Abstract
Background: Obesity has become a global issue, since there are numerous indications that a greater number of children and adolescents are becoming overweight each year. In Greece, increasing trends in the proportion of overweight children have been observed. Nonetheless, epidemiological information is lacking as these findings pertained to different population samples, and robust conclusions could not be drawn.
Aim: The purposes of the present study were to assess the body mass index (BMI) and prevalence of overweight and obesity in a population-based, representative sample of Greek children at the ages of 7 and 18, and to examine changes in the proportion of children classified as overweight or obese from childhood to adolescence. In addition, we examined the association between BMI z-score at 7 and 18 years, and change in BMI z-score from 7 to 18 years, and perinatal, child, parental, socio-economic and adolescent lifestyle characteristics. Finally, cross-national comparisons aimed ...
Background: Obesity has become a global issue, since there are numerous indications that a greater number of children and adolescents are becoming overweight each year. In Greece, increasing trends in the proportion of overweight children have been observed. Nonetheless, epidemiological information is lacking as these findings pertained to different population samples, and robust conclusions could not be drawn.
Aim: The purposes of the present study were to assess the body mass index (BMI) and prevalence of overweight and obesity in a population-based, representative sample of Greek children at the ages of 7 and 18, and to examine changes in the proportion of children classified as overweight or obese from childhood to adolescence. In addition, we examined the association between BMI z-score at 7 and 18 years, and change in BMI z-score from 7 to 18 years, and perinatal, child, parental, socio-economic and adolescent lifestyle characteristics. Finally, cross-national comparisons aimed to reveal similarities and differences in the weight status of children and adolescents, and in characteristics among adolescents of a Northern (Finland) and Southern (Greece) European country.
Population and methods: A total of 11048 neonates, that represented 8% of the country’s annual births, were recorded nationwide from the 1st to the 30th of April 1983 and constituted a representative, population-based sample. Children were followed up through two, separate, questionnaire surveys conducted in 1990 and 2001. The first follow-up survey was conducted through primary schools nationwide, whereby a total of 8130 completed questionnaires from 7-year-olds nationwide were returned, and 6143 were merged with corresponding data from the 1983 survey. During the second follow-up participants were located at high-schools throughout the country. Of the 4675 questionnaires distributed, 3500 responded (75% response rate). For each survey, the questionnaires were kept anonymous. Finally, data was matched-up through the three study periods using key-variables, whereby complete data from birth up to 18 years was available for 2180 children. Due to the fact that the Greek adolescents’ responded to the questionnaire shortly before their university entry examinations, the attrition was relatively high. Nevertheless, sensitivity analyses showed that the follow-up data was well representative of the initial Greek birth cohort from whom it was selected.
Analysis was restricted to those who had valid weight and height measurements (6847 participants aged 7, and 2842 participants aged 18) to determine the BMI and prevalence of obesity among children aged 7 and 18 years. To determine the tracking of overweight and obesity, and the factors associated with BMI z-score and change in BMI z-score from 7 to 18 years, analysis was restricted to 2180 participants matched-up throughout the three study periods.
The independent variables analyzed in the present study from the three study periods included peripartum and parental factors, early childhood and adolescent lifestyle characteristics, and socio-economic status. Univariate and multivariate linear regression analyses were used to examine the associations between BMI z-score at 7 and 18 years and change in BMI z-score between 7 and 18 years with the independent variable studied. Chi-squared tests and logistic regression analyses were also used to examine the binary status of overweight and obesity and the included variables. A significance level of 0.05 was used and analysis was performed for both genders separately.
Results: Some of the most significant findings in the present study were:
1. The levels of overweight and obesity were similar for both genders in childhood, given that overweight boys and girls were 15.7% and 17.6%, respectively, and obese boys and girls were 6.1% and 5.6% each. Yet, during adolescence boys had a significantly higher prevalence of overweight and obesity than girls (19.1% vs. 7.8%, and 3.7% vs. 1.0% respectively, P< 0.001).
2. The tracking of subgroups showed that overall 73.7% of the 7-year-olds retain their classification of weight status at 18 years. For the obesity categories, 24.6% of the overweight 7-year-olds remained overweight at 18 years of age, and 15.0% of the obese 7-year-olds remained obese at the age of 18 years. However, there was a significant difference in the trends followed by each gender, as there was a greater shift from overweight/obesity in childhood to normal weight in adolescence among girls (77.9%) than among boys (51.8%). In fact, data revealed that overweight/obesity persisted in 48.2% of the boys and only in 22.1% of the girls.
3. BMI z-score at 7 years was positively associated with higher parental education, smaller number of siblings and responsiveness with food in boys. In girls, a positive association was also found with birthweight, maternal smoking during pregnancy and maternal pre-pregnancy obesity.
4. Parental weight status was positively associated with BMI z-score in boys and girls at 18 years, and with change in BMI z-score between 7 and 18 years, with the father mostly influencing his son, and the mother her daughter.
5. BMI z-score at 18 years was significantly influenced by psychosocial factors and eating behaviors in both genders.
6. The cross-national comparison confirmed higher levels of overweight youth at 7 years in Greece, compared to England, and a higher rate of overweight adolescents in Greece compared to Finland.
Conclusions: The high prevalence and tracking of overweight and obesity among Greek boys is an important public health problem. Those developing policy and programmes for obesity prevention should place more emphasis on appropriate measures in preschool years. A high priority for the future should be to further examine the associations between psychosocial parameters and obesity by gender, and to establish the extent to which these determinants can explain the obesity problem in our country.
show more