1.gif (1892 bytes)

Original Articles

Indian Pediatrics 1999; 36:145-156


A.N. Kanade, S.B. Joshi and S. Rao

From the Department of Biometry and Nutrition, Agharkar Research Institute, Pune 411 004, India.

Reprint requests: Dr. (Mrs.) Shobha Rao, Agharkar Research Institute, Law college Road,
411 004, India.

Manuscript received: January 13, 1998; Initial review completed: May 3, 1998;
 Revision accepted: August 31, 1998


Objective: To evaluate impact of undernutrition on various adolescent growth parameters among rural Indian boys. Design: Adolescent boys covering 8-18 yr age group were observed longitudinally for a period of 3 years. Setting: Adolescent boys (n = 673) from seven different villages within 30 to 40km from Pune were studied. Method: Six monthly measurements on weight (upto 50 g) and height (upto 0.1 cm) were recorded and age assessment was done from school records with reasonable accuracy. Results: Stunted and underweight boys were lighter (by 4 kg) and shorter (by 8 cm) at 10 yr of age compared. to their normal counterparts but this difference increased to 12 kg and 10 cm respectively by adulthood. Undernourished boys however, revealed significant height gains at later ages especially beyond 14+ yr, compared to normals suggesting' slow, gradual but continual growth. Undernutrition delayed age at take-off and age at PHV by about 2 yr, and lowered attained height at PHV (by 5 cm) and adult height (by 7 cm). Conclusion: Normal4nd malnourished children from the same rural community show wide differences in their adolescent growth performance. Nutritional deprivation thus seems to affect almost all growth parameters and final adult size too.

Key words: Adolescent growth, Rural boys, Undernutrition.

Children pass through adolescent phase of accelerated growth at widely different chronological ages. As such marked differences are observed between children of same age-sex with regard to age at which spurt begins and ends, as also in magnitude of the peak height velocity (PRV). Although, social class related differences in growth have been documentedO,2) by many, the fact that large variation is seen even within the same community has not received much attention. Differences in dietary intakes alone are unlikely to explain between individual variations in adolescent growth as there is hardly any evidence for the relationship between the two during. adolescence(3,4). It cannot be overlooked that rural children exhibit a wide spectrum of undernutrition throughout. pre- school and pre-pubertal period. Thus children entering adolescence with differing nutritional status are likely to differ greatly in their adolescent. growth performance, resulting in the large between individual variations. 'The iIl1plications of undernutrition in terms of weight and height velocities, and adolescent growth parameters such as peak height velocity (PRV), 'age at PRY, adult height, etc. need 'critical evaluation. Data reported in literature mostly being cross-sectional, don't permit such investigation. The present report therefore, attempts to evaluate impact of undernutrition on various adolescent growth parameters in a longitudinal study carried out over 3 years period.

Subject and Methods

Seven villages around Pune are considered for present investigation. Several departmental projects were carried out in the past 10 yr in these villages which are within 30. to 40
km from Pune. As such good rapport is established with this rural community. All children in the age group 8-18 from the selected villages were considered.. Most boys were attending school and about 15% children were school drop outs who were contacted from house to house, leading to almost 'complete enumeration of boys in this age range. Most villages had school up to 7th standard only, and boys had to go to another village for further education. These children were contacted in their new schools if they had changed their school, in order to minimize loss to follow-up.

Rural Community

All villages represented typical rural community. Thus main occupation of most families (79%) was agriculture. However, in view of small land holding few families (17%) had to work as agricultural laborers. Parents of children in the study had low levels of education, in fact 68% women were illiterate. The habitual diet consisted of cereal, pulse and vegetable while meat, milk or fruits are rarely consumed. Lack of safe drinking water supply and poor hygiene and sanitation was Seen in all the villages. The study villages were thus similar in several aspects and none of the villages was too rich or too poor outwardly.


