical support workers 182 8.7     Technicians and associate professionals 162 7.7     Professionals 183 8.8     Managers 113 5.4 Mother social and professional class (ISCO-08b)     Other people unemployed 466 22.5     Elementary occupations 121 5.8     Plant and machine operators, and assemblers 322 15.5     Craft and related trades workers 25 1.2     Skilled agricultural, forestry and fishery workers 25 1.2     Service and sales workers 352 17.0     Clerical support workers 299 14.4     Technicians and associate professionals 85 4.1     Professionals 346 16.7     Managers 32 1.5 Family affluence scale score, mean (SD) 2113 5.8 (1.7) Anthropometric measurements     BMI (kg/m2), mean (SD) 2113 25.6 (3.9)     BMI Z-score, mean (SD) 2113 1.4 (0.8)     Obesity (IOTFc classification) 344 16.3     WC (cm), mean (SD) 2112 85.4 (10.6)     High WCd (McCarthy classification) 1701 80.5     WHtR, mean (SD) 2112 0.5 (0.1)     High WHtR 1019 48.2     Overweight self-perception 704 37.3 N % Socioeconomic Age (year), mean (SD) 2113 15.3 (0.7) Gender (girls) 1188 56.2 School type     Vocational high school 720 34.1     General and technological high school 1013 48.0     Middle schools 378 17.9     Scholarship holder 507 29.2 Parental place of birth     Both outside France 156 7.9     One in France 144 7.3     Both in France 1668 84.8  Perception of family financial level     Not at all 28 1.3     Slightly 151 7.2     Moderately 919 43.6     Rather affluent 1012 48.0 Father education level (ISCEDa)     < Level 3 1556 25.1     ≥ Level 3 444 22.2 Mother education level (ISCEDa)     < Level 3 1364 20.9     ≥ Level 3 712 34.3 Father employment status     Job search 101 5.3     Others 151 7.9     Employed 1668 86.9 Mother employment status     Job search 179 8.8     Others 391 19.1     Employed 1475 72.1 Father social and professional class (ISCO-08b)     Other people unemployed 243 11.6     Elementary occupations 125 6.0     Plant and machine operators, and assemblers 776 37.1     Craft and related trades workers 101 4.8     Skilled agricultural, forestry and fishery workers 42 2.0     Service and sales workers 164 7.8     Clerical support workers 182 8.7     Technicians and associate professionals 162 7.7     Professionals 183 8.8     Managers 113 5.4 Mother social and professional class (ISCO-08b)     Other people unemployed 466 22.5     Elementary occupations 121 5.8     Plant and machine operators, and assemblers 322 15.5     Craft and related trades workers 25 1.2     Skilled agricultural, forestry and fishery workers 25 1.2     Service and sales workers 352 17.0     Clerical support workers 299 14.4     Technicians and associate professionals 85 4.1     Professionals 346 16.7     Managers 32 1.5 Family affluence scale score, mean (SD) 2113 5.8 (1.7) Anthropometric measurements     BMI (kg/m2), mean (SD) 2113 25.6 (3.9)     BMI Z-score, mean (SD) 2113 1.4 (0.8)     Obesity (IOTFc classification) 344 16.3     WC (cm), mean (SD) 2112 85.4 (10.6)     High WCd (McCarthy classification) 1701 80.5     WHtR, mean (SD) 2112 0.5 (0.1)     High WHtR 1019 48.2     Overweight self-perception 704 37.3 Data are N (%) unless indicated. BMI, body mass index; WC, waist circumference; WHtR, waist to height ratio; a International Standard Classification of Education; b International Standard Classification of Occupations—08; c International Obesity Task Force; d Waist Circumference. Table 1 Baseline socioeconomic and anthropometric characteristics of adolescents N % Socioeconomic Age (year), mean (SD) 2113 15.3 (0.7) Gender (girls) 1188 56.2 School type     Vocational high school 720 34.1     General and technological high school 1013 48.0     Middle schools 378 17.9     Scholarship holder 507 29.2 Parental place of birth     Both outside France 156 7.9     One in France 144 7.3     Both in France 1668 84.8  Perception of family financial level     Not at all 28 1.3     Slightly 151 7.2     Moderately 919 43.6     Rather affluent 1012 48.0 Father education level (ISCEDa)     < Level 3 1556 25.1     ≥ Level 3 444 22.2 Mother education level (ISCEDa)     < Level 3 1364 20.9     ≥ Level 3 712 34.3 Father employment status     Job search 101 5.3     Others 151 7.9     Employed 1668 86.9 Mother employment status     Job search 179 8.8     Others 391 19.1     Employed 1475 72.1 Father social and professional class (ISCO-08b)     Other people unemployed 243 11.6     Elementary occupations 125 6.0     Plant and machine operators, and assemblers 776 37.1     Craft and related trades workers 101 4.8     Skilled agricultural, forestry and fishery workers 42 2.0     Service and sales workers 164 7.8     Clerical support workers 182 8.7     Technicians and associate professionals 162 7.7     Professionals 183 8.8     Managers 113 5.4 Mother