Update Of Growth Percentiles for Children of an Iranian Population

 

Shams Vazirian MD^•, Abolfazl Sedighnezhad MD

 

Department of Pediatrics, Kermanshah University of Medical Sciences, Kermanshah, Iran

•Correspondence: S. Vazirian MD, Department of Pediatrics, Razi Hospital, Kermanshah University of Medical Sciences, Kermanshah, Iran. Fax: +98-831-367985.

 

---

Background – Growth assessment is an essential component of pediatric health surveillance because almost any problem within physiologic, interpersonal, and social domains can adversely affect growth. The growth chart is the most powerful tool in growth assessment. Standard growth charts are based on the data collected by the American National Center for Health Statistics (NCHS).

Growth percentiles require periodic revision because of the changes in the ethnic mix of the population as well as socioeconomic and environmental conditions. This paper describes the new reference percentile curves for weight, height, and head circumference in an Iranian population between birth and 6 years of age.

Methods – Fifteen-hundred and forty children (808 boys and 732 girls) were studied using cross-sectional method. Based on the primary data, 5 percentiles (5, 25, 50, 75, and 95) were determined for each sex and age group separately. The smoothed curves were then prepared. The smoothed curves of the 50^th percentiles in our study compared to those of NCHS. 

Results – Using Run test, there were no significant differences between the percentiles found in our study and those in NCHS except for three groups: the height of males at birth (p = 0.0268), the weight of females at 18^th month (p = 0.0456), and the weight of males at 4 ½  years of age (p = 0.0109). However, after 36^th month, Iranian children were found to be generally lighter and smaller than the subjects in NCHS references.

Conclusion – The new reference smoothed curves are similar to NCHS curves.
The means, NCHS/WHO standards are appropriate for growth assessment in our
 community in this range of age (0 – 6 years). Nevertheless, for determination of standard growth charts especially for weight and height, more extensive studies on other age groups and in different locations of Iran are recommended.

 

---

Keywords · growth charts · head circumference · height · percentiles · weight

 

Introduction

 

nowledge of the normal growth pattern of children is essential for detecting and preventing pediatric diseases by recognizing overt deviations from normal patterns. Deviations in growth patterns are nonspecific yet important indicators of serious medical disorders. They often provide the first clue to learn that something is wrong, at times, even when the parents do not suspect a problem.¹

The most important growth criteria are anthro-

pometric indices (weight, height, and head circumference) and the most powerful tool in growth assessment is the growth chart. Thus, an accurate measurement of height weight and head circumference should be obtained at every health supervision visit.²

The standard growth charts are based on the data collected, from 1963 to 1975, by the American National Center for Health Statistics (NCHS). New growth charts are scheduled to be released in 1999 based on a nationally representative sample collected from 1988 to 1994 as a part of the National Health and Nutrition Examination Survey (NHANES-III).

The NCHS charts have been accepted by WHO as the international standard of growth for the first 5 years of life² but, over the last decade, these references have become more and more out of date due to secular trend in body size. Therefore, age-related reference ranges have received considerable attention in recent years.^{3 – 8}

Growth varies from population to population, hence extrapolating growth patterns from one population to another may either over- or underestimate excessive or deficient growth.⁹ The controversy over whether or not the growth standards for children developed in Europe and North America are universally applicable, appears now to be settled in favor of those who maintain that they are. Thus, growth percentiles require periodic revision because of the changes in the ethnic mix of a population and socioeconomic factors. This paper presents growth percentiles for children of an Iranian population between birth and 6 years of age.

 

Patients and Methods

 

Using descriptive cross-sectional method, parameters of growth (weight, length/stature, and head circumference) were determined in a 3-month period from birth to 36 months of age and a 6 month period between the ages of 3 and 6 years. For infants, two examiners measured their lengths with the infant supine on a measuring board, and for older children, the stature was measured with the child standing on a stadiometer. Head circumferences were determined only between birth and 3 years. 

Fifteen-hundred and forty children (808 boys and 732 girls) without any previous medical problems and no specific malformations were studied. The sampling method was random strata cluster. The age of the subjects were expressed as the number of months whose days were complete (complete months) and percentile values were derived accordingly, i.e. using the WHO convention that the age of each subject is recorded as the number of months whose days are complete; for example, 3 months and 29 days was recorded as 3 months. Five percentiles (5, 25, 50, 75, and 95) were separately determined for each criterion of both sexes.

The growth curves were then drawn for each age and sex group. Using Run test, our study’s percentiles were compared to those of NCHS; the probability value (p value) was calculated for determination of significant differences.

 

Results

 

There were 808 weight and length/stature measurements for boys and 732 for girls between birth and 6 years of age. There were 423 head circumference measurements for boys and 393 for girls between birth and 3 years of age. The percentiles are shown in Tables 1 and 2. According to the determined percentiles, the smoothed curves for the 5^th, 50^th, and 95^th percentiles were drawn (Figures 1 to 6).

A comparison of the smoothed curve of the 50^th percentiles for each parameter with that reported by NCHS are shown in Figures 7 to 12. There are no differences based on a visual comparison of the growth curves in the period between birth and 36 months during which only small deviations are seen. However, after 36 months of age, the curves of weight and length/stature of both sex groups drop slightly. When compared to NCHS/WHO reference, the heights and weights of Iranian children were lower in both sexes, but followed a similar pattern to that of weight and height for age in NCHS/WHO reference.

Using Run test, there were no significant statistical differences between the percentiles of our study and those of NCHS (p > 0.05) except for three groups: the length of males at birth (p = 0.0268), the weight of females at 18 months (p = 0.0456), and the weight of males at 4 ½ years (p = 0.0109).

 

Discussion

 

Nowadays, the universal use of NCHS/WHO reference for all populations is controversial. Sullivan et al (1991) outlined the minimum criteria for the development of reference curves for specific populations. However, as well as recommending the NCHS/WHO international reference, they suggested that "in developing countries, the resources needed to produce a local growth reference might be more effectively used to meet other public health needs".¹⁰ Goldstein and Tanner strongly argued that developing countries, in particular, should create their own standards for clinical use.¹¹ These beliefs explain the importance and necessity of our study.

Our growth reference consists of the percentiles for the weight, length/stature, and head circumference of under 6-year-old children in a middle class Iranian population. As we expected, a comparison of the boys’ and girls’ percentiles revealed larger values for boys.

Comparing our percentiles with those of NCHS, we found no significant difference. Although, with these provisos, the reference sample is representative of a part of Iranian population and hence it is a growth reference and not a growth standard, but the NCHS standard is useful and appropriate for evaluation of growth assessment in our community. Thus, more extensive studies are recommended in order to determine standard growth charts for Iranian children.

After an exhaustive review of growth studies worldwide, Eveleth et al declared that the growth patterns of healthy populations in different parts of the world are the same (at least up to 5 years of age) and concluded that they should be represented by a universal standard.¹²

Graitcer et al compared the growth indices of privileged groups of children in Haiti, Togo and Egypt with NCHS reference; they concluded that NCHS growth standards are appropriate for measuring child growth in developing countries;¹³ this is in agreement with the results of our study. Guaran et al represented the growth percentiles for infants born in an Australian population during the 1980s and emphasized that birth-weight standards should be updated every 5 – 10 years.

 British growth reference percentiles were determined by Cole et al in 1990 and turned out to be rather different from NCHS standards.¹⁴

Mohammadi et al (1997) compared the growth indices of a group of children in Tehran, Iran with those of NCHS and found that the head circumference percentiles of his subjects in the first year of life was similar to NCHS percentiles.¹⁵ Talebian et al achieved similar results in Kashan, Iran in 1998.¹⁶ Both of these studies demonstrated the same facts as our study.

In 1998, Sharif et al compared the 50^th percentiles smoothed curves for the weight and length of breastfed infants (in the first year of life) in Kashan, Iran with those of NCHS and concluded that the curves they found were similar to NCHS curves up to 3 – 4 months of age, though they dropped gradually afterwards. Weight values dropped more prominently than length values.¹⁷

Hams P et al concluded that although there are no significant differences for head circumference percentiles in different areas, there is not a unique standard pattern for weight and height.¹⁸ Therefore, according to our results and other studies in different areas of Iran, we can use the head circumference standard for Iranian children with confidence. However, due to the disagreement among the results of different studies, we recommend more extensive studies to determine standard growth charts in Iranian children.

 

Acknowledgment

 

We are grateful to Dr. A. Seyedzadeh, Dr. A. Biglari, and A. Hashemian for their genuine support.

            

References

 

1       Howard R, Foye J. Growth and behaviors in pediatrics. In: Behrman RE, Kleigman R, eds. Nelson Essentials of Pediatrics. 3rd ed. Philadelphia: WB Saunders; 1998: 1 – 20.

2       Needlman RD. Growth and development. In: Behrman RE, Kleigman R, Janson HB, eds. Nelson Textbook of Pediatrics. 16th ed. Philadelphia: WB Saunders; 2000: 23 – 65.

3       Healy MJ, Rasbash J, Yang M. Distribution and free estimation of age-related percentiles. Ann Hum Biol. 1988; 15: 17 – 22.

4       Cole TJ. Fitting smoothed percentiles curves to reference data. J Royal Stat Soc. 1988; 151:
 385 – 418.

5       Pan HQ, Goldstein H, Yang Q. Non-parametric estimation of age-related percentiles over wide age ranges. Ann Hum Biol. 1990; 17: 475 – 81.

6       Thompson ML, Theron GB. Maximum likelihood estimation of reference percentiles. Stat Med. 1990; 9: 539 – 48.

7       Pan HQ, Goldstein H, Gi G. A two-level cross- sectional model using grafted polynomials. Ann Hum Biol. 1992; 19: 337 – 46.

8       Cole TJ, Green PJ. Smoothing reference percentiles curves: the LMS method and penalized likelihood. Stat Med. 1992; 11: 1305 – 19.

9       Guaran RP, Wein P, Sheedy M, et al. Update of  growth percentiles for infants born in an Australian population. Aust N Z J Obstet Gynaecol. 1994; 34: 39 – 50.

10    Sullivan K, Trowbridge F, Gorstein J, et al. Growth references. Lancet. 1991; 337: 1420 – 1.

11    Goldstein H, Tanner JM. Ecological considerations in the creation and the use of child standards. Lancet. 1980, 1: 582 – 5.

12    Eveleth PB, Tanner JM. Worldwide Variations in Human Growth. 2nd ed. Cambridge [England]; New York: Cambridge University Press; 1990.

13    Graitcer PL, Gentry EM. Measuring of children: one reference for all. Lancet. 1981, 2: 297 – 9.

14    Cole TJ, Freeman JV, Preece MA. British 1990 growth reference percentiles for weight, height, body mass index, and head circumference fitted by maximum penalized likelihood. Stat Med. 1998; 17: 407 – 29.

15    Mohammadi M. Do the Iranian growth charts are similar to NCHS curves? Iran J Ped. 1997; 10: 17 – 23.

16    Talebian A, Azimi A, Sharif M. Growth rate of head circumference in first year of life in breastfed infants in Kashan, Iran.  Feiz J Sci Res. 2000; 14: 47 – 53.

17    Sharif M, Azimi A. Growth rate of weight and height in first year of life in breastfed infant in Kashan, Iran. Feiz J Sci Res. 2000; 14: 37 – 46.

18    Hams P. The state of world’s children. United Nation Children's Fund (UNICEF). WHO Bull. 1994: 82 – 4.

---

AIM Home | Table of Contents