The prevalence of overweight and obesity was 31% among Azorean youth 11–15 years of age, for both sexes, while the prevalence of over-fatness was 27%. About 64.3% of Azorean adolescents were classified as having a normal weight and normal fat, while 239 youth (19.8%) were classified both over-fat and overweight/obese. The two classifications thus agreed in 84.2% of cases (boys: 85.3%; girls: 82.9%). Among youth discordant for fatness and weight status, 82 youth (6.8%) were over-fat but classified as having a normal weight status by the BMI, while 111 youth (9.2%) were classified as having an acceptable level of fatness but were classified overweight or obese according to the BMI. The current study (like FITNESSGRAM) used skinfolds as the field method to estimate body fatness. Skinfolds had proven to be one of the most effective field methods for estimating body fatness with standard errors of estimate of 3 to 4% body fat. Although BMI is often viewed as an alternative to estimates of body fatness, however, it is not as effective in identifying moderately over-fat children
The observed prevalence of overweight and obesity in the present study of Azorean youth, aged 11–15 years, was 33% in males and 28% in females (Table
2). This prevalence was higher than estimates for a larger sample of Portuguese youth, 10–18 years of age, which was 23% for males and 21% for females
. Both surveys used the IOTF criteria. Studies differed, of course, in consent rate, sampling, and so on, but the possibility of differences in lifestyle between the islands and the mainland needs consideration. A recent study noted an inverse association between intake of milk intake with BMI and percentage of body fat in Azorean girls
The relationship between variability in chronological age, maturity status and CRF, on one side, and inter-individual variability in the combined prevalence of overweight/obesity and over-fatness, on the other side, was also considered in the present study. Percentage of predicted adult height attained at the time of measurement was used as a non-invasive parameter of biological maturity status. Youth of both sexes classified as on-time (average) and early maturing had a higher probability of being classified as overweight or obese. In contrast, only early maturing boys were more likely to be classified as over-fat, while both on-time and early maturing girls were more likely to be over-fat compared to late maturing boys and girls, respectively. These results were consistent with an earlier study of Portuguese boys and girls 10–15 years of age; early maturation based on clinically-assessed stages of puberty (genitalia for males, breasts for females) was associated with an increased risk of overweight/obesity
However, it is worth highlighting that the mentioned study
 grouped the sample using the quartiles of the decimal age adjusted for stages of sexual maturation and sex. For example in stage 1, children were considered as ‘early maturers’ if they were in the first quartile of the decimal age for that stage and sex. The same procedure was used for the other stages. Those in the fourth quartile of the decimal age were considered as ‘late maturers’. Afterwards, in the data analysis, differences in the prevalence of overweight among the quartiles of sexual maturation were tested using a Chi-square test, separately for boys and girls. Erroneously, the literature often interprets sexual maturation given by stages of pubic hair interpreted as an indicator of timing.
Current stature expressed as a percentage of the predicted mature value has been used in studies of PA
[36, 37]. A model for adolescent involvement in PA that incorporates individual differences in timing and tempo of biological maturation has been recently presented in a cross-sectional sample of adolescents
. The decline in PA with age was related to physical and physiological changes associated with pubertal maturation and growth spurt, which include changes in body composition and body proportions.
Low CRF has been associated with the presence of cardiovascular and metabolic risk factors in youth
[38, 39]. Low levels of CRF in childhood and adolescence have also been associated with increased cardiovascular risk in adulthood
. About 53% of Azorean youth in the current study were classified as unfit in the 20-m shuttle run test. More importantly, unfit boys and girls were more likely to be overweight/obese than aerobically fit youth. In cross-sectional
 and longitudinal
 analysis from Portuguese youth, low levels of CRF were associated with increased BMI and body fatness. Note, however, that biological maturation was positively related to both CRF, especially in boys, and fatness during adolescence
[16, 42]. These relationships need further studies. Allowing for the limitation of the current study, boys and girls who were simultaneously over-fat and had a higher BMI tended to be more advanced in estimated maturity status and were also less fit aerobically (Table
4). BMI may temporarily be confounded by inter-individual differences in the timing of the growth spurts of height and weight. This, unfortunately, was not captured in cross-sectional surveys. As such, use of a multi-method approach may help to avoid potentially erroneous classification of BMI for chronological age during adolescence. These issues influence the diagnostic accuracy of BMI to diagnose overweight/obesity status among adolescents. The issue has been examined in adults
 and more recently in youth
Previous studies with Azorean adolescents indicated a significant association between higher levels of PA and a lower occurrence of overweight and obesity, and metabolic risk
[6, 44]. The results of these studies and also of the present study claim a need for promotion of PA programs in Portuguese youth. Organized sport may be a relevant correlate. Portuguese adolescents spent 11% to 13% of total daily energy expenditure in organized sports, which corresponded to 35% to 42% of the moderate-to-vigorous portion of daily energy expenditure
. Unfortunately, many currently organized sport programs do not fit the interests and readiness level of the overweight and obese youngster.
Although the results highlighted an association between low CRF, advanced biological maturation, weight status and over-fat in youth, several limitations of the study should be noted. First, the cross-sectional design of the current study precludes causal inferences to be drawn. Second, PA was not assessed. Third, a non-invasive indicator of maturity status was used. Although the results were consistent with several studies using secondary sex characteristics, validation of attained height expressed as a percentage of predicted adult height is needed.
However, the observed associations between sexual maturation and obesity presented in the current study may have important implications for the classification, management, and prevention of child and adolescent obesity. There is a risk to uncritically apply BMI cutoffs for classifying child and adolescent overweightness and obesity without fully recognizing their limitations and the potential misclassifications. To our knowledge, none of the existing anthropometry references are able to account for maturation status.