social and professional class (ISCO-08b)     Other people unemployed 466 22.5     Elementary occupations 121 5.8     Plant and machine operators, and assemblers 322 15.5     Craft and related trades workers 25 1.2     Skilled agricultural, forestry and fishery workers 25 1.2     Service and sales workers 352 17.0     Clerical support workers 299 14.4     Technicians and associate professionals 85 4.1     Professionals 346 16.7     Managers 32 1.5 Family affluence scale score, mean (SD) 2113 5.8 (1.7) Anthropometric measurements     BMI (kg/m2), mean (SD) 2113 25.6 (3.9)     BMI Z-score, mean (SD) 2113 1.4 (0.8)     Obesity (IOTFc classification) 344 16.3     WC (cm), mean (SD) 2112 85.4 (10.6)     High WCd (McCarthy classification) 1701 80.5     WHtR, mean (SD) 2112 0.5 (0.1)     High WHtR 1019 48.2     Overweight self-perception 704 37.3 N % Socioeconomic Age (year), mean (SD) 2113 15.3 (0.7) Gender (girls) 1188 56.2 School type     Vocational high school 720 34.1     General and technological high school 1013 48.0     Middle schools 378 17.9     Scholarship holder 507 29.2 Parental place of birth     Both outside France 156 7.9     One in France 144 7.3     Both in France 1668 84.8  Perception of family financial level     Not at all 28 1.3     Slightly 151 7.2     Moderately 919 43.6     Rather affluent 1012 48.0 Father education level (ISCEDa)     < Level 3 1556 25.1     ≥ Level 3 444 22.2 Mother education level (ISCEDa)     < Level 3 1364 20.9     ≥ Level 3 712 34.3 Father employment status     Job search 101 5.3     Others 151 7.9     Employed 1668 86.9 Mother employment status     Job search 179 8.8     Others 391 19.1     Employed 1475 72.1 Father social and professional class (ISCO-08b)     Other people unemployed 243 11.6     Elementary occupations 125 6.0     Plant and machine operators, and assemblers 776 37.1     Craft and related trades workers 101 4.8     Skilled agricultural, forestry and fishery workers 42 2.0     Service and sales workers 164 7.8     Clerical support workers 182 8.7     Technicians and associate professionals 162 7.7     Professionals 183 8.8     Managers 113 5.4 Mother social and professional class (ISCO-08b)     Other people unemployed 466 22.5     Elementary occupations 121 5.8     Plant and machine operators, and assemblers 322 15.5     Craft and related trades workers 25 1.2     Skilled agricultural, forestry and fishery workers 25 1.2     Service and sales workers 352 17.0     Clerical support workers 299 14.4     Technicians and associate professionals 85 4.1     Professionals 346 16.7     Managers 32 1.5 Family affluence scale score, mean (SD) 2113 5.8 (1.7) Anthropometric measurements     BMI (kg/m2), mean (SD) 2113 25.6 (3.9)     BMI Z-score, mean (SD) 2113 1.4 (0.8)     Obesity (IOTFc classification) 344 16.3     WC (cm), mean (SD) 2112 85.4 (10.6)     High WCd (McCarthy classification) 1701 80.5     WHtR, mean (SD) 2112 0.5 (0.1)     High WHtR 1019 48.2     Overweight self-perception 704 37.3 Data are N (%) unless indicated. BMI, body mass index; WC, waist circumference; WHtR, waist to height ratio; a International Standard Classification of Education; b International Standard Classification of Occupations—08; c International Obesity Task Force; d Waist Circumference. Identification of socioeconomic dimensions The following social indicators were linearly and significantly associated with anthropometric measurements and were selected for PCA: FAS score, scholarship holder, perception of family financial level, parents’ (mother and father) education level, social and professional category and employment status. PCA identified three dimensions labelled as followed: (i) ‘Family social status’ (including scholarship holder, mother employment status and father and mother social and professional category), (ii) ‘Family education level’ (including education level of each parent) and (iii) ‘Family income level’ (including FAS score, perception of family financial level and father’s employment status) (table 2). SEM confirmed an acceptable fit of the three dimensions (RMSEA = 0.08 (95% CI 0.07–0.09), CFI = 0.91 and NFI = 0.90) and allowed a calculation of each dimension score using path coefficient: Table 2 Principal component analysis of family socioeconomic dimensions Dimension 1 Dimension 2 Dimension 3 Family social status Family education level Family income level Parents’ education level     Father 0.06 0.88 0.08     Mother 0.19 0.83 0.07 Parents’ employment status     Father −0.01 0.02 0.64     Mother 0.73 −0.02 0.17 Parents’ social and professional classes (ISCOa classification)     Father 0.57 0.30 0.06     Mother 0.86 0.14 0.00 Family affluence scale in 5 classes 0.31 0.13 0.61 Scholarship holder 0.51 0.06 0.44 Perception of family financial level 