Indian Pediatrics 2001; 38: 1217-1235  

Physical Growth Assessment in Adolescence

K.N. Agarwal, A. Saxena*, A.K. Bansal and D.K. Agarwal+

From the Department of Pediatrics, University College of Medical Sciences, Delhi 110 057, India and *Sanjay Gandhi Postgraduate Institute of Medical Sciences, Lucknow, India and +Department of Pediatrics, Institute of Medical Sciences, Varanasi, India.

Correspondence to: Dr. K.N. Agarwal, D-115, Sector 36, Noida 201 301, Gautam Buddha Nagar, U.P. India. Email: kna_ped@yahoo.com

Manuscript received: May 4, 2001, Initial review completed: May 30, 2001,
Revision accepted: June 15, 2001.

Objective: To describe ponderosity indices body mass index - ‘BMI’ and ponderal index ‘PI’ and skin fold thicknesses (triceps, biceps, subscapular and suprailiac) for affluent Indian school going adolescents. Design: Cross sectional. Methods: Measurements were recorded in healthy affluent school going adolescents in public schools of 12 cities in India (boys=11,863 and girls 7,694). Means and percentiles of ponderosity indices and skinfold thicknesess at yearly intervals were derived for each sex and related to sexual maturity. Results: BMI, PI and skin fold thickness (SFT) were higher in girls. There was lower variability of these parameters with sexual maturity rating (breast/genital development stages) as compared to age, suggesting use of these indices in relation to sexual maturity for assessment of adolescent growth. Conclusion: Pediatricians and endocrinologists can use these indices for assessment of thinness and obesity, in adolescent Indian children, in relation to sexual maturity for the age.

Key words: Adolescence, Body mass index, Ponderal index, Skin fold thickness, Sexual maturity.

ADOLESCENCE contributes >20% of total growth in stature and up to 40-50% of body weight as somatic growth. The spurt in somatic growth which is initiated by the sex hormones is accompanied by sexual develop-ment. However, there is marked variation in timing of these maturational changes, there-fore, relationship of somatic growth with chronological age may be misleading, especially when applied to individuals. To assess somatic growth in adolescence, per-centiles for height and weight and related indices need to be calculated in relation to sexual maturity of the child rather than age.

Morphometeric differences in the timing of height and weight growth among populations particularly Asiatics and South Americans generate different patterns of body ponderosity indices [body mass index (BMI) and ponderal index (PI)] and subcutaneous fatness(1-3). As Asians living in USA have lower BMI than other US populations(2), this could necessitate different cut off’s for the diagnosis of obesity which is a known risk factor for insulin resistance(4), abnormal lipids and lipoproteins(5), elevated blood pressure(3), and adult morbidity and mortality(6). It is likely that Indian adoles-cents at upper extremes of BMI and skin fold thickness (SFT) are also at risk for similar health consequences(7). On the other hand, in India short stature and low BMI may be determinants of concurrent functional impairment due to reduced lean body mass, changes in muscle metabolism and deficiencies in muscular strength and working capacity(8), and reduction in higher mental functions(9,10).

The WHO expert group(11) has recommended that thinness as well as obesity should be evaluated by body ponderosity indices particularly BMI and skin fold thickness (triceps and subscapular). It has recommended 85th percentile of BMI as cut off point for overweight and >95th centile for defining obesity for its association with hypertension and other morbidites in children. The present article describes BMI, ponderal index and skinfold thicknesses in affluent Indian adolescents. The height, weight, BMI and SFT (peripheral-triceps and biceps and central-subscapular and suprailiac) percentiles are also evaluated in relation to the sexual maturity. As values for these indices were not available for Indian adolescents, the calculated BMI, Ponderal index and SFT data will help in serving as a benchmark for 1988-1991, to relate to health consequences and secular trend.

Subjects and Methods

Study Population

The data on affluent school children (collected between 1988-91) used for the study has been presented earlier for other anthropometric measurements(12). The data were collected from 23 public schools of different cities of India, namely Delhi, Shimla, Dehradun and Nainital (North Zone); Bombay (Mumbai), Madras (Chennai) and Udaipur (West South Zone); Lucknow, Allahabad and Varanasi (Central Zone), and Dhanbad and Calcutta (East Zone). Children who attended these schools come from high socioeconomic group (well to do families), where families are well educated and have no financial constraints. Since the data on 19,557 school children were collected from all the four zones of the country, it can, therefore, be expected that the data by and large represent affuent Indian children.

Cross-sectional anthropometeric mea-surements were collected on 11,863 healthy boys and 7,694 healthy girls, aged 4-18 years of age, studying in classes 1-12th and percentiles were calculated(12). Khadgawat et al.(13), evaluated the data and recom-mended it for use as reference for Indian school children.

Anthropometeric measurements was collected using standard techniques(11,12). Weight (Wt) was taken on Chattilon weighing scale (John Chattilon and Sons Inc. N.Y.), with standard minimum clothing. Height (Ht) was measured with an anthropometeric rod, subject standing straight and head held in Frankfurt horizontal plane. Skin fold thick-ness (SFT) – triceps, biceps, subscapular and suprailiac were measured using Lange’s cali-pers (Cambridge Scientific Ind. Cambridge MD, USA) with the subject positioned appropriately for the measurement. Mean of the three measurements was taken for this purpose. All the measurements were taken by the same team of workers. Sexual development was assessed as per Tanner’s recommendations(14). Quality control was maintained by checking intra and inter observer error.

Statistical Analysis

BMI was calculated as weight (in kg)/ht2 (in meter) and Ponderal index was computed as weight (in kg)/ht3 (in meters). Percentiles for BMI, height, weight and SFT were calculated in relation to age and sexual maturity (breast development in girls and genital development in boys) using SPSS version 7.0 for windows.

Results

BMI percentiles and mean (±SD) values are presented in Tables I-II. Girls had marginally higher values than boys in adoles-cence. These measurement increased signifi-cantly with progress in sexual maturity (SMR) as well as age (Table III). In the same age group maturity is showing narrow BMI range of <1.0, e.g., 20.7–21.6 kg/m2 in SMR-5 for 50th percentile. In contrast 13 year old girls have 50th percentile values of 15.4, 17.4 18.9 and 21.6 kg/m2 (range being 6.2) for breast development stages 2, 3, 4, and 5 respectively (Table III). These differences were signficant (p <0.001).

Ponderal index (kg/m3) showed little variation with age, 50th percentile range being 10.9-11.5 for boys 10-18 years of age and 11.6-12.9 kg/m3 for girls 9-17 years of age. The 5th percentile values were 9.0-9.6 and 9.5-10.3 kg/m3 and 95th percentile 15.6-15.5 and 16.4-17.8 kg/m3 for boys and girls, respectively. The girls had higher values (Tables IV & V).

Weight and height percentiles in relation to sexual maturity show that it is important to assess an adolescent’s anthropometery for the sexual maturity attained at that particular age (Table VI & VII). This is apparent for the 50th centile values for height range, i.e., 149.5, 151.9, 153.3 and 155.1 cm in the SMR-stages 2, 3, 4, and 5 respectively, for girls of 13 years of age.

Skin fold thicknesses (tricep, bicep, subscapular and suprailiac) means (±SE) in relation to age were described earlier(12). These measurements as percentiles in relation to the sexual maturity rating showed that 50th centile values do not change for boys. How-ever, in girls 50th centile values were more than double for SMR-5 as compared to SMR-2 for age groups 12-13 years (Tables VIII & IX).

Discussion

Rosner et al.(2) have documented significant ethnic variations in BMI. There was therefore need to establish normal reference for BMI and skin fold thicknesses for diagnostic purposes of obesity and thinness for Indian children. Secondly, well nourished Indian children over the past two decades are showing an increasing trend in somatic/physical growth as established by anthro-pometeric measurements(12). In Britain new anthropometeric standards were developed in 1990, but less than a decade later it has become evident that these standards no longer reflect the distribution of weight in British school children(15). Ponderal index did not change much during adolescence and we need to explore its utility for an analogous tool like mid arm circumference during preschool period.

Cole et al.(16) generated internationally applicable BMI cut off for overweight and obesity using datasets from six countries. However, in puberty the sensitivity of their curves was affected. Given that ethnic differences exist(2,16), ethnic group specific standards are more appropriate for comparing health-compromised children, especially in our country where the problem is more pronounced for undernutrition than over-nutrition. So far there are no available standards for BMI and ponderal index percentiles for Indian affluent (well nourished healthy) children on an acceptable number of subjects. Kaur and Singh(17) on a smaller sample of well nourished Delhi girls observed higher BMI values as compared to the present study (which has data collected from several towns of India). The NCHS(2) percentiles for age and sex are higher for 5th to 95th percentiles as compared to the present study.

N = Number of children.

N = Number of children.  

Table III__ Body Mass Index (BMI) in Adolescents in Relation to Sexual Development and Age.

SMR = Sexual maturity rating; G = Girls and B = Boys.  

Table IV__ Ponderal Index Mean (±D) and Percentiles for Indian Boys