Simple anthropometric measurements were recorded every six months (:!: 1 week) on all the children, but only annual measurements are considered for analysis presented in this paper. A total of 673 rural boys were thus observed for the period of three years during 1992-1995. Body weight was measured up to 50 g (Avery, India) while height was measured up to 0.1 cm (UNA, India). Boys were asked to stand on wooden platform with bare- feet, having feet parallel and closed together, head held comfortably erect and back of the head touching the upright steel rod. Height was recorded by lowering gently the horizontal metal bar making contact with the top of the head. Inter-observer study was undertaken before starting the main study. Thirteen subjects were measured twice by each of the five investigators (including authors). Interas well as intra-observer variation was observed to be negligible (CV <1 %). Nevertheless, investigator recording a particular measurement was same throughout the study to reduce errors of measurement.

Age Assessment

Special care was taken for assessment of are. Thus school records, village records and also use of specially developed local calendars in case of few, was used for getting reasonable accuracy in age assessment.

Assessment of Nutritional Status

In order to study impact of undernutrition on adolescent growth parameters, conventional indicators based on weight and height were used for assessing undernutrition. The weight for age (W/A) and height for age (H/A) ratios were computed using NCHS(5) standards and W/A was used to classify boys as normal (W/A >=75; combining normal and mild malnutrition) or underweight (W/A <75; representing moderate and. severe malnutrition) given by Gomez et al.(6). Similarly, children were classified using H / A indicator as normal (H/A >=90) and stunted (IDS <90) on the basis of cut-off points suggested in Waterlow(7) classification.

Statistical Methods

Two sample mean tests are carried out using Student's 't' test for comparing means of weights and heights for normal and under- nourished groups of children. In order to estimate various adolescent growth parameters, popular non-linear model suggested by Preece-Baines(8) was fitted to stature growth data of rural boys. Parameters of this model were estimated using an iterative method(9).

The age distribution of study population along with the extent of underweight and stunting in each yearly age group is given in.
Table I. It can be observed that only 15% boys .had normal weight status while 40% boys showed normal height status during adolescence. The magnitude of wasting was more
than stunting in this rural community. The mean BMI (Body Mass Index) values for broad age groups, viz., < 12yrs, 12-14yrs, 14- 16 yrs and 16-18 )yrs were 13.81, 14.35, 15.43 and 16.63, respectively. More than 70% boys had in fact BMI values as low as IS, indicating. frank undernutrition.

 Extent of Malnutrition According to Age Among Rural Boys

Age (yr)
     8- 9       86 73.3 45.3
9-10 114 77.1 34.2
10-11 80 81.3 43.8
11-12 74 81.1 43.2
i2-13 74 85.1 55.4
13-14 67 89.6 ,67.2
14-15 68 89.7 60.3
15-16 45 86.7 60.0
16-17 42 92.9 59.5
17-18 17 76.4 29.4
18-19      6 83.3 16.7

Underweight: Weight for age <75%
Stunted: Height for age <90%

In view of large undernutrition prevailing in this rural community we examined adolescent growth performance of normal, under- weight and stunted boys on the basis of V:1IA and H/A ratios computed for their measurements in the initial round. All the annual measurements on each boy were considered to develop, growth curves shown in Figs. 1 &2. The mean measuremerits in yearly age groups were compared for normal and underweight (Table ll) groups and for normal and stunted (Table Ill) groups.

Growth curve for boys with normal weight status fell below the NCHS median and was
closer to 30th percentile while that for under- weight boys was even below 3rd percentile. Thus, normal and underweight boys from the same community differed widely especially during adolescence. It can be observed that underweight boys had significantly (p <0.01) lower body weights and heights throughout adolescence than normal boys. The difference in the two groups at the age of II yr is about 4 kg for weight and 8 cm for height, but in- creases to almost 12 kg and 10 cm by adult- hood. Underweight boys are thus clearly at a disadvantage with respect to adolescent growth and final adult size. These trends are almost similar if we compare normal and stunted boys (Table Ill).

In order to further examine the impact of undernutrition on growth velocities during adolescence, annual increments were computed as difference between two successive annual measurements for each
boy. The whole year velocities were then assigned to the midpoint of the two examination ages. For example, the mean annual velocities observed for boys in a given age group (say 11-12 yr) were plotted against its mid point (11.5+ yr) and velocity curves were developed (Fig. 3 & 4) using the cumulative data (n
= 400) of all annual velocities. The mean observed velocities are compared for normal and underweight boys in Table IV and for normal and stunted in Table V.


Fig. 1. Growth curves for normal and underweight boys.