0.07 0.04 0.68 Dimension 1 Dimension 2 Dimension 3 Family social status Family education level Family income level Parents’ education level     Father 0.06 0.88 0.08     Mother 0.19 0.83 0.07 Parents’ employment status     Father −0.01 0.02 0.64     Mother 0.73 −0.02 0.17 Parents’ social and professional classes (ISCOa classification)     Father 0.57 0.30 0.06     Mother 0.86 0.14 0.00 Family affluence scale in 5 classes 0.31 0.13 0.61 Scholarship holder 0.51 0.06 0.44 Perception of family financial level 0.07 0.04 0.68 a International Standard Classification of Occupations. Data are factor scores. Table 2 Principal component analysis of family socioeconomic dimensions Dimension 1 Dimension 2 Dimension 3 Family social status Family education level Family income level Parents’ education level     Father 0.06 0.88 0.08     Mother 0.19 0.83 0.07 Parents’ employment status     Father −0.01 0.02 0.64     Mother 0.73 −0.02 0.17 Parents’ social and professional classes (ISCOa classification)     Father 0.57 0.30 0.06     Mother 0.86 0.14 0.00 Family affluence scale in 5 classes 0.31 0.13 0.61 Scholarship holder 0.51 0.06 0.44 Perception of family financial level 0.07 0.04 0.68 Dimension 1 Dimension 2 Dimension 3 Family social status Family education level Family income level Parents’ education level     Father 0.06 0.88 0.08     Mother 0.19 0.83 0.07 Parents’ employment status     Father −0.01 0.02 0.64     Mother 0.73 −0.02 0.17 Parents’ social and professional classes (ISCOa classification)     Father 0.57 0.30 0.06     Mother 0.86 0.14 0.00 Family affluence scale in 5 classes 0.31 0.13 0.61 Scholarship holder 0.51 0.06 0.44 Perception of family financial level 0.07 0.04 0.68 a International Standard Classification of Occupations. Data are factor scores. ‘Family social status’ = +0.4988*(Scholarship holder) + 0.5898*(mother’s employment status) + 0.5147*(father’s social and professional category) + 0.7638*(mother’s social and professional category) ‘Family education level’ = +0.6378*(father’s education level) + 0.8635*(mother’s education level) ‘Family income level’ = +0.3936*(perception of family financial level) + 0.6840*(FAS) + s0.2727*(father’s employment status). Each score was them normalized ranging from 0 to 10 (the higher social class). The mean score of the dimensions were 5.2 ± 2.4 for ‘Family social status’, 4.7 ± 3.4 for ‘Family education level’ and 7.1 ± 2.0 for ‘Family income level’. For internal validated, the same analyses in a random 50% subsample led to similar dimensions and scoring with a good agreement (Cramer’s V statistic ranged from 0.77 to1.0). Social gradient analysis We found a significant social gradient in weight status for each identified SES dimension (table 3). Form all the three dimensions, the higher the score, the lower the weight status (BMI, BMI Z-score, WC and WHtR) with significant linear trend test. For the ‘Family income level’ dimension, c with highly less advantaged, highly advantaged adolescents had lower weight: −2.35 (−3.65; −1.05) kg/m2 for BMI, −0.45 (−0.71; −0.19) for BMI Z-score, −5.5 (−9.01; −2.02) cm for WC, −3.9 (−5.99; −1.81) for WHtR. Similar but slightly less significant results were found for ‘Family social status’ and ‘Family education level’ dimensions. The figure 1 shows the association between measured or perceived weight excess and socioeconomic dimensions. For measured weight excess indicators, there was a significant social gradient. Proportion of obesity (IOTF), high WHtR and high WC (McCarthy) significantly increased with the decrease SES. In contrast to measured weight excess, no social gradient was observed with the perceived weight status. Socially less advantaged adolescents did not perceived themselves more overweight than socially advantaged adolescents. Table 3 Association of family socioeconomic dimensions and adolescent weight measurements BMI BMI Z-score WC WHtR ratio (*100) β (95% CI)# P β (95% CI)# P β (95% CI)# P β (95% CI)# P Dimension 1: family social status     Highly less advantaged (17.2%) 0.0 0.0 0.0 0.0 –  Less advantaged (23.9%) −0.44 0.14 −0.07 0.22 −0.48 0.56 −0.54 0.26 (−1.02; 0.15) (−0.19; 0.04) (−2.08; 1.12) (−1.47; 0.39)  Intermediate (22.8%) −0.73 0.02 −0.14 0.02 −1.31 0.10 −1.55 0.001 (−1.33; −0.13) (−0.26; −0.02) (−2.89; 0.27) (−2.5; −0.6)  Advantaged (26.3%) −0.67 0.02 −0.12 0.04 −1.59 0.05 −1.51 0.001 (−1.25; −0.09) (−0.24; −0.01) (−3.22; 0.03) (−2.43; −0.59)  Highly advantaged (9.8%) −1.64 <0.0001 −0.33 <0.0001 −3.06 0.003 −2.53 <0.0001 (−2.39; −0.89) (−0.48; −0.18) (−5.1; −1.02) (−3.71; −1.34) P value for linear trend 0.0001 <0.0001 0.003 <0.0001 Dimension 2: Family education level     Highly less advantaged (37.5%) 0.0 0.0 0.0 0.0 –  Less advantaged (10.5%) 0.11 0.71 −0.01 0.92 0.58 0.48 0.28 0.56 (−0.48; 0.7) (−0.12; 0.11) (−1.04; 2.2) (−0.66; 1.22)  Intermediate (22.8%) −0.17 0.46 −0.03 0.46 −0.33 0.60 −0.56 0.13 (−0.62; 0.28) (−0.12; 0.06) (−1.56; 0.9) (−1.27; 0.16)  Advantaged (13.3%) −0.61 0.03 −0.12 0.03 −1.58 0.04 −1.21 0.006 (−1.15; −0.07) (−0.23; −0.01) (−3.06; −0.1) (−2.07; −0.35)  Highly advantaged (15.9%) −0.86 0.0008 −0.15 0.004 −2.42 0.0007 −2.08 <0.0001 (−1.37; −0.36) (−0.25; −0.05) (−3.8; −1.03) (−2.89; −1.27) P value for linear trend 0.0003 0.001 0.0003 <0.0001 Dimension 3: Family income level     Highly less advantaged (1.9%) 0.0 0.0 0.0 0.0 –  Less advantaged (14.4%) −1.32 0.049 −0.25 0.05 −3.22 0.07 −2.09 0.05 (−2.63; −0.01) (−0.51; 0.005) (−6.74; 0.3) (−4.20; 0.02)  Intermediate (16.8%) −1.50 0.02 −0.29 0.03 −3.79 0.03 −2.60 0.02 (−2.81; −0.2) (−0.55; −0.03) (−7.29; −0.3) (−4.69; −0.50)  Advantaged (49.1%) −1.90 0.003 −0.37 0.003 −4.94 0.004 −3.40 0.001 (−3.16; −0.64) (−0.62; −0.13) (−8.32; −1.55) (−5.43; −1.37)  Highly advantaged (17.8%) −2.35 0.0004 −0.45 0.0006 −5.52 0.002 −3.90 0.0003 (−3.65; −1.05) (−0.71; −0.19) (−9.01; −2.02) (−5.99; −1.81) P value for linear trend <0.0001 <0.0001 <0.0001 <0.0001 BMI BMI Z-score WC WHtR ratio (*100) β (95% CI)# P β (95% CI)# P β (95% CI)# P β (95% CI)# P Dimension 1: family social status     Highly less advantaged (17.2%) 0.0 0.0 0.0 0.0 –  Less advantaged (23.9%) −0.44 0.14 −0.07 0.22 −0.48 0.56 −0.54 0.26 (−1.02; 0.15) (−0.19; 0.04) (−2.08; 1.12) (−1.47; 0.39)  Intermediate (22.8%) −0.73 0.02 −0.14 0.02 −1.31 0.10 −1.55 0.001 (−1.33; −0.13) (−0.26; −0.02) (−2.89; 0.27) (−2.5; −0.6)  Advantaged (26.3%) −0.67 0.02 −0.12 0.04 −1.59 0.05 −1.51 0.001 (−1.25; −0.09) (−0.24; −0.01) (−3.22; 0.03) (−2.43; −0.59)  Highly advantaged (9.8%) −1.64 <0.0001 −0.33 <0.0001 −3.06 0.003 −2.53 <0.0001 (−2.39; −0.89) (−0.48; −0.18) (−5.1; −1.02) (−3.71; −1.34) P value for linear trend 0.0001 <0.0001 0.003 <0.0001 Dimension 2: Family education level     Highly less advantaged (37.5%) 0.0 0.0 0.0 0.0 –  Less advantaged (10.5%) 0.11 0.71 −0.01 0.92 0.58 0.48 0.28 0.56 (−0.48; 0.7) (−0.12; 0.11) (−1.04; 2.2) (−0.66; 1.22)  Intermediate (22.8%) −0.17 0.46 −0.03 0.46 −0.33 0.60 −0.56 0.13 (−0.62; 0.28) (−0.12; 0.06) (−1.56; 0.9) (−1.27; 0.16)  Advantaged (13.3%) −0.61 0.03 −0.12 0.03 −1.58 0.04 −1.21 0.006 (−1.15; −0.07) (−0.23; −0.01) (−3.06; −0.1) (−2.07; −0.35)  Highly advantaged (15.9%) −0.86 0.0008 −0.15 0.004 −2.42 0.0007 −2.08 <0.0001 (−1.37; −0.36) (−0.25; −0.05) (−3.8; −1.03) (−2.89; −1.27) P value for linear trend 0.0003 0.001 0.0003 <0.0001 Dimension 3: Family income level     Highly less advantaged (1.9%) 0.0 0.0 0.0 0.0 –  Less advantaged (14.4%) −1.32 0.049 −0.25 0.05 −3.22 0.07 −2.09 0.05 (−2.63; −0.01) (−0.51; 0.005) (−6.74; 0.3) (−4.20; 0.02)  Intermediate (16.8%) −1.50 0.02 −0.29 0.03 −3.79 0.03 −2.60 0.02 (−2.81; −0.2) (−0.55; −0.03) (−7.29; −0.3) (−4.69; −0.50)  Advantaged (49.1%) −1.90 0.003 −0.37 0.003 −4.94 0.004 −3.40 0.001 (−3.16; −0.64) (−0.62; −0.13) (−8.32; −1.55) (−5.43; −1.37)  Highly advantaged (17.8%) −2.35 0.0004 −0.45 0.0006 −5.52 0.002 −3.90 0.0003 (−3.65; −1.05) (−0.71; −0.19) (−9.01; −2.02) (−5.99; −1.81) P value for linear trend <0.0001 <0.0001 <0.0001 <0.0001 BMI, body mass index; WC, waist circumference; WHtR, waist to height ratio; # Beta coefficient of linear regression (95% confidence interval) adjusted for age and gender. Table 3 Association of family socioeconomic dimensions and adolescent weight measurements BMI BMI Z-score WC WHtR ratio (*100) β (95% CI)# P β (95% CI)# P β (95% CI)# P β (95% CI)# P Dimension 1: family social status     Highly less advantaged (17.2%) 0.0 0.0 0.0 0.0 –  Less advantaged (23.9%) −0.44 0.14 −0.07 0.22 −0.48 0.56 −0.54 0.26 (−1.02; 0.15) (−0.19; 0.04) (−2.08; 1.12) (−1.47; 0.39)  Intermediate (22.8%) −0.73 0.02 −0.14 0.02 −1.31 0.10 −1.55 0.001 (−1.33; −0.13) (−0.26; −0.02) (−2.89; 0.27) (−2.5; −0.6)  Advantaged (26.3%) −0.67 0.02 −0.12 0.04 −1.59 0.05 −1.51 0.001 (−1.25; −0.09) (−0.24; −0.01) (−3.22; 0.03) (−2.43; −0.59)  Highly advantaged (9.8%) −1.64 <0.0001 −0.33 <0.0001 −3.06 0.003 −2.53 <0.0001 (−2.39; −0.89) (−0.48; −0.18) (−5.1; −1.02) (−3.71; −1.34) P value for linear trend 0.0001 <0.0001 0.003 <0.0001 Dimension 2: Family education level     Highly less advantaged (37.5%) 0.0 0.0 0.0 0.0 –  Less advantaged (10.5%) 0.11 0.71 −0.01 0.92 0.58 0.48 0.28 0.56 (−0.48; 0.7) (−0.12; 0.11) (−1.04; 2.2) (−0.66; 1.22)  Intermediate (22.8%) −0.17 0.46 −0.03 0.46 −0.33 0.60 −0.56 0.13 (−0.62; 0.28) (−0.12; 0.06) (−1.56; 0.9) (−1.27; 0.16)  Advantaged (13.3%) −0.61 0.03 −0.12 0.03 −1.58 0.04 −1.21 0.006 (−1.15; −0.07) (−0.23; −0.01) (−3.06; −0.1) (−2.07; −0.35)  Highly advantaged (15.9%) −0.86 0.0008 −0.15 0.004 −2.42 0.0007 −2.08 <0.0001 (−1.37; −0.36) (−0.25; −0.05) (−3.8; −1.03) (−2.89; −1.27) P