1000

Asia Pacific J Clin Nutr (1997) 6(4): 265-272

The assessment of malnutrition in children under 3 years of age in Sichuan areas of China using two different growth standards

Mao Meng¹, MD, Qian Yougiong¹, MD, Xu Bo¹, MD, Zak Sabry², PhD, Liu Zhen¹, MD, Li Giang¹, MD, Zhen Deyuan¹, MD, Liu Zhenyue¹, MD, Tang Zheyuan¹, MD

 

¹Department of Pediatrics, West China University of Medical Sciences, Chengdu, Sichuan Province, PR China²Public Health Nutrition, School of Public Health, University of California, at Berkeley, California, USA

---

6,666 children under 3 years of age were selected from 231 villages in 25 townships of the Sichuan Province using a stratified random sampling design. The subjects consisted of 3,356 boys (50.3%) and 3,310 girls (49.7%); 2,510 from plain areas (37.8%),1,857 from hill areas (27.9%), and 2,299 from mountain areas (34.4%). All the children in this study had four anthropometric variables measured: body weight, length, head circumference (HC), upper-arm circumference (UAC). Three anthropometric indicators were used to measure the prevalence of malnutrition: weight-for-age (underweight), length-for-age (stunting) and weight-for-length (wasting). The prevalence of malnutrition and Z scores were calculated for male and female, respectively, by age to compare the Chinese rural children with the Chinese standard (1985) and NCHS Standard. The prevalence of moderate plus severe underweight, stunting and wasting were 20.8%,40.7% and 1.78% in boys and 19.5%, 38.3% and 2.26% in girls. The prevalence of malnutrition also varied with age. The group with the highest prevalence of malnutrition was in the first year of life, with weaning time proving to be a critical period influencing child growth and development.

Z score indicated that the growth and development of Chinese rural children was behind the Chinese Urban and NCHS populations, especially after 6 months of age.

Furthermore, the mean head circumference and upper arm circumference was lower in mildly malnourished children in each age gr 1000 oup than in normal children, and much lower still in moderately and severely malnourished children. The influences of malnutrition were also manifested on bone and muscle development.

Key words: malnutrition, preschool children, growth standards, Sichuan, rural China, plains, hills, mountains

---

Introduction

A global assessment of malnutrition in children under five has been completed by UNICEF in 1990¹. It was based on national representative data from 76 developing countries, which did not include China. The study indicated that about 36% of children under five in the developing world (excluding China) are seriously malnourished in overall terms of their weight-for-age; about 39% are stunted in terms of their height-for-age, and 8% are wasted in terms of their weight-for-height.

Growth monitoring in children has proven to be a valuable tool in the assessment of their nutritional status and in the evaluation of efforts to alleviate child malnutrition¹. The UNICEF study is, therefore, of particular value both globally and at the national level since the data may be taken to reflect the nature and magnitude of the problem of malnutrition.

China’s population exceeds 1.1 billion, with nearly 70 million under 3 years of age, 85% of whom live in rural areas. Data on prevalence of malnutrition in children under 3 years of age in rural Chinese areas are limited. Only a few studies related to this area of interest have been done^2-4.

This paper deals with an epidemiological investigation on the prevalence of malnutrition and malnutrition risk factors among children under 3 years of age. The study was carried out between March and May of 1990, in the rural areas of Sichuan Province, which is the largest province of China in both population and agricultural production. A set of systematic sensitive parameters were obtained in this study to provide a useful data base on the nutritional status and the prevalence of malnutrition and malnutrition risk factors among Chinese rural children. The anthropometric indicators of malnutrition were: weight-for-age (under-weight), length-for-age (stunting) and weight-for-length (wasting). Low weight-for-age is taken to represent a composite measure of stunting and wasting. Low length-for-age (stunting) stems from a slowing in the linear growth of the child and results in a failure to achieve the expected level of stature of the reference population. Low weight-for-length relative to the level in the reference population in indicative of a state of wasting. This paper will describe the prevalence of malnutrition of children under 3 years of age in the Sichuan rural areas of China, using Chinese national⁵ and NCHS standards⁶.

Population sample and methods

Population sample

The sample consisted of 6,666 children under 3 years of age. selected from 231 villages in 25 townships of the Sichuan Province, using a stratified random sampling design. The subjects consisted of 3,356 boys (50.3%) and 3,310 girls (49.7%), 2,510 from plain areas (37.8%), 1,857 from hill areas (27.9%), and 2,299 from mountain areas (34.4%).

Methods

All the children in this study had four anthropometric variables measured: body weight, length, head circumference (HC), upper-arm circumference (UAC). All the somatic measurements were carried out by specially trained 1000 team members following a physical examination of the children and the administration of a questionnaire to their parents.

For the body weight measurements, a Chinese level Balance, made in Factory of Scales in Chengdu China, with the scale measuring up to a maximum 20kg with increments of 50g was used. With this type of level balance the child was placed in a specially designed hanging ‘bag’. The reading was taken to the nearest 50g. Body weight was obtained on children without clothing or diaper.

The length measuring scales were made in Italy, and donated to Chengdu Nutrition Center by the Italy Association. They were in the form of special beds for child length measurement. The children were measured following the procedures adopted by WHO^6,7.

The head circumference and upper-arm circumference were measured using soft cotton tapes. The reading was taken to the nearest 0.1 cm. To maintain accuracy, each tape was discarded after measuring about 100 children.

About 320 medical and health workers and fifth year medical students, from the Department of Pediatrics at the West China University of Medical Sciences and local hospitals, participated in the survey teams. All of them had been strictly trained and certified before starting the survey. Each team member was specially trained for one variable only, in order to build expertise and reduce errors. Each measurement was obtained by two examiners. The measurement manager remeasured randomly selected subjects to ensure the quality and measurements.

The Kappa Test was used to describe agreements between weight and length measurements obtained by two different observers⁸. For weight, a difference was defined as exceeding 50g and for length, 1.0cm. Based on an interpretation of Kappa index where > 0.75 is excellent, 0.74-0 50 good and 0.49-0.4 fair, the measurements in this study were considered acceptable. The Kappa index for weight was > 0.66 and for length >0.65.

The data were entered into the computer at the School of Public Health, West China University of Medical Sciences (1990 and 1992) and computer-managed and analysed using SAS package at the School of Public health, University of California at Berkeley (1993). The prevalence of malnutrition and Z scores^9,10 were calculated for male and female, respectively, by age to compare the Chinese rural children with the Chinese and the NCHS standards.

The anthropometric indicators of malnutrition were: weight-for-age (underweight), length-for-age (stunting) and weight-for-length (wasting). Low weight-for-age is taken to represent a composite measure of stunting and wasting. Low length-for-age (stunting) stems from a slowing in the linear growth of the child which results in a failure to achieve the expected level of stature of the reference population. Low weight-for-length relative to the level in the reference population in indicative of a state of wasting. The reference populations used in this study are those of the Chinese standard⁵ and the NCHS standard⁶. The cut-off points of the three indicators are defined in terms of the standard deviation of the two reference populations. The actual measurements -1 to -2 SD of the reference population reflect a mild state of malnutrition; those -2 to -3 SD moderate malnutrition, and < -3 SD severe malnutrition.

Results

The children were divided into 4 groups by age: from 1 to 6 months, then to 12 months, then to 24 months and then to 36 months. The malnutrition prevalence rates were expressed as a percentage of individuals below minus one (mildly malnourished), two (moderately malnourished) or three (severely malnouris 1000 hed) standard deviations from the median value of each of the two reference populations (Chinese and NCHS). The results of these for the weight-for-age, length-for-age, and weight-for-length, as related to the Chinese and the NCHS standards, are shown in Tables 1 to 6. The results for girls were similar to those for boys.

The prevalence of malnutrition by weight-for-age (underweight), relative to the Chinese and the NCHS standards, are presented in Tables 1 and 2, respectively. Although the standards used for data analysis gave divergent values of magnitude, both consistently showed that the prevalence of malnutrition increased with age. Of the males under 6 months, 41.8% and 9.9% were classified as malnourished (Including mild, moderate or severe malnutrition. -1 to -2 SD of the reference population reflect a mild state of malnutrition; those -2 to -3 SD a moderate malnutrition, and <3 SD a severe malnutrition), by the Chinese and the NCHS standards, respectively. By comparison, the prevalence in those aged 24-36 months of age were 50% and 71.5%.

It may be observed that the prevalence of moderate underweight at each age group, relative to the NCHS standard, was much higher than that defined by the Chinese standard. However, there was not a great difference in the prevalence of severe underweight, when related to either of these two reference populations.

The prevalence of underweight defined by the NCHS standard increased more drastically with age, progressing from 9.9% to 71.5%, than that defined by the Chinese standard (from 41.8% to 50%). The most sensitive age group to the use of a specific standard was the <6-month old (9.9% and 41.8%, NCHS and Chinese, respectively). Only in this age group did the NCHS standard diagnose considerably fewer underweight children than did the Chinese standard.

The Z scores of the weight-for-age values defined by the Chinese and the NCHS standards are presented in Figures 1a and 1b, for boys and girls, respectively. It may be readily observed that before 8 months, the Z score values of weight-for-age of the Chinese rural children, relative to the Chinese standard, are much smaller than those defined by NCHS Reference. The decline in values defined by NCHS standard in the rural sample appears to continue until about the end of the third year of life and get much smaller than those values defined by the Chinese standard.

Table 1. Prevalence of malnutrition in Chinese rural children < 3 years of age based on weight-for-age of Chinese standard population.

1000

Months	N	>+2SD		+1SD-+2SD		SD-+1SD		-1SD--2SD		-2SD--3SD		<-3SD
		N	%	N	%	N	%	N	%	N	%	N	%
Boys
-6	465	2	0.4	18	3.8	248	53.3	142	30.0	39	8.4	16	3.4
-12	616	3	0.5	30	4.8	318	51.6	218	35.4	35	5.7	12	1.9
-24	1107	3	0.3	24	2.2	528	47.6	415	37.4	96	8.7	40	3.6
-36	1169	1	0.01	27	2.3	557	47.6	445	38.1	98	8.4	41	3.5
Total	33 1000 57	9	0.3	99	2.9	1651	49.2	1220	36.3	268	8.0	109	3.2
Girls
-6	478	5	1.0	22	4.6	255	53.3	126	26.3	40	8.4	30	6.3
-12	616	9	1.5	32	5.2	337	54.7	182	29.5	42	6.8	14	2.3
-24	1094	5	0.5	21	1.9	539	49.2	389	35.6	101	9.2	39	3.6
-36	1122	3	0.3	35	3.1	533	47.5	458	40.8	104	9.3	43	3.8
Total	3310	22	0.6	850	2.6	1664	50.2	1155	34.9	287	8.7	126	3.8

Chang, 1985

Table 2. Prevalence of malnutrition in Chinese rural children< 3 years of age based on weight-for-age of NCHS population.

Months	N	>+2SD		+1SD-+2SD		SD-+1SD		-1SD- -2SD		-2SD- -3SD		<-3SD
		N	%	N	%	N	%	N	%	N	%	1000 N	%
	Boys
-6	465	8	1.7	72	15.4	339	72.9	39	8.4	7	1.5	0	0
12	616	3	0.5	32	5.2	332	53.9	201	32.6	50	8.1	7	1.1
-24	1107	0	0	7	0.6	272	24.5	489	44.1	2X6	25.8	52	4.7
-36	1169	0	0	2	0.2	331	28.3	541	46.3	264	22.6	30	2.6
Total	3357	11	0.3	113	3.3	1274	37.9	1270	37.8	607	1000 18.1	89	2.7
Girls
-6	478	23	4.8	89	18.6	296	6.2	56	11.7	13	2.7	1	0.2
-12	616	7	1.1	35	5.6	362	58.7	164	26.6	40	6.5	7	1.1
-24	1094	2	0.2	5	0.4	325	29.7	494	45.1	242	22.1	26	2.4
-36	1122	0	0	4	0.3	308	27.4	502	44.7	276	24.6	42	3.7
Total	3310	32	0.9	133	4.0	1291	39	1216	36.7	571	17.2	76	2.3

The data base of the Fels Research Institute, Yellow Springs, Ohio (WHO, 1982).

Table 3. Prevalence of malnutrition in Chinese rural children <3 years of age based on length-for-age of Chinese standard population.

Months	N	>+2SD		+1SD-+2SD		SD-+1SD		-1SD- -2SD		-2SD- -3SD		<-3SD
		N	%	N	%	N	%	N	%	N	%	N	%
Boys
-6	465	1	0.2	6	1.2	143	30.7	182	39.1	72	15.5	61	13.1
-12	616	6	1.0	16	2.6	285	46.2	298	4X.4	85	13.8	97	15.7
-24	1107	4	0.3	15	1.3	334	30.1	425	38.4	174	15.7	155	14.0
-36	1169	0	0	5	0.4	371	31.7	420	35.9	221	18.9	152	13.0
Total	3357	11	0.3	42	1.2	1133	33.7	1325	39.4	552	16.4	465	13.9
Girls
-6	478	2	0.4	3	0.6	140	29.3	167	34.9	80	16.7	87	18.2
-12	616	7	1.1	10	1.6	256	41.5	194	31.5	84	13.6	65	10.6
-24	1093	8	0.7	14	1.3	331	30.3	406	37.1	177	16.2	157	14.4
-36	1122	0	0	9	0.8	360	32.1	419	37.3	190	16.9	144	12.8
Total	3309	17	0.5	36	1.1	1087	32.8	1186	35.8	531	16.0	453	13.7

*Chang, 1985

Table 4. Prevalence of malnutrition of Chinese rural children < 3 years of age based on length-for-age of NCHS standard population.

Months	N		>+2SD				+1SD-+2SD				SD-+1SD				-1SD- -2SD				-2SD- -3SD				<-3SD
			N		%		N		%		N		%		N		%		N		%		N		%
Boys
-6	465		2		0.40		14		3.0		246		52.9 1000		154		33.1		40		8.6		9		1.9
-12	616		1		0.16		8		12.9		226		36.7		228		37.0		116		18.8		37		6.0
-24	1107		1		0.09		6		0.5		180		16.3		415		37.5		345		31.2		133		12.0
-3 1000 6	1169		1		0.08		0		0		108		9.2		318		27.2		407		34.8		282		24.1
Total	3357		5		0.14		28		0.8		760		22.6		1115		33.2		908		27.0		461		13.7
Girls
-6	478		1		0.21		8		1.7		266		55.6		136		28.5		46		9.6		21		4.4
-12	616		4		0.64		8		1.3		279		45.3		215		34.9		82		13.3		27		4.4
-24	1093		3		0.27		11		1.0		224		20.5		419		38.3		312		28.5		122		11.2
-36	1122		0		0		0		0		126		11.2		339		30.2		428		38.1		229		20.4
Total	3309		8		0.24