It can be observed that the maximum weight Increments for normal (5.14 kg) and underweight boys (4.68 kg) are not significantly different but there was considerable difference in the age at which the peak weight ve10cityoccurs between normal (13.5 yrs) and underweight (14.5 yrs) boys. The peak height velocities, on the other hand, were quite similar for normal (7.01 cm) and underweight (6.64 cm) group, but the difference in the ages at which they occurred was considerable, viz.,12.5 yr Vs 14.5 yr. Comparison of velocities for normal and stunted boys too, shows (Table V) similar estimates of PHV but ages of their occurrences are considerably different. This indicates that more than the growth rates; it is the timing of peak height velocity which is more sensitive to undernutrition. These findings support Tanner' s( 10) observation that. it is the age of onset of adolescent event that is more sensitive indicator than the growth itself.


Fig. 2. Growth curves for normal and stunted boys.

It can be noted further that the delay in the onset of growth spurt among undernourished boys shifts the velocity curve to the right. Moreover, it is interesting to note that under- weight (Table IV) or stunted boys (Table V) continue to have significantly (p < 0.01) higher velocities right upto the age of 17.5 yrs, compared to their respective normals.



Attained (Mean':1:SD) Weights and Heights for Nonnal and Underweight Boys

Age No. of boys   Weight (kg)    Height (cm) 
(yr) Normal Underweight Normal Underweight Difference Normal Underweight Difference
8-9 23 63 21.33 1.3 18.121.22 3.21 * * 123.44.05 116.4 4.66 7.0**
9-10 45 145 23.591.56 19.6865 3.91 ** 128.13.79 121.24.62 6.9**
10-11 59 200 25.962.37 21.591.76 4:37** 133.64.36 125.94.74 7.7**
11-12 52 200 29.133.49 23.552.39 5.58** 139.14.59 130.2 5.37 8.9**
12-13 38 170 33.994.83 26.243.13 7.75** 146.55.81 134.5 6.22 12.0**
13-14 23 170 39.73 3.09 28.81 4.28 10.92** 154.84.09 140.4 7.09 14.4**
14-15 21 174 44.00:t 3.06 32.19 5.09 11.81 ** 159.44.35 146.37.63 13.1 **
15-16 19 144 48.212.83 35.94 5.35 12.27** 162.44.37 151.47.49 1\.0**
16-17 12 110 51.212.61 39.53 5.21 11.68** 165.54.98 156.17.65 9.4**
17-18 9 67 54.434.12 42.30 4.56 12.13** 168.23.48 159.07.03 9.2**
18 4       37 57.935.97 45.264.97 12.67** 170.63.66 160.96.90 9.1**

** p < 0.01
Underweight: Weight for age <75%
Stunted: Height
for age < 90%


Attained (Mean
SD) Weights and Heights for Normal and Stunted Boys

Age No. of boys   Weight (kg)   Height (cm)
(yr) Normal Stunted Normal Stunted Difference Normal Stunted Difference
8-9 47  39 20.121.58 17.611.19 2.51 ** 122.23.49 113.53.00 8.7**
9-10 117 73 21.751.98 18.771.54 2.98** 126.13.48 117.63.28 8.5**
10-11 158 101 23.882.39 20.571.57 3.31 ** 131.13.91 122.33.22 8.8**
11-12 153 99 26.373.18 22.132.07 4.24** 135.94.44 126.13.74 9.8**
12-13 110 98 30.424.31 24.552.46 5.87** 142.75.50 131.54.29 11.2**
13-14 83 110 34.644.55 26.69 3.07 7.95** 149.65.30 136.54.90 13.1 **
14-15 74 121 39.074.58 30.034.09 9.04** 156.0 4.92 142.75.64 13.3**
15-16 61 102 42.674.91 34.205.04 8.47** 160.0 4.46 148.36.29 11. 7* *
16-17 47 75 45.354.44 37.705.51 7.65** 163.74.51 152.86.63 10.9**
17-18 31 45 47.635.55 41.064.63 6.57** 165.44.20 156.46.81 9.0**
18 21 20 49.835.33 42.995.29 6.85** 165.34.08 158.28.02 7.1 **