value for linear trend 0.0003 0.001 0.0003 <0.0001 Dimension 3: Family income level     Highly less advantaged (1.9%) 0.0 0.0 0.0 0.0 –  Less advantaged (14.4%) −1.32 0.049 −0.25 0.05 −3.22 0.07 −2.09 0.05 (−2.63; −0.01) (−0.51; 0.005) (−6.74; 0.3) (−4.20; 0.02)  Intermediate (16.8%) −1.50 0.02 −0.29 0.03 −3.79 0.03 −2.60 0.02 (−2.81; −0.2) (−0.55; −0.03) (−7.29; −0.3) (−4.69; −0.50)  Advantaged (49.1%) −1.90 0.003 −0.37 0.003 −4.94 0.004 −3.40 0.001 (−3.16; −0.64) (−0.62; −0.13) (−8.32; −1.55) (−5.43; −1.37)  Highly advantaged (17.8%) −2.35 0.0004 −0.45 0.0006 −5.52 0.002 −3.90 0.0003 (−3.65; −1.05) (−0.71; −0.19) (−9.01; −2.02) (−5.99; −1.81) P value for linear trend <0.0001 <0.0001 <0.0001 <0.0001 BMI BMI Z-score WC WHtR ratio (*100) β (95% CI)# P β (95% CI)# P β (95% CI)# P β (95% CI)# P Dimension 1: family social status     Highly less advantaged (17.2%) 0.0 0.0 0.0 0.0 –  Less advantaged (23.9%) −0.44 0.14 −0.07 0.22 −0.48 0.56 −0.54 0.26 (−1.02; 0.15) (−0.19; 0.04) (−2.08; 1.12) (−1.47; 0.39)  Intermediate (22.8%) −0.73 0.02 −0.14 0.02 −1.31 0.10 −1.55 0.001 (−1.33; −0.13) (−0.26; −0.02) (−2.89; 0.27) (−2.5; −0.6)  Advantaged (26.3%) −0.67 0.02 −0.12 0.04 −1.59 0.05 −1.51 0.001 (−1.25; −0.09) (−0.24; −0.01) (−3.22; 0.03) (−2.43; −0.59)  Highly advantaged (9.8%) −1.64 <0.0001 −0.33 <0.0001 −3.06 0.003 −2.53 <0.0001 (−2.39; −0.89) (−0.48; −0.18) (−5.1; −1.02) (−3.71; −1.34) P value for linear trend 0.0001 <0.0001 0.003 <0.0001 Dimension 2: Family education level     Highly less advantaged (37.5%) 0.0 0.0 0.0 0.0 –  Less advantaged (10.5%) 0.11 0.71 −0.01 0.92 0.58 0.48 0.28 0.56 (−0.48; 0.7) (−0.12; 0.11) (−1.04; 2.2) (−0.66; 1.22)  Intermediate (22.8%) −0.17 0.46 −0.03 0.46 −0.33 0.60 −0.56 0.13 (−0.62; 0.28) (−0.12; 0.06) (−1.56; 0.9) (−1.27; 0.16)  Advantaged (13.3%) −0.61 0.03 −0.12 0.03 −1.58 0.04 −1.21 0.006 (−1.15; −0.07) (−0.23; −0.01) (−3.06; −0.1) (−2.07; −0.35)  Highly advantaged (15.9%) −0.86 0.0008 −0.15 0.004 −2.42 0.0007 −2.08 <0.0001 (−1.37; −0.36) (−0.25; −0.05) (−3.8; −1.03) (−2.89; −1.27) P value for linear trend 0.0003 0.001 0.0003 <0.0001 Dimension 3: Family income level     Highly less advantaged (1.9%) 0.0 0.0 0.0 0.0 –  Less advantaged (14.4%) −1.32 0.049 −0.25 0.05 −3.22 0.07 −2.09 0.05 (−2.63; −0.01) (−0.51; 0.005) (−6.74; 0.3) (−4.20; 0.02)  Intermediate (16.8%) −1.50 0.02 −0.29 0.03 −3.79 0.03 −2.60 0.02 (−2.81; −0.2) (−0.55; −0.03) (−7.29; −0.3) (−4.69; −0.50)  Advantaged (49.1%) −1.90 0.003 −0.37 0.003 −4.94 0.004 −3.40 0.001 (−3.16; −0.64) (−0.62; −0.13) (−8.32; −1.55) (−5.43; −1.37)  Highly advantaged (17.8%) −2.35 0.0004 −0.45 0.0006 −5.52 0.002 −3.90 0.0003 (−3.65; −1.05) (−0.71; −0.19) (−9.01; −2.02) (−5.99; −1.81) P value for linear trend <0.0001 <0.0001 <0.0001 <0.0001 BMI, body mass index; WC, waist circumference; WHtR, waist to height ratio; # Beta coefficient of linear regression (95% confidence interval) adjusted for age and gender. Figure 1 View largeDownload slide Associations between socioeconomic dimensions and objective or perceived weight excess Figure 1 View largeDownload slide Associations between socioeconomic dimensions and objective or perceived weight excess Discussion Main results This population-based study demonstrates that an adequate combination of parental socioeconomic position is an effective way of measuring WSG in adolescence. The study also highlighted two interesting points. First, mother’s employment status was found to be a component of ‘Dimension 1: Family social status’ while father’s employment status was in the ‘Dimension 3: Family income level’. Consequently, in study including adolescents, father and mother employment status may not be substitutable social indicators. Second, the WSG was significant for objective measures of weight status while it was not for subjective measures. Adolescents of socially less advantaged classes perceived themselves less overweight than they are. Identified SES dimensions The combinations of parental SES indicators yielded three clear and consistent family-related dimensions: ‘Family social status’, ‘Family education level’ and ‘Family income level’. This result confirmed that SES is a composite indicator which combines education level, social status and income level and with a different weight for each SES among each dimension. It should be noted that being employed reflected different SES component of adolescent depending on whether it was the status of mother or father. Studies often used the head of household SES (education level or employment status) as the SES of adolescent which can result in different findings depending on which parent is consider as head of household.27,28 Moreover, each component of identified SES dimensions had a specific weight and