January 2010, Volume 60, Issue 1

Original Article

Prevalence and risk factors for Stunting among children under 5 years: a community based study from Jhangara town, Dadu Sindh

Syed Farid-ul-Hasnain  ( Department of Community Health Sciences,The Aga Khan University, Karachi, Pakistan. )
Raafay Sophie  ( Medical Student,The Aga Khan University, Karachi, Pakistan. )

Abstract

Objective: To assess the prevalence and the associated risk factors for stunting in Jhangara Town, Dadu Sindh.
Methods: A cross-sectional survey was conducted in Jhangara Town, located in district Dadu, rural Sindh. A total of 800 children less than 5 years of age were enrolled. A questionnaire was used to elicit required information and anthropometric measurements were made.
Results: The overall prevalence for stunting was 61% in the study population, which was higher than the prevalence reported by Pakistan Demographic Health Survey (PDHS) 1990/91. In multivariate analysis various risk factors for stunting were ethnicity (OR = 2.1, 95% CI 1.27- 3.57) and birth interval < two years (OR = 1.5, 95% CI 1.09 - 1.99).
Conclusion: A large number of children below 5 years of age were stunted. Ethnicity and birth interval less than 2 years were the important risk factors (JPMA 60:41; 2010).

Introduction

Adequate nutrition is not only essential for proper growth and physical development but ensures optimal working capacity, normal reproductive performance and adequacy of immune mechanism from conception to adulthood.1
Children under 5 years, who constitute about 9.4%2 of the total global population, are the primary victims of malnutrition. The association between malnutrition (measured as poor anthropometric status) and mortality has been well established.3 In addition, malnutrition has been shown to increase mortality associated with diseases, especially diarrhoea and acute respiratory infections.4
Malnutrition is responsible, directly or indirectly, for 54% of the 10.8 million deaths per year in children under five and contributes to every second death (53%) associated with infectious diseases among children under five years of age in developing countries.5,6
Malnutrition is of particular concern in developing countries. A report by UNICEF published in 2006 states that around 146 million children in developing countries are underweight - that is one out of every fourth child. Out of these, over half of the world's underweight children live in just three countries: Bangladesh, India and Pakistan.7
Pakistan Demographic and Health Survey (1990/91) examining the nutritional status of children less than 5 years revealed that, 50% of children were stunted, 9% were wasted while 40% were underweight (as compared to NCHS standard).8 The Pakistan National Nutrition Survey (1985-87) had showed similar results.9 Since then, Pakistan has made moderate progress with statistics for 2000-2006 showing that 37% of children are stunted, 13% are wasted and 38% are underweight.10 This progress is inadequate to meet the Millennium Development Goals and more that 8 million children under 5 years are malnourished in Pakistan.7
The aetiology of malnutrition is complex and multi-factorial. It is usually a consequence of inadequate dietary intake and disease. However, this occurs in combination with multiple social, economic, cultural and political elements. Factors reported in various studies include initiation of breast feeding and improper weaning,11 incomplete vaccination,12 low socioeconomic status, mother's age less than 20 years, poor sanitation of the area, child age <36 months,13 large family size,14 low birth weight,15 maternal education and birth interval.16
Child growth is internationally recognized as an important public health indicator for monitoring nutritional status and health in populations. A number of studies have established the association between increasing severity of anthropometric deficits and mortality.17
Of the various anthropometric indices that can be used to assess child growth status, three are more widely used: height-for-age portrays performance in terms of linear growth, and essentially measures long-term growth faltering (stunting); weight-for-height reflects body proportion, or the harmony of growth, and is particularly sensitive to acute growth disturbances (wasting); and weight-for-age represents a convenient synthesis of both linear growth and body proportion (underweight).18
In this study we will be focusing on stunting as an indicator of malnutrition. The objective of this study is to determine the prevalence of stunting and its associated factors among children under the age of five years in a rural town of Sindh province.