**p <0.01
Underweight: Weight for age <75%

Stunted.: Height
for age < 90%

Finally, we attempted fitting of Preece- Baines(8) model to mean heights of normal, underweight and stunted boys to have sound estimates of biological parameters of adolescent growth and especially adult size for comparison (Table VI). Although this model was originally developed to fit individual longitudinal growth data, several workers(Il,12) have demonstrated that it is also useful to estimate average population growth of height. It can be observed that underweight and stunting during adolescence delays age at take-off (A TO) by almost 2 to 3 yrs and age at PHV (APHV) by about 2 yrs. Undernutrition seemed to lower PHV only marginally, i.e., by 2 cm. The differences in height at PHV (5 cm) between normal and malnourished groups was observed to remain similar by adulthood. Thus, most of. the differences in final adult size of normal and malnourished boys are already developed even before the onset of growth spurt. These observations are similar to those reported in case of rural Hyderabad boys(13).

Fig. 3. Velocity Curves for normal and underweight boys.


Fig. 4. Velocity curves for normal and stunted boys


Although, growth is known to be influenced by genetic as well as environmental factors, the latter are of greater importance. For, it is only under .optimal levels of environ- mental conditions that genetic factors become effective. Rural Indian children living in poor environmental conditions reveal varying degrees of wasting and stunting. It is, how- ever, not known how children from the same socio-economic class but witnessing differing types of malnutrition, vary in their adolescent growth performance and this was the main aim of the present study.

It was observed that differences in attained weights and heights for normal and under- weight or stunted boys were even larger in magnitude than that for rural-urban groups or low and high socio-economic groups from urban populations reported in Indian studies( 14- 17).


SD) Velocity in Weights and Heights for Normal and Underweight Boys

Age No. of boys   Weight Velocity (kg/yr)   Height Velocity (cm/yr)
(yr) Normal Underweight Normal 'Underweight Difference Normal Underweight Difference
8-9 20 58 1.920.93 1.610.72 0.31(NS) 5.030.64 4.750.91 0.28 (NS)
9-10 43 133 2.120.78 1.740.75 0.38** 4.830.77 4.650.82 0.18 (NS)
10-11 54 188 2.591.59 1.960.84 0.63** 4.851.13 4.50 0.79 0.35**
11-12 48 174 3.681.80 2.381.28 1.30** 5.721.74 4.881.29 0.84**
12-13 28 155 5.042.29 3.1O1.77 1.94** 7.012.13 5.471.73 1.54**.
13-14 20 153 5.142.14 3.891.97 1.25** 5.42.26 6.252.17 -0.83(NS)
14-15 13 146. 4.342.13 4.682.09 -0.34(NS) 3.652.65 6.642.16 -2.99**
15-16 12 111 2.712.25 4.561.98 -1.85** 2.231.92 5.72 2.33 -3.49**
16-17 8 73 2.310.98 3.991.95 -1.68** 1.491.11 4.142.34 -2.65**
17-18 2 40 0.180.03 3.072.24 - 1.60.10 2.531.90 -
18 1 23 2.40 2.492.09 - 0.0 1.852.33 -


SD) Velocity in Weights and Heightsfor Normal and Stunted Boys

Age No. of boys   Weight Velocity (kglyr) Height Velocity (cm/yr) 
(yr) Normal Underweight Normal Stunted Difference Normal Stunted Difference
8-9 43 35 1.830.95 1.570.49 0.26 (NS) 4.860.76 4.780.96 0.08 (NS)
9-10 111 65 1.960.77 1.610.74 0.35** 4.750.77 4.600.87 0.15 (NS)
10-11 143 96 2.291.25 1.810.66 0.48** 4.620.96 4.510.77 0.11 (NS)
11-12 138 84 2.981.67 2.140.98 0.84** 5.371.56 4.561.04 0.81**
12-13 93 90 4.332.09 2.441.29 1.89** . 6.471.98 4.911.38 1.56**
13-14 69 104 5.002.15 3.291.73 I. 71 ** 6.372.65 6.012.10 0.36 (NS)
14-15 55 104 4.822.00 4.562.14 0.26(NS) 5.692.65 6.772.08 -1.08**
15-16 44 79 3.622.18 4.791.89 -1.17** 3.702.46 6.312.01 -2.61 **
16-17 28 53 2.89:1.61 4.321.92 -1.43** 1.841.22 4.952.13 -3.11 **
17-18 13 29 1.661.55 3.482.34 -1.82** 1.230.96 3.041.90 -1.81 **
18 11 13 1.901.82 2.992.09 -1.09(NS) 1.122.17 2.322.27 -1.20(NS)

** P < 0.01; NS : Not significant, Underweight: Weight for age <75%, Stunted: Height for age < 90%


Adolescent Growth Parameters Estimated by Fitting Preece-Baines Model

Biological Parameteres   Weight for age Height for age
  Normal Underweight Normal Stunted
ATO (yr) 7.65 10.65 7.78 9.95
APHV (yr) 12.61 14.31 13.05 14.38
HTO (em) 119.15 126.52 119.26 119.68
HPHV (em) 147.63 145.06 146.40 141.80
Vel at TO (emlyr) 4.36 4.55 3.90 4.26
PHV (emlyr) 7.58 5.71 7.01 6.00
Adult height (em) 168.62 161.99 166.22 159.83

Underweight: Weight for age <75%
Stunted: Height for age <90%
ATO : Age at take off; APHV : Age at peak height velocity;
HTO : Height at take off; HPHV : Height at peak height velocity

The estimates of peak height velocity for rural boys observed in this study ranged 'between 6.47 to 7.01 cm for different groups. These are similar to those reported(18) for rural Hyderabad boys, and are marginally lower in comparison to that observed for well- nourished Indian(19) boys (8.13 cm) or that for British(20) boys (8.2 cm).

The timing of occurrences of PHV be- tween normal and undernourished boys was studied through velocity curves. Impact of undernutrition on age at PHV using weight for age and height for age or respective z-scores was similar. Thus age at PHV was observed to be delayed by almost two years in case of undernourished. boys compared to normals. Among rural Hyderabad boys too, undernutrition was observed to delay adolescent growth spurt by two years. More importantly, we observed that the delay in onset of growth spurt shifted the velocity curves to the right indicating significant height gain at later ages. The growth in height had not ceased ( < 1 cm) as the height gain at 17+ yrs remained in the range of 3 to 4 cm in our study. In fact, similar gains reported in rural Hyderabad (Group IV) boys appeared even larger (about 6 cm), but did not received their attention. The above observation suggests that the height growth probably continues beyond 18 ysr among undernourished boys.

Our attempt of fitting Preece-Baines model to normal and malnourished boys was further useful to quantify the differences in various adolescent growth parameters of these groups. Brown et al.(21) have used this model for comparing growth of aboriginals in Australia with British boys. Similarly, Dasgupta e( al.(22), studied growth in sitting height and leg height in adolescence for middle class Bengal boys with the help of this model. Satyanarayana  et al.(13). using PB-I model observed that age at PHV was significantly delayed by severe malnutrition in childhood among rural Hyderabad boys. Although, we do not have data on growth during early life, our analysis indicates that even current under- nutrition exerts significant influence on adolescent growth. Thus boys entering adolescence with poor nutritional status have delayed onset of spurt and are likely to con- tinue stature growth even beyond 18 yrs.

The review of Indian data by Sathyavathi et al. (23) indicates that age at PHV for boys from different rural communities varies from 13 to 14.5 yrs. This is more reflective of the fact that types of malnutrition (wasting or stunting) prevalent in different rural communities perhaps differ, and are unlikely to be mere reflection of socioeconomic diversity. The fact that differences in adult sizes among normal and undernourished boys were developed much before the onset of spurt, highlights importance of good nutrition while entering adolescence. In view of this, efforts are necessary to improve nutritional ,status of rural boys, perhaps through short term interventions near age at take off, for achieving better adult size.


Authors are grateful to Dr. A.D. Agate, pi rector, Agharkar Research Institute, for providing facilities to undertake the, longitudinal 'Adolescent Growth Study'. We are also thankful to Dr. (Mrs) Gokhale, Mrs. Desai, Mrs. Kelkar and Mr. Gosavi for their help in the field work.



1. Prabhakar AK, Sundaram KR, Ramanujacharyalu TKTS, Taskar AD. Influence of socio economic factors on the age at appearance of different puberty signs. Indian J Med Res 1972; 60: 789-792.

2. Banik NPP. Semi Io.ngitudinal growth evaluation of children from birth to. 14 year in different socio. economic groups. Indian J Pediatr 1982; 19: 353-359.

3. Durnin JVGA, Lonergan ME, Good J, Ewan A. A gross sectional nutritional and anthropometpic study with an interval of  7 yrs on 611 young adolescent school children Brit J Nutt: 1974; 32: 169-179.

4. Hacket AF. A two. year Iongitudinal study of dietary intake in relation to. the growth of 405 English children initially aged 11-12 years. Ann Hum Biol 1984; 11: 545-553.

5. World Health, Organization Measuring Change in Nutritional Status. Guidelines for Assessing the Nutritional Impact of Feeding Programmes for Vulnerable Groups. Geneva, World Health Organization 1983; pp 63-101.

6. Gomez F, Ramos R, Frenk S, Craviofo J, Chevez R, Vazquez J. Mortality in second and third degree malnutrition, J Trop Pediatr 1956; 2: 77-81.

7. Waterlow JC. Classificatio.n and definition of protein energy malnutrition. BMJ 1972; 2: 566-569.

8. Preece MA, Baines MJ. A new family of mathematical models describing the human growth curve. Ann Hum BioI 1978; 5: 1-24.

9. Snedecor GW, Cochran WG. Statistical Methods. Iowa. The lOWA state University Press, 1967; pp 465-467,419-443.

10. Tanner JM. Growth at Adolescence, Oxford , Blackwell Scientific Publications, 1962.

11. Lindgren GW, Hauspie RC. Heights and weights of Swedish school children born in 1955 and 1967. Ann Hum Biol, 1989; 16: 397- 406.

12. Cameron N, Tanner JM, Whitehouse RH. A Iongitudinal analysis of the growth of limb segments in adolescence. Ann Hum BioI 1982; 9: 211-220.

13. Satyanarayana K, Radhaiah G, Murlimo.han KR, Thimmayamma BVS, Rao. NP, Narsinga Rao. BS. The adolescent growth spurt of height among rural Indian boys in relation to. childhood nutritional background. An 18 years Iongitudinal study. Ann Hum Biol.1989; 16: 289- 300.

14. Sathyavathi K, Agarwal KN, Khare BB, Agarwal DK. The growth pattern of weight and height during adolescence. Indian J Moo Res 1981; 74: 857-865.

15. Sikri SD. A comparative study of height and weight of Government and Public school children of Punjabi population. Indian 1 Med Res 1972; 60: 491-500.

16. Kaul KK, Sundaram KR, Rajput VT. Mukerji B, Khurana O, Phadnis RV, et al. Influence of socioeconomic deprivation on physical and sexual growth during adolescence in school and college boys. Indian 1 Med Res 1982; 75: 624-631.

17. Pushpamma P, Geervani P, Devi NL. Food in-take nutrient adequacy and anthropometry of adolescents in Andhra Pradesh. Indian 1 Med Res 1982; 75: 61-67.

18. Satyanarayana K, Naidu N, Rao BSN. Adolescent growth spurt among rural Indian boys in relation to their nutritional status in early child- hood, Ann Hum Biol 1980; 7: 359-365.

19. Pathmanathan G, Prakash S. Growth of sitting height, subisschial leg length and weight in

well-off northwestern Indian children. Ann Hum Bioi 1994; 21: 325-334.

20. Tanner 1M, Whitehouse RH, Takashi M. Stan- dards from birth to maturity for height, weight, height velocity and weight .velocity: British children 1965 Part II. Arch Dis Child 1966; 41: 613-635.

21. Brown T, Townsend GC. Adolescent growth and height of Australian Aboriginals analyzed by Preece-Baines function: A longitudinal study. Ann Hum Biol 1982; 9: 495-505.

22. Dasgupta P, Das SR. A cross-sectional growth study of trunk and limb segments of the Bengali boys of Calcutta. Ann Hum Biol 1997; 24: 363-369.

23. Sathyavathi K, Agarwal KN. Adolescent growth studies - Part B. Growth characteristics. Indian Pediatr 1979; 16: 271-280.



Past Issue

About IP

About IAP



 Author Info.