highlighted the importance of building composite index of SES when investigating socioeconomic. To ensure a great understanding of socioeconomic disparities, studies should consider these components simultaneously. Unfortunately, this studies often focus on one or some of them with debatable combinations methods.28 The result of this study can be the base of creating a composite scoring of adolescents SES using parents SES. Weight socioeconomic gradient Weight status was significantly and inversely associated with the SES of adolescents (for each SES dimension) confirming the presence of WSG as described in numerous studies. Adolescents with low SES were more likely to have high weight status. Despite the fact that overweight or obesity prevalence has remained relatively stable in recent years, it still increasing in low SES classes resulting in the worsening of this WSG.29–32 Worst still, the perceived WSG was not significant while the measured one was. It highlights a potential weight misperception according to the SES. Socially less advantaged adolescents may perceive themselves less corpulent than they were or advantaged adolescents may perceive themselves more corpulent than they were. This studies have pointed the importance of weight misperception among adolescents. Lu et al. showed that only 21% of boys and 36% of girls who are overweight, and 58% of boys and 73% of girls with obesity, accurately perceive their weight status.33 An accurate perception of one’s weight is believed to be necessary to motivate weight loss intention because misperception of weight status is known to be associated with unhealthy weight-related behaviours.34,35 Differential weight misperception according to adolescents’ SES may contribute to maintain or worsen WSG. Public health intervention may also act on reducing weight misperception for more effectiveness. Although the WSG was significant for all the three SES dimensions when using objective weight status, the dimension of the ‘family income level’ seemed to have the strongest association with adolescent weight status and it raises the question of how relevant is each SES dimension in social inequalities in health. The SES dimensions may not equally impacted adolescents weight status. It is well established that health-related behaviours such as physical activity, sedentary behaviour and diet are significantly associated with adolescents’ weight status. It is also known that the affluence of the adolescents’ family can impact their health-behaviours.35 So understanding the causal mechanism of possible mediating effect of the relationship between health-related behaviours and weight status through SES (mediation effect of SES) can help developing more effective public health interventions to reduce WSG in adolescents. Limitations and strengths The main limitation of this study is the cross-sectional design with could not allowed for a causal relationship between weight status and SES. Even if analyzing the relationship between SES and weight status is a relevant issue, in this study we were more interested in how to measure efficiently the WSG but combining SES indicators rather than analyzing the causality relationship between weight status and SES. Other limitations include the use of self-reported questionnaire to assess SES with a potential reporting bias. However, adolescents’ self-reported SES was completed by data collected from the Academic Board of Education with a view to ensure consistency of these data. Despite these limitations, a number of aspects of the study should provide confidence in its findings. This is the first study to describe a socioeconomic gradient in weight status among adolescents by combining parental SES indicators. The identified SES dimensions were consistent to what we expected. Other strengths were the large sample size, multicentre design and valid measure of weight status. Conclusion Beyond the confirmation of significant WSG in adolescence, the result of this study highlights the importance of an adequate combination of parental SES to determine adolescents SES. The results of this study can help to develop multicomponent tool for measuring socioeconomic gradient in adolescence. A difference between objective and perceived weight status according to the SES of the adolescent was also observed. Adolescent misperception of one’s weight status according to their SES may be one of the explaining factors of maintaining or worsening social inequalities in health. Reducing weight misperception especially in socially less advantaged adolescents can help changing health-related behaviour in public health intervention. Longitudinal studies focusing on weight misperception are needed. Acknowledgements Many people worked together selflessly and enthusiastically to make the PRALIMAP-INÈS trial a success. The PRALIMAP-INÈS trial group warmly acknowledges the students and their parents who participated in the measurements and interventions as well as the school professionals (nurses, teachers, administrative staff and headmasters’ staff) who helped recruit students and deliver the interventions. PRALIMAP-INÈS Trial Group: Philip Böhme (PB), Serge Briançon (SB), Rozenn De Lavenne (RDL), Cécile Gailliard (CG), Johanne Langlois (JL), Edith Lecomte (EL), Karine Legrand (KL), Laurent Muller (LM), Abdou Y. Omorou (AO), Céline Pourcher (CP), Marie-Hélène Quinet (MHQ), Laura Saez (LS), Elisabeth Spitz (ES), Brigitte Toussaint (BT). Funding The PRALIMAP-INÈS trial is funded by the French National Cancer institute (Institut National du Cancer N°2011-239). It also received support from public institution (Conseil Régional de Lorraine and Agence Régionale de Santé). The funding or sponsoring agency had and will have no role in all trial steps, design, data collection, analysis, write-up and reports. Conflicts of interest: None declared. Key points Adolescents’ SES is multidimensional and each dimension is composed of different parental socioeconomic indicator. Mother’s and father’s employment status are components of two different dimensions of adolescents’ SES (father’s employment status is a component of family income dimension while mother’s employment status is a component of family social status). Weight social gradient is significant when used objective measures (BMI, BMI Z-score and obesity proportion) but not significant when using adolescent perceived weight status. Socially less advantaged adolescents perceived themselves less overweight than they were. References 1 Lobstein T , Baur L , Uauy R . IASO International Obesity TaskForce . Obesity in children and young people: a crisis in public health . Obes Rev Off J Int Assoc Study Obes 2004 ; 5 : 4 – 104 . Google Scholar Crossref Search ADS 2 Verdot C , Torres M , Salanave B , Deschamps V . Corpulence des enfants et des adultes en France métropolitaine en 2015. Résultats de l’étude Esteban et évolution depuis 2006 . Bull Epidémiol Hebd 2017 ; 13 : 234 – 41 . 3 Magarey AM , Daniels LA , Boulton TJ , Cockington RA . Predicting obesity in early adulthood from childhood and parental obesity . Int J Obes Relat Metab Disord J Int Assoc Study Obes 2003 ; 27 : 505 – 13 . Google Scholar Crossref Search ADS 4 Dietz WH . Childhood weight affects adult morbidity and mortality . J Nutr 1998 ; 128 : 411S – 4S . Google Scholar Crossref Search ADS PubMed 5 Cole TJ , Bellizzi MC , Flegal KM , Dietz WH . Establishing a standard definition for child overweight and obesity worldwide: international survey . BMJ 2000 ; 320 : 1240 – 3 . Google Scholar Crossref Search ADS PubMed 6 McCarthy HD , Jarrett KV , Crawley HF . The development of waist circumference percentiles in British children aged 5.0-16.9 y . Eur J Clin Nutr 2001 ; 55 : 902 – 7 . Google Scholar Crossref Search ADS PubMed 7 Marmot M . Social justice, epidemiology and health inequalities . Eur J Epidemiol 2017 ; 32 : 537 – 46 . Google Scholar Crossref Search ADS PubMed 8 Preda A , Voigt K . The social determinants of health: why should we care? Am J Bioeth AJOB 2015 ; 15 : 25 – 36 . Google Scholar Crossref Search ADS PubMed 9 Mackenbach JP , Stirbu I , Roskam A-JR , et al. Socioeconomic inequalities in health in 22 European countries . N Engl J Med 2008 ; 358 : 2468 – 81 . Google Scholar Crossref Search ADS PubMed 10 Saurel-Cubizolles M-J , Chastang J-F , Menvielle G , et al. Social inequalities in mortality by cause among men and women in France . J Epidemiol Community Health 2009 ; 63 : 197 – 202 . Google Scholar Crossref Search ADS PubMed 11 Leclerc A , Chastang J-F , Menvielle G , Luce D . Socioeconomic inequalities in premature mortality in France: have they widened in recent decades? . Soc Sci Med 2006 ; 62 : 2035 – 45 . Google Scholar Crossref Search ADS PubMed 12 Marmot M , Ryff CD , Bumpass LL , et al. Social inequalities in health: next questions and converging evidence . Soc Sci Med 1997 ; 44 : 901 – 10 . Google Scholar Crossref Search ADS PubMed 13 Whitehead M , Dahlgren G . Concepts and Principles for Tackling Social Inequities in Health: Levelling up Part 1. World Health Organization, Studies on Social and Economic Determinants of Population Health, 2006 , No. 2. Available at: http://www.enothe.eu/cop/docs/concepts_and_principles.pdf (26 July 2017, date last accessed). 14 Haut Conseil de la Santé Publique. Indicateurs de suivi des inégalités sociales de santé. Haut Conseil de la Santé Publique, Paris, 2013 . Available at: http://www.hcsp.fr/explore.cgi/avisrapportsdomaine? clefr=360 (7 April 2017, date last accessed). 15 Junek W . Developmental Health and the Wealth of Nations: Social, Biological and Educational Dynamics . Can Child Adolesc Psychiatr Rev 2003 ; 12 : 26 – 7 . 16 Legrand K , Lecomte E , Langlois J , et al. Reducing social inequalities in access to overweight and obesity care management for adolescents: the PRALIMAP-INÈS trial protocol and inclusion data analysis . Contemp Clin Trials Commun 2017 ; 7 : 141 – 57 . Google Scholar Crossref Search ADS PubMed 17 Ashwell M . Charts based on body mass index and waist-to-height ratio to assess the health risks of obesity: a review . Open Obes J 2011 ; 3 : 78 – 84 . Google Scholar Crossref Search ADS 18 Ashwell M , Gibson S . Waist-to-height ratio as an indicator of ‘early health risk’: simpler and more predictive than using a ‘matrix’based on BMI and waist circumference . BMJ Open 2016 ; 6 : e010159 . Google Scholar Crossref Search ADS PubMed 19 Sørensen TI , Stunkard AJ , Teasdale TW , Higgins MW . The accuracy of reports of weight: children’s recall of their parents’ weights 15 years earlier . Int J Obes 1983 ; 7 : 115 – 22 . Google Scholar PubMed 20 Currie CE , Elton RA , Todd J , Platt S . Indicators of socioeconomic status for adolescents: the WHO Health Behaviour in School-aged Children Survey . Health Educ Res 1997 ; 12 : 385 – 97 . Google Scholar Crossref Search ADS PubMed 21 Boyce W , Torsheim T , Currie C , Zambon A . The family affluence scale as a measure of national wealth: validation of an adolescent self-report measure . Soc Indic Res 2006 ; 78 : 473 – 87 . Google Scholar Crossref Search ADS 22 Kaiser HF . The varimax criterion for analytic rotation in factor analysis . Psychometrika 1958 ; 23 : 187 – 200 . Google Scholar Crossref Search ADS 23 Kaiser HF . An index of factorial simplicity . Psychometrika 1974 ; 39 : 31 – 6 . Google Scholar Crossref Search ADS 24 Hooper D , Coughlan J , Mullen M . Structural Equation Modelling: Guidelines for Determining Model Fit . Electronic Journal of Business Research Methods 2008 ; 6 : 53 – 60 . 25 Cramér H . Mathematical Methods of Statistics . Princeton : Princeton University Press, Vol. IX , 1999 : 575 . 26 Omorou AY , Coste J , Escalon H , Vuillemin A . Patterns of physical activity and sedentary behaviour in the general population in France: cluster analysis with personal and socioeconomic correlates . J Public Health Oxf Engl 2015 ; 38 : 483 – 92 . Google Scholar Crossref Search ADS 27 Lioret S , Touvier M , Dubuisson C , et al. Trends in child overweight rates and energy intake in France from 1999 to 2007: relationships with socioeconomic status . Obes Silver Spring Md 2009 ; 17 : 1092 – 100 . Google Scholar Crossref Search ADS 28 Salanave B , Peneau S , Rolland-Cachera M-F , et al. Stabilization of overweight prevalence in French children between 2000 and 2007 . Int J Pediatr Obes IJPO Off J Int Assoc Study Obes 2009 ; 4 : 66 – 72 . 29 Chung A , Backholer K , Wong E , et al. Trends in child and adolescent obesity prevalence in economically advanced countries according to socioeconomic position: a systematic review . Obes Rev Off J Int Assoc Study Obes 2016 ; 17 : 276 – 95 . Google Scholar Crossref Search ADS 30 Hardy LL , Mihrshahi S , Gale J , et al. 30-year trends in overweight, obesity and waist-to-height ratio by socioeconomic status in Australian children, 1985 to 2015 . Int J Obes 2005 2017 ; 41 : 76 – 82 . 31 Lu H , Tarasenko YN , Asgari-Majd FC , et al. More overweight adolescents think they are just fine: generational shift in body weight perceptions among adolescents in the U.S . Am J Prev Med 2015 ; 49 : 670 – 7 . Google Scholar Crossref Search ADS PubMed 32 Datar A , Chung PJ . Accuracy of weight perceptions in a nationally representative cohort of US 8th grade adolescents . Acad Pediatr 2016 ; 16 : 267 – 74 . Google Scholar Crossref Search ADS PubMed 33 Fredrickson J , Kremer P , Swinburn B , et al. Weight perception in overweight adolescents: associations with body change intentions, diet and physical activity . J Health Psychol 2015 ; 20 : 774 – 84 . Google Scholar Crossref Search ADS PubMed 34 Storey KE , Forbes LE , Fraser SN , et al. Adolescent weight status and related behavioural factors: web survey of physical activity and nutrition . J Obes 2012 ; 2012 : 1 . Google Scholar Crossref Search ADS 35 Langlois J , Omorou AY , Vuillemin A , et al. Association of socioeconomic, school-related and family factors and physical activity and sedentary behaviour among adolescents: multilevel analysis of the PRALIMAP trial inclusion data . BMC Public Health 2017 ; 17 : 175 . Google Scholar Crossref Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Journal

The European Journal of Public HealthOxford University Press

Published: Dec 1, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off