Subjects and Methods

A cross sectional design was followed to carry out the study. This study was conducted in Jhangara Town and nine villages within 20km area around the Jhangara Town. Jhanjara Town is located in district Dadu, rural Sindh which is 15 km away from Sehwan Sharif and 320 km from Karachi. The area is inhabited by Sindhi speaking Muslims, most of them are farmers and labourers and majority of the population is illiterate.
The total sample size was calculated as 692 at least, with a power of 90%, at a 95% confidence level (two sided) which can easily detect an odds ratio of at least 2, provided that the prevalence of various risk factors in the source population is between 15% and 75%.19
A total of 800 children less than five years of age were enrolled for the study giving an allowance of 15% for both expected refusal and incomplete information. (A census was conducted as the total number of children in the study site was around 800).
The inclusion criteria were to enroll those children who were under five year of age and he or she had not celebrated his/her fifth birthday at the time of interview. The selection of child was irrespective of ethnicity, religion and gender. Only the last born child of the family who was under five years of age was enrolled. If a person had more than one wife, then the youngest child of either of the wife was enrolled, if person had twins than one child was selected randomly. The couples who had an adopted child or stepchild (for either of the parents) was excluded from the study, similarly those children with visible congenital deformity were also excluded. As we were not able to measure the parental height, we cannot be sure about the inclusion of genetic or familial short stature children in the study.
A questionnaire was designed to elicit information from respondents with precision. The interviewers were trained based on the manual of instructions. Protocol for anthropometric measurements were physically demonstrated on children during the training sessions.20
The anthropometric measurements were converted into three indices: weight for age, weight for height and height for age. To calculate anthropometric indices the information regarding the child that is age (months), sex, weight (kg) and height / length (cm) were entered into nutritional anthropometric programme in Epi Info. These indices were then expressed as Z-scores relative to the international [National Center for Health Statistics (NCHS) / Center for disease control and prevention/ World Health Organization] reference population. A child nutritional status was then categorized by his / her Z-score.
A child was defined as stunted if the Z Score was less than that of a child with a value 2 SD below the reference median (NCHS)
Multivariate analysis was done through Multiple Logistic Regression model (SPSS Package version 7.5) to adjust for confounding.

Results

The descriptive results are based on 800 records. Out of 800 records, 21 records were flagged (based on nutritional anthropometry package). Flagging occurs when the value for indices becomes out of range which happens due to incorrect measurements. Bivariate analysis is based on 779 records.
We approached 816 eligible respondents, with a refusal rate of 1.5% and incomplete information 0.5%. A total of 800 questionnaires were completed form 771 household and the anthropometric measurement of the children less than 5 years was recorded. Overall 61% of the children were stunted.
In the study population boys were in majority (56.3%) while there were 43.8% girls. The mean age of the boys were 25.34 ±16.48 months while mean age of the girls was 25.75 ± 15.90 months (Table-1).
Only 1.9% of the parents were able to tell the documented birth weight of their child (Table-1). When asked about the appearance of their child in parents perspective 56.5% of the parents responded that their child looked normal at the time of birth while 28.4% said that their child was under weight and 15.1 % of the parents were not sure about the child appearance at birth (Table-1).


When child factor were considered for univariate analysis, the gender was not identified as a risk factor for stunting (OR=0.9, 95% CI 0.69-1.26) (Table-2). Similarly low birth weight (parents perspective) was not identified as a risk factor for stunting (OR=1, 95% CI 0.73-1.44) (Table-2).Breast feeding was unrelated to child's stunting (OR=1.3, 95% CI, 0.61-2.89) (Table-2). Prolonged breast-feeding also did not play any role in stunting (OR=0.9, 95% CI 0.60-1.23) (Table-2). Risk for stunting did not vary with the weaning age (OR = 1, 95% CI 0.70-1.38) (Table-2). Ethnicity (OR = 2.1, 95% CI 1.23-3.70) and birth interval (OR = 1.5, 95% CI 1.08-2.00) were identified as a risk factors for stunting (Table-2).


After assessing the variables for multicollinearity, various possible subsets were tried to obtain a best-fit model. Ethnicity and birth interval were identified in the main effect model. Interaction term that is ethnicity and birth interval was introduced which was not significant (P-value criteria).
Ethnicity played a role in stunting among children. Baluchi children were 2.1 times more likely to be stunted as compared to Sindhi children (OR=2.1, 95% CI 1.27 - 3.57). As there were 11% Baluchis, the results should be applied with caution. Birth interval was also identified as a risk factor for stunting. Mothers who had previous birth interval of less than 24 months were 1.5 times more likely to bear stunted children as compared to those mothers whose previous birth interval was 24 months or more from the index child (OR=1.5, 95% CI 1.09- 1.99). The final multivariate model for stunting is shown in Table-3.

Discussion

The over all prevalence in our study for stunting in children less than 5 years in a rural area of Pakistan was 61.0%. This is higher than the prevalence of 54.9% reported for rural areas in the last Demographic Health Survey in Pakistan carried out in 1990-91,6 as well as the prevalence of 46% reported in the National Nutrition Survey conducted in 1985-87.7 It was not unusual to observe a higher prevalence of malnutrition, because the study population of Jhangara belongs to a low socio-economic strata, with minimal health facilities, in sanitation, illiteracy and lack of infrastructure, these features are reflective of poor nutritional status of the children in community.
Data from WHO studies over the period 1996-2005 (period after our data was collected) showed an improvement in this parameter (37% prevalence).10 This may be due to child health programmes like IMCI that incorporate nutritional counseling or may be due to the data representing urban and rural data.
In India the prevalence of stunting among children (< 5 years) was reported as 52% in 1997.21 Similar improvements can be observed here as the prevalence improved from 78.6% to 45.5% from 1975-1979 to1998-1999.22 Data for the year 2007 estimates the current prevalence between 23.9% - 46.06%.23,24
In Bangladesh, a similar picture of stunting is seen. Prevalence was reported as 71.5% in 1985-1986, 66.5% in 1989-1990, 57.4% in 1995-1996,25 and 44% in 2007.26
In this study majority of the population were Sindhis while the rest were Baluchis. Analysis showed that Baluchi children were at risk for stunting and the relationship remained as such in multivariate model. These results are well in accordance with national figures, which showed that Baluchistan had the highest prevalence of stunting i.e. 70% as compared to other provinces.8 Within Baluchistan people may have peculiar food habits or diet or may be the weaning age is higher, which could be responsible for high stunting among Baluchi children. This study has not explored the feeding practices
In this study, birth interval less than 24 months was identified as a risk factor for stunting, this could be due to lack of awareness or lack of counseling regarding family planning practices and birth spacing. Birth interval of less than 24 months has been shown as one of the determinant of nutritional status of the children, in several studies.13,27 A systematic review on effect of birth interval on child nutritional status published in 2007 concluded that reduction in stunting from birth interval > 36 months ranged from 10%-50% in some, but not all populations.28 Lack of breast feeding was observed as a risk factor for stunting in the univariate analysis but was not significant, probably due to random variation.
Every study has its limitatio0ns; a follow up study design is much better to assess the nutritional status of the children. This study has not explored the parental height which is one of the determinants for stunting but nevertheless this study points out about the risk factors for stunting. The effect of both environmental and genetic factors is expected to have been obscured because children from most effected (e.g. most socially deprived) families happened to be most malnourished and already passed away, this created a built-in bias in the study, because the children who were alive were enrolled for the study. Another limitation, regarding this study, was that it was conducted in a rural area with high illiteracy rate, hence, it was expected that problems of recalling, like difficulty in reporting own age, child age and birth interval etc. could affect the results of the study. To obtain a good quality data, interviewers were trained to use local events and Islamic calendar to get a more precise response. Other measures like close motioning of field workers, daily editing and validated double entry ensured the quality of the data.
The data for this study was collected between July and December 1997. The publication was delayed due to administrative reasons.

Conclusion and Recommendations

This study identified shorter birth interval and ethnicity (Baluchi) as risk factors for stunting. Based on the results of this study, we suggest specific counseling to increase birth interval to more than 2 years and a recommendation for health policy makers at a local or broader level with emphasis on Baluchi population.
More improvement is needed in community based nutrition programs as a direct means of improving nutrition, and as a concrete focus for nutritional concerns and policies.

References

1.David AB. Introduction to Nutrition and Metabolism. 3rd ed.: Taylor & Francis, London; 2002.
2.International Data Base (IDB) of US Census Bureau. (Online) [Cited 2008 Aug 10]; Available from URL: http://www.census.gov.
3.Pelletier DL. Methodology for estimating the contribution of malnutrition to child mortality in developing countries. J Nutrition 1994; 124: 2106-22.
4.Rice AL, Saccol, Hyder A, Black RE. Malnutrition as an underlying cause of childhood deaths associated with infectious diseases in developing countries. Bull World Health Organ 2000; 78: 1207-21.
5.Schaible UE, Kaufnaann SH. Malnutrition and Infection: Complex Mechanisms and Global Impacts PLoS Med. 2007; 4: e115.
6.World Health Organization. Nutrition: Challenges 2005. (Online) 2005 (Cited 2009 Aug 03). Available from URL: http://www.who.int/nutrition/challenges
7.Progress For Children, A Report Card On Nutrition. Unicef May 2006.
8.Pakistan Demographic and Health Survey 1990/91 National Institute of Population Studies, Islamabad, Pakistan 1992; 149-63.
9.Pakistan National Nutrition Survey 1985-87 National Institute of Health, Islamabad, Pakistan 1988: pp 19-36.
10.State of the world's children. (Online) 2008 (Cited 2008 August 4). Available from URL: http://www.unicef.org/sowc08/.
11.Kumar D, Goel NK, Mittal PC, Misra P. Influence of Infant-feeding Practices on Nutritional Status of Under-five Children. Indian J Pediatr 2006; 73: 417-21.
12.Siddique B, Jamal A, Aslam R. Assessment of risk factors & case fatality rate of malnourished admitted children. Med Channel 2006; 12: 47-51.
13.Vitolo MR, Gama CM, Bortolini GA, Campagnolo PD, Drachler Mde L. Some risk factors associated with overweight, stunting and wasting among children under 5 years old. J Pediatr (Rio J) 2008; 84: 251-7.
14.Salim F. Growth of children; effect of family size. Professional Med J 2005; 12: 14-6.
15.Rayhan IM, Hayat SK. Factors Causing Malnutrition among under Five Children in Bangladesh. Pak J Nutrit 2006; 5: 558-62.
16.Mozumder AB, Barkat-e-Khuda, KaneTT, Levin A, Ahmed S. The effect of birth interval on malnutrition in Bangladeshi infants and young children. J Biosoc Sci 2000; 32: 289-300.
17.de Onis M, Blössner M. The World Health Organization Global Database on Child Growth and Malnutrition: methodology and applications. Int J Epidemiol 2003; 32: 518-26.
18.Group WW. Use and interpretation of anthropometric indicators of nutritional status. Bull of the World Health Organ 1986; 64: 924-41.
19.Centers for Disease Control and Prevention (CDC). Epi Info. 6.04d ed. Atlanta; 2001.
20.World Health Organization. Physical status: The use and interpretation of anthropometry. Geneva: WHO, 1995.
21.The state of the world children. UNICEF 1997; pp 82-99.
22.Sachdev HPS, Shah D. Recent Trends in Nutritional Status of Children in India. 10th ASCON; 2002. (Online) Available from URL: http://www.icddrb.org/pub/publication.jsp?Classification ID=08 pub ID=1420.
23.Bose K BS, Bisai S, Ganguli S, Khatun A , Mukhopadhyay A , Bhadra M. Stunting, underweight and wasting among Integrated Child Development Services (ICDS) scheme children aged 3-5 years of Chapra, Nadia District, West Bengal, India. Matern Child Nutr 2007; 3: 216-21.
24.Mittal A, Singh J, Ahluwalia SK. Effect of Maternal Factors on Nutritional Status of 1-5-Year-Old Children in Urban Slum Population. Ind J Community Med 2007; 42: 264-7.br>25.Mannan MA, Azmatullah ZA, Ali EM, Akhter D. Nutrition Scenario in Bangladesh: Changes Over the Decades 10th ASCON 2002.
26.Rahman A, Chowdhry S. Determinants of chronic malnutrition among pre-school children in Bangladesh. J Biosoci Sci 2007; 39: 161-73.
27.Som SP, Bharati P. Role of individual and household level factors on stunting: a comparative study in three Indian states. Ann Hum Biol 2007; 34: 632-42.
28.Dewey KG. Does birth spacing affect maternal or child nutritional status? A systematic literature review. Matern Child Nutr 2007; 3: 151-73.
29.McDade TW, Blackinton P, Tanner S, Huanca T, Leonard WR. Ethnobotanical knowledge is associated with indices of child health in the Bolivian Amazon. Proc Natl Acad Sci USA 2007; 104: 6134-9.
30.Berggren WL, Wray JD. Positive deviant behavior and nutrition education. Food Nutr Bull 2002; 23: 7-8.

Journal of the Pakistan Medical Association has agreed to receive and publish manuscripts in accordance with the principles of the following committees: