Grace Ettyang

PSYCHO-SOCIAL AND NUTRITIONAL STATUS
OF STREET CHILDREN IN COMPARISON
TO SCHOOL CHILDREN:
A CASE OF ELDORET TOWN

Ayuku D, Ettyang E and W Odero

ABSTRACT

David Ayuku

In most developing countries the number of children growing up on the streets of the world’s cities is rising at an alarming rate. There is scanty research on the effect of such a life style on the psychosocial and nutritional status of these children. The goal of this study was therefore to examine the social networks, temperament and nutritional status of street based children compared to schoolchildren with a poor housing background. Our psychosocial research methods included qualitative and ethnographic techniques combined with standardised socio-metric and psychometric tests. Snowball sampling was used to identify the study groups. Four distinct groups of 100 children each were randomly selected. The street children were grouped as either living ‘on’ or ‘of’ the street. The school children comprised of those either living at home with a family or those living in a shelter under the care of social services. The Maastricht Social Network Analysis (MSNA) and psychometric Revised Dimensions of Temperament Survey (DOTS-R) tests were used to determine the social networking capacity and temperament of the children. Nutritional status was based on weight and height. Skinfold measurement was used to determine percentage body fat (%BF) and fat free mass (FFM). Our results show a mean age of 13.34 years for all the four groups of children. Both the “on” the street and “of” street children did not go beyond lower primary school and they lived by scavenging. The children living at home and at the shelter had a balanced network between family, friends and social service providers. Both the ‘on and ‘of’ the street children had a significantly higher level of underweight (p <0.05) and stunting (p<0.05) as well as highly significant (p<0.1) lower body mass index (BMI), % BF and FFM. Our findings indicate that homelessness and absence of family support do not necessarily have adverse implication for psychological and physical development of children in especially difficult circumstances. However, efforts should be directed towards improving the home environment and household food security so that these children stop scavenging for food in the streets.

Key words: street, social network, nutrition, homelessness, temperament.

FRENCH

L'état psychosocial et nutritionnel des enfants de la rue en comparaison
avec les enfants scolarisés : cas de la ville d'Eldoret

Resumé

Dans la plupart des pays en développement, le nombre d'enfants qui grandissent dans les rues des centres urbains augmente à un taux alarmant. Les recherches sur l'effet qu'un tel style de vie produit sur l'état psychosocial et nutritionnel de ces enfants sont limitées. En conséquence, la présente étude avait pour objectif d'examiner les réseaux sociaux, le tempérament et l'état nutritionnel des enfants de la rue en comparaison avec les enfants scolarisés vivant dans de mauvaises conditions de logement. Les méthodes de recherche psychosociale utilisées comprennaient les techniques qualitatives et ethnographiques en association avec les tests socio-métriques et psychométriques standardisés. L'identification des groupes d'étude a été faite par échantillonage en boule de neige. Quatre groupes distincts comprenant chacun 100 enfants choisis au hasard ont été formés. Les enfants de la rue ont été groupés en deux catégories : les sans-abri qui vivent dans la rue et ceux qui ont un abri mais dépendent de la rue pour leurs ressources. Quant aux enfants fréquentant l'école, ils ont été groupés en deux catégories : ceux vivant chez eux dans le foyer familial et ceux vivant dans les centres d'acceuil pris en charge par des services sociaux.

L'analyse des réseaux sociaux de Maastricht (MSNA) et les tests psychométriques des Dimensions Révisées de l'Etude des Tempéraments (DOTS-R) ont été utilisés pour déterminer la capacité de développement de réseaux sociaux et le tempérament des enfants. L'état nutritionnel a été déterminé sur base du poids et de la taille/hauteur, et le taux des graisses corporelles et la masse maigre ont été déterminés par les mesures du pli cutané. Les résultats de l'étude dégagent un âge moyen de 13,34 ans pour tous les quatre groupes d'enfants. Les deux groupes d'enfants de la rue ont abandonné leurs études dès les premières années de l'école primaire et font les poubelles. Les enfants vivant dans un foyer dans le cadre familial et dans les abris des services sociaux ont un réseau équilibré entre la famille, les amis et les prestataires de services sociaux. Les deux groupes d'enfants de la rue présentent une insuffisance pondérale (p <0,1) et un retard de croissance (p <0,05) graves, un indice de masse corporelle (IMC) très bas (p <0,1), un faible taux de graisse corporelle et un très faible pourcentage de masse maigre. Nos résultats indiquent que le manque de domicile fixe et de soutien familial n'ont pas nécessairement un effet négatif sur le développement psychologique et physique des enfants dans les circonstances particulièrement difficiles. Cependant, des efforts devraient être focalisés sur l'amélioration de l'environnement domestique et la securité alimentaire des ménages afin d'empêcher ces enfants de faire les poubelles dans les rues.

