Tracking of weight status and body fatness in Italian children
- First Online:
- Cite this article as:
- Toselli, S., Brasili, P. & Di Michele, R. Eat Weight Disord (2013) 18: 383. doi:10.1007/s40519-013-0074-3
- 58 Views
The prevalence of weight disorders among school-aged children is an increasing phenomenon and it is of great importance to identify the characteristics of individuals at risk of gaining or retaining weight. This study aimed to examine the prevalence of weight disorders and their tracking over a 3-year period in a sample of Italian children. Body mass, body height and selected skinfold thicknesses were assessed in 355 children at the age of 7 and 10 years. Tracking of body mass index (BMI), inverted BMI (iBMI) and skinfold-based body fat were analyzed and the relationships between changes in BMI and body fat were examined. Children presenting with overweight or obesity at 7 years old showed a trend toward lower weight categories at 10 years old. Conversely, a trend to become overweight was observed among normal weight boys, and a trend to become underweight was observed among normal weight girls. BMI, iBMI and body fat showed good levels of tracking, with high correlations between measurements performed at 7 and 10 years of age. Furthermore, BMI and iBMI changes were correlated to body fatness changes. The present study shows the importance of carefully following children’s development over time because weight disorders may appear even in previously normal weight children.
KeywordsBody fatBody mass indexObesitySkinfoldsWeight disorders
In recent decades, overweight and obesity have been recognized as being among the fastest growing non-transmissible health problems worldwide [1, 2]. A number of factors have been shown to play a role in the development of these diseases, such as eating disorders [3–5], chronic distress  and hormonal imbalances . Apart from their origins, the increase in overweight and obesity is of particular concern as they are linked to numerous chronic diseases and other health disorders among both adults  and children [9–11]. Besides worsening the quality of living, obesity increases public and personal health expenditures [12, 13]. Obesity is becoming a public health priority in developed and developing countries and numerous studies have focused on the relationship between childhood obesity and young adult obesity in the perspective of preventing such a phenomenon. The rates of childhood  and adult obesity  are rapidly increasing and it is of interest to understand the extent to which childhood obesity persists over time . On the other hand, the prevalence of underweight individuals is an additional aspect of weight disorders, with lesser attention to date, that may have relevant consequences and therefore needs attention as well.
Longitudinal studies are valuable since they can reveal the tracking of weight status in a single or sample population and allow the identification of the characteristics of individuals who are at the highest risk of gaining or retaining weight . Several authors have shown a trend of individuals who are overweight or obese as children to remain overweight or obese as adolescents and adults [1, 18–21]. Various studies have focused on tracking body mass index (BMI) from childhood into adulthood , and have shown that a high BMI substantially persists over time [23–26]. However, there are some challenges to using BMI to assess adiposity (see, e.g., [27, 28]). Moreover, it has been argued that tracking BMI can reflect the tracking of body composition rather than fatness . Several methods are available to calculate body fat from other anthropometric measurements such as selected body perimeters  or skinfold thickness . Some authors assessed the tracking of body fat using skinfold-based adiposity [2, 32], thus showing the possibility to analyze the tracking of body composition parameters differing from BMI.
The aim of this study was to longitudinally assess the weight status group of Italian children born during or after the obesity epidemic was recognized. BMI and skinfold-based body fat were used to analyze how these parameters track over time and to evaluate the relationship between changes occurring in such parameters. Furthermore, we evaluated the inverted BMI (iBMI, expressed in cm2/kg) that was recently shown to be a sound alternative to BMI as a proxy for body fatness .
Materials and methods
From children participating in a prospective research project assessing physical and motor development during the school age, 173 girls and 172 boys were selected. All of the children attended the same primary school in a residential, middle/high class neighborhood of Bologna, northern Italy. The inclusion criterion for the present study was the complete availability of all the examined anthropometric measures at the age of 7 and 10 years. No exclusion criteria based on anthropometric data were used.
The examined data were collected at a duration of about 3 years. The mean ± SD children age was 7.0 ± 0.3 and 10.0 ± 0.3 years at the first and second assessment, respectively. The majority of children at age 7 (boys: 82 %; girls: 57 %) and the majority at age 10 (boys: 92 %; girls: 76 %) practiced at least one sport. The mean (SD) weekly time spent by boys practicing sport at age 7 and 10 years was 1.9 (0.8) and 3.3 (1.5) h, respectively, whereas in girls, at the same ages, sports practice per week was 1.7 (0.7) and 2.4 (1.7) h.
Parents or guardians provided written informed consent for participation. All of the procedures were designed and carried out in accordance with the Helsinki Declaration. The study protocol was approved by the university ethical committee and authorized by the schools involved.
Height and weight were measured; BMI (weight/height2, expressed in kg/m2) and iBMI (height2/weight, expressed in cm2/kg) were derived. According to their BMI, the children were classified as underweight, normal weight, overweight or obese using the International Obesity Task Force (IOTF) thresholds  for gender and exact age. The Cole et al.  thresholds were used for specific attribution of participants to the underweight category. Skinfold thickness (triceps, subscapular, suprailiac and medial calf) were measured with a Lange caliper on the left side of the body. Skinfold measurement was performed in all the children by the same experienced anthropometrist at 7 and 10 years. All anthropometric measurements were taken according to standard techniques described by Weiner and Lourie . The percentage of body fat was calculated from the triceps and subscapular skinfolds using Slaughter et al.’s  equations. Pearson’s correlation coefficients were calculated to measure, in boys and girls, the relationships between anthropometric variables assessed at 7 and 10 years of age, and between changes which occurred in BMI/iBMI and body fat at the follow-up examination.
Descriptive statistics of anthropometric characteristics
Triceps skinfold (mm)
Subscapular skinfold (mm)
Suprailiac skinfold (mm)
Calf skinfold (mm)
Body fat (%)
Triceps skinfold (mm)
Subscapular skinfold (mm)
Suprailiac skinfold (mm)
Calf skinfold (mm)
Body fat (%)
In boys, from 7 to 10 years, being overweight increased (9.9 to 25.6 % of cases), while being underweight (4.1 to 2.3 %), normal weight (77.9 to 70.9 %) and obese (8.1 to 1.2 %) decreased. In girls, being underweight increased (6.9 to 16.8 % of cases), being overweight was similar (both 17.9 %), while being normal weight (71.1 to 64.2 %) and obese decreased (4.0 to 1.2 %). Tracking rates of being overweight were 70.6 % in boys and 61.3 % in girls, while tracking rates of obesity were 14.3 % in boys and 28.6 % in girls.
In both sexes, besides a clear trend to move toward the lower weight categories shown by overweight and, even more evidently, obese individuals, it was observed that a certain portion of normal weight individuals moved toward being underweight or overweight, although with a different trend between the genders. Specifically, passing from the normal category to the overweight was particularly evident among boys: 16.4 % of normal weight boys at 7 years moved to the immediately higher weight category at 10 years. Instead, girls showed a trend to move toward underweight: 15.4 % of previously normal weight girls become underweight at 10 years (Fig. 1).
Correlations between anthropometric characteristics at 7 and 10 years tended to be strong. Height showed the highest correlations (r = 0.95 for both boys and girls, both p < 0.001), followed by weight (r = 0.85 for boys and r = 0.88 for girls, both p < 0.001), whereas both BMI and iBMI showed an r value of 0.85 for girls and 0.79 for boys (both p < 0.001). Regarding skinfold thickness, in girls, the subscapular showed the highest correlation value (r = 0.75, p < 0.001), followed by the calf (r = 0.71, p < 0.001), suprailiac (r = 0.70, p < 0.001) and triceps (r = 0.43, p < 0.001). The calf skinfold thickness presented the highest values in boys (r = 0.64, p < 0.001), while the other skinfold thickness showed values in the 0.55–0.6 range (all p < 0.001). The body fat percentage showed correlation coefficient values of 0.66 (p < 0.001) and 0.67 (p < 0.001) for boys and girls, respectively.
Changes in BMI occurring between 7 and 10 years of age were well correlated with changes in body fat percentage (girls: r = 0.69; boys: r = 0.65; both p > 0.001). Correlations of similar magnitude were observed between changes in iBMI and changes in body fat percentage (girls: r = −0.67; boys: r = −0.64; both p > 0.001).
The results showed that at 7 years of age, girls presented with the highest proportion of being overweight, while boys presented with the highest proportion of obesity. After 3 years of follow-up, a remarkable increased proportion of underweight children was observed among girls, while an equally considerable increased proportion of overweight children was observed among boys. This trend can be better understood by analyzing the tracking of weight status over time. Indeed, many individuals who presented with obesity and being overweight at the beginning of the study showed a disappearance or a reduction of such disorders at the follow-up. On the contrary, negative variations were observed in normal weight subjects: 5.7 % of girls and 16.4 % of boys belonging to the normal weight category became overweight. This result seems alarming and indicates that more attention must be paid to following children’s development over time. Another relevant aspect emerging from this study is that the risk of becoming underweight is also an important consideration, especially in the European population and particularly among girls. The different tendency of the two sexes to become underweight may be linked to a different perception of one’s body image [37, 38]. Gualdi-Russo et al. , using the sample population from the same geographical area as the present study, reported that both sexes desired a thinner body image, with a higher degree of dissatisfaction among girls than boys. Thus, possible problems with body image perception may be important in the evolution of weight disorders, and problematic cases have to be identified as early as possible and carefully followed over time.
Studies carried out in the USA and Europe that tracked obesity from childhood to adulthood generally found that about one-third of all overweight and obese children remained overweight or obese as adults [19, 39–41]. In the study by Starc and Strel , the percentage of overweight and obese 7-year-olds who became overweight or obese 18-year-olds was higher (over 65 % among boys and below 50 % among girls) than in this study. The prevalence of obesity in the study by Leitão et al.  on school-aged children from childhood to adolescence followed over a 6-year period decreased from 21.9 to 14.8 % in boys, while in girls it increased from 14.3 to 19.5 %. The percentage of children who went from being obese to being non-obese was more than threefold higher in boys (9.7 %) than in girls (3.0 %). The authors also reported a moderate tracking of obesity in both sexes. Juhola et al.  found a significant and strong tracking between childhood and adulthood BMI values. Wright et al.  found a strong tendency for children to remain in the obese BMI category over time (75 %), as well as a tendency for those in the overweight category to progress toward obesity (63 % were still overweight and 16 % had progressed to becoming obese). Reilly et al. , valuing the probability that overweight children often progress to obesity, reported that 34 % of overweight children at age 7 became obese at age 13 years. Our data, in comparison with the longitudinal data of Wright et al. , carried out on age-matched children, showed that a lower percentage of obese individuals were still obese at the follow-up. Tracking of overweight girls in our study was instead comparable with that of Wright et al. , while boys in our study presented a higher percentage of tracking. Moreover, it is important to notice that in our study, a progression from overweight to obesity was not observed. However, Valerio et al. , studying southern Italian children, reported tracking rates of overweight (73 %) and obesity (80 %) higher than those observed in the present sample. These differences can be explained by the fact that in Italy, a strong regional gradient of overweight and obesity has been observed, with the highest values in the central and southern regions [44, 45].
A strength of the present study is that although the period of follow-up was relatively short, it provides important information on the tracking of measurements other than BMI, as the majority of previous large-scale studies, especially in Italy, focused on BMI. We observed that not only BMI, but also some of the examined skinfolds and skinfold-based body fat percentage showed good tracking, in agreement with previous studies [46–48]. The utilization of BMI as an indicator of adiposity is often questioned in the literature, since it simultaneously evaluates both the fat and fat-free mass. According to Demerath et al. , BMI percentile changes may not accurately reflect changes in adiposity in children over time, particularly among children of lower BMI, in whom the differences in BMI percentile are largely attributable to differences in fat-free mass rather than fat mass. A discrepancy between obesity classification methods was reported by Leitão et al. . In the study by Leitão et al. , the prevalence of obesity based on body fat percentage was more than twice as high as that obtained by BMI at age 9 years and fourfold higher than at age 15 years. In the present study, however, the tracking of each skinfold and of skinfold-based body fat was correlated to that of BMI, thus not invalidating the use of BMI when analyzing the tracking of adiposity in children. Together with BMI, we analyzed the iBMI, because this index was suggested to be more biologically sound than the BMI and to represent a better proxy of fat mass . However, we observed no substantial difference between BMI and iBMI in tracking, or in the relationships between changes of these indices over time and changes of skinfolds or skinfold-based body fat. Therefore, considering iBMI would not add any additional information than using BMI when analyzing the tracking of body composition and adiposity in children. Anyway, the relationship between BMI and the other examined indices do not seem to challenge the usefulness of accurate surveys examining aspects other than BMI, since the growth patterns are more understandable when there are more data available for analysis.
Future research on later age groups will be necessary over the next few years to validate the results obtained in the present study.
The research was supported by grants PRIN 2009—MIUR.
Conflict of interest