Mots-clés  : rue, réseau social, nutrition, manque d'abri fixe, tempérament

INTRODUCTION

More than a decade ago, the United Nations International Children’s Fund (UNICEF) estimated that approximately 100 million children and adolescents are growing up on the streets of the world’s cities [1]. Increasing poverty, war, famine and disease occurring over the past decade have substantially increased this number. The phenomenon of street children is prevalent throughout Africa; in Kenya alone an estimated one million children are now orphaned by AIDS with most ending up in the streets [2]. The lack of a stable parental figure has been linked to a variety of negative outcomes [3,4]. Children’s ties to their families and their compensatory relationships with significant others are of developmental significance [5]. Evidence pertaining to the vulnerability and coping of street children, for example, is contradictory; peer relationships are reported as erratic and unstable in some publications and mutual caring in others [6]. UNICEF [7] defines the “on” the street children as usually engaged in some sort of child labour while maintaining strong family ties expressed by sleeping and basing themselves in the family home; and “of” the street children as having their ties in the street with little or no social support from the family. Abandoned children may be on the streets because of the death or rejection by their parents. Such children may end up in a shelter run with help from the government or non-governmental organizations. The purpose of this study was to examine the social network, temperament and nutritional status of street based and school children in Eldoret.

METHODS

The study was a cross-sectional design. The study site was Eldoret, a Kenyan town with an estimated population of 300,000 people. An adaptive snowball sampling strategy was applied to recruit representative “on” and “of” the street children samples [8,9,10]. This sampling strategy was employed because of non-existence of a sampling frame for the “on” the street and “of” the street children samples. Since these street children constituted the “experimental” group in our research design, we wanted to go beyond a convenience sample to draw the most representative sample that was possible. We allowed the street children to “drive” the sampling process offering ideal and material incentives for their participation [11]. One hundred children in each group representing “on” the street, “of” the street children, shelter children and a control group of primary school children from poor Eldoret neighbourhoods were recruited for the study. In each group a random sample of 51 children were selected making four sub-samples of 204 boys from the cohort of 400 children This group of children was further assessed with psychometric test, the Revised Dimensions of Temperament Survey (DOTS-R) a 54-item, factor-analytically developed self-report instrument that measures nine temperament dimensions. A standardised psychometric instrument for cross-cultural assessment of children’s’ temperament validated for Kenyan samples was also used [12,13]. A Salter stand on validated weighing scale was used to take weight. A height metre was used to take height [14]. Body mass index (BMI) was calculated as weight/Height (m²). The prediction formulas developed by Deurenberg [15] was used to determine % body fat (%BF) and fat free mass (FFM) where %BF = 1.51X BMI – 0.70 X age – 3.6 X 1 + 1.4(R² 0.38) for children aged < 15 years and %BF = 1.20 X BMI + 0.23 X age – 10.8 X 1-5.4 (R² 0.79) for children aged > 16 years. Total body fat (TBF) was calculated as TBF (kg) = %BF × weight (kg).

Fat free mass was calculated as FFM (kg) = weight (kg) - TBF (kg). Levels of underweight, stunting, and wasting were determined based on the NCHS reference standards. A flexible tape measure was used to measure mid-upper-arm circumference (MUAC). A holtain skinfold thickness calliper (UK) was used to measure in quadruple biceps, suprailiac and subscapular skinfold thickness [14]. One investigator made all the measurements. The mid-upper-arm muscle area (M) results from the mid upper-arm circumference (MUAC) and the triceps skinfold thickness (TSK) and is used widely as an indicator of the total body muscle mass. The following equation was used: M (mm2) = [MUAC- (p × TSK)] 2/4p [14].

RESULTS

All the children were boys with a mean age across the groups of 13.34 with no significant differences. The “on” the street and “of” the street children scavenged for a living and the “on” street children came from single parent home while the school children came from two parent home (“on” the street 46% single parents and school 57% two parents). Of the children ‘on’ and of street 16% and 21% had not attended any schooling respectively (Table 1).

Table 2 illustrates aggregating the network members and cross tabulating the distribution of the boys network members across sectors by groups. The Pearson Chi. Square test showed that there were significant differences between groups and sectors (c² = 279.4; 6; p < 0.000). The “on” the street group had the largest percentage of the network in the family sector (53.7) followed by the “of” the street group (47.9) the “on” and “of” street children have a relatively larger percentage of family members in their networks compared to the control groups (shelter = 40.1; school = 42.2).

Similar results were observed for the friends sector. The “on” and “of” the street children had a relatively larger percentage of their social network members in the friends sector (“on” the street = 36.5; “of” the street = 41.5) compared to the control (shelter = 30.8; school = 30.3). An inverse pattern of relationship was observed for the social service sector. The “on” and “of” the street groups had relatively lower percentage of social service in their networks (“on” street =9.8; “of” street =10.6) compared to the shelter and school children groups (shelter = 29.1 and school children = 27.5).

As shown in Table 3 the statistical analysis for the DOTS-R dimensions showed statistically significant differences on the approach/withdrawal dimension (F =12.384, p <0.001). The activity level-sleep (F= 4.196, p <0 .01) and the task orientation (F=3.616, p<0.01) dimensions were next in rank in terms of significance level. The difference on the activity level-general dimension (F=3.006, p<0.05) was also found to be statistically significant. Statistical trends could be seen in the flexibility/rigidity (F=2.034, p<. 10) and rhythmicity-daily habits (F=2.166, p<0.10). No other significant differences between the groups were detected.

Over all the prevalence of severe underweight (z <-2.00) and stunting (z <-2.00) was 30.7% and 25.9% respectively. Severe underweight and stunting was 35.3% and 35.3% for “on” the street; 23.5% and 21.6% for “of” the street, 45.1%; and 27.5% for shelter, 19.6% and 19.6% for school children respectively. Body Mass index was highly correlated with measures of leanness i.e. Fat Free Mass (FMM, r =0.78 p<0.01). A Body Mass Index (BMI) of <17.5 indicative of severe malnutrition and chronic energy deficiency (CED) was found in 64.9% of the children. Children with a MBI < 17.5 and living at home with a family, in a shelter, “on” the street and “of” the street were 68.6%, 62,5%, 80.4% and 49.0% respectively. The mean (±SD) MUC, underweight, levels of stunting, BMI, % body fat and FFM statistics are shown in Table 4. The mean (±SD) BMI for children in school, shelter,”on” and “of” the street was 16.9(1.8), 16.9(2.2), 15.7(2.1) and 17.7(2.2) respectively.

The significance of the differences in nutritional status among the groups is shown in Table 5. Compared to the “of” the street children, the “on” the street children’s nutritional status shows a significantly higher level of underweight (P<0.05) and stunting (p<0.05). Very highly significant (p<0.01) are a lower BMI, %BF and FFM. A different picture images (results not shown) when similar comparisons are made between “on” the street, “of” the street and school children. Compared to school children “on” the street children have highly significantly (p<0.01) higher levels of under weight and stunting. Children “of” the street children have a highly significantly (p<0.01) higher mean MUC and BMI.

One-way analysis of variance showed that there were significant differences between and within the four study groups. The mean statistics were BMI (F 7.735 df =3 p < 0.01), FFM (F 15.9, df =3 p <0.01), underweight (F 6.60, df = 3 p < 0.01) and stunting (F 4.11, df = 3 p < 0.01). The difference was in favour of the “of” the street children with a mean (±SD) BMI of 17.7, 2.2. A similar pattern in the prevalence of underweight and stunting was also observed.

DISCUSSION

Our primary aim was to provide a description and interpretation of a representative sample of Eldoret’s street children population that can be compared to schoolchildren as a control group. Consistent with the Eastern African literature, street children are, for the great majority, a male population [16,17]. The school control group also had a large majority of males, though the gender difference is not statistically significant. Almost all of the “of” the street children are scavenging for a living, but, also, somewhat surprisingly, the majority of “on” the street children are also engaged in this activity. While more than half of the school children live with two parents only a quarter of the “on” the street children are based in intact families. Issues relating to female-headed households are definitely pertinent to this group, as has been found by other Kenyan studies [18]. With respect to nutritional well being, a stable socio – economic environment is imperative for the cognitive development of the children [19]. Surprisingly, the prevalence of wasted children was relatively low. This becomes more obvious by comparing the levels of stunting in the “on” and “of” the street with the shelter and school children. We also observed a difference in the average weight, which favours the “of” street children when comparing them with their siblings at the shelter. Furthermore, their astute begging might actually enhance the nutritional intake [20].

According to the causal model of malnutrition proposed by UNICEF [21] and accepted internationally by FAO/WHO [22], caring capacity plays a central role. Children particularly rely on the physical and emotional support of others [23]. Several studies have shown that an inappropriate caring capacity leads to under-nutrition [24]. Following this rationale, the lower prevalence of underweight and stunting in “of” street children may be due to the support received from their peers on the streets. The prevalence of underweight and stunting are significantly lower in “of” the street children and may indicate their ability to survive in a very hostile environment. But the disadvantaged life led by these street children is reflected in the levels of body mass index. A person with a low BMI is most certainly physiologically and physically disadvantaged [25,26]. The primary social network deficiencies of street children seem to be related to size and imbalance in sector composition. The low percentages of service providers in both the “of” and “on” the street groups contradicts the view presented by Campos et al. [27], that the street-based children have more contact with service providers than their home-based counterparts as it is the case in Brazil. In our study the shelter and school groups seem to have a more balanced network of family, friends and social service. We also note that the “on” and “of” the street children have a noticeable lack of social service in their social networks. This deficit in social service in the “on” and “of” the street groups seems to be compensated for by family for the ‘on’ the street group and friends for the ‘of’ the street group.

The psychological measures findings supports research suggesting that the majority of street children do not present mental health problems [20]. Based on the findings of the psychological test analysis, “of” the street and shelter children temperament stylistic trait seems to have a good fit with their lifestyle and institutional demands. The religiously oriented shelter programs, in their local and international fund-raising activities, actively employ these children. They are encouraged to approach prospective donors and assigned highly supervised tasks of singing, acting and/or proselytising to obtain support for the programs as well as the household chores in the shelter. The paradoxical easy temperament style of street children may be an expression of a continuous childhood developmental struggle to find a niche in a hostile and dangerous world constituted by especially difficult circumstances. Although not statistically significantly different than the control group, the “on” the street children’s relatively higher score on the mood dimension indicates perhaps a conditioned contextual pathway formed by the cumulative reactions of others. Expressions of good mood and friendliness to most people in the street by street children fits well with their dependence on strangers for donations and other favours. However, to maintain this temperament style, as other African studies have documented, they sniff glue and take other drugs to cope with the hunger, cold and hostility [28]. However, such a process-oriented theory to explain the paradoxical temperament of street children, as has also been suggested by the American adolescent research, can only be tested with longitudinal research. Contrary to the situation observed among homeless families in the West. Street children in Eldoret are not necessarily at the bottom of the heap (in terms of growth and physical health) and adoption of street-life may, for some at least, represent both a rational and successful response to their prior circumstances. Others who have studied children in developing countries [7,17, 20, 29] have expressed similar views. Although, our data are cross – sectional rather than longitudinal, this suggests that the “of” the street children are able to maintain growth despite a lack of permanent shelter and parental care.

CONCLUSION

The linking of the MSNA methodology to a snowball sampling proved to be an efficient way of obtaining a relatively large and representative sample for the study. Qualitative field observations, key informant interviews and focus groups supplemented the MSNA and psychosocial and nutritional data. These findings indicate that, at least in Eldoret, homelessness and absence of family support may not necessarily have adverse implication for psychological and physical health development of street based children. However, the finding of this study suggests that efforts should be directed toward improving the home environment and household food security so that these children stop scavenging for food in the streets.

ACKNOWLEDGEMENTS

The study was supported by an educational development grant from the Dutch International Educational Organization (NUFFIC) and the Directorate General of the Dutch Ministry of International Cooperation (DGIS) to the Maastricht University Centre for International Cooperation in Academic Development (MUNDO) and the Moi University Faculty of Health Sciences. We would like to especially thank Dean B. O. Khwa-Otsyula of the Moi University Health Sciences and Margreet te Wierik and Geraldine van Kasteren, the MUNDO Long Term Experts, the Eldoret Street Children and Faculty of Health Sciences Students.

REFERENCES

1. UNICEF. United Nations Children’s Fund. Annual Report, UNICEF, New York, 1989.

2. World Bank. Kenya Poverty Assessment Report. World Bank, Washington, 1995.

3. Krazler EM Parent Death in Childhood In: Childhood stress. Arnold LE (Ed.),
. Wiley, New York, 1990, 405-421.

4. Damon W Social and Personality Development. Norton, New York, 1983.

5. Kimchi J and B Schaffner Childhood Protective Factors and Stress Risk In: Childhood Stress. Arnold LE (Ed.), Wiley, New York, 1990; 475-500.

6. Donald D and J Swart-Kruger The South African Street Child: Developmental Implications. S. African J. Psychology, 1994; 24: 169-174.

7. UNICEF. United Nations Children’s Fund. Children in especially difficult circumstances. UNICEF, New York. 1986.

8. Spreen M Rare Population, Hidden Populations, and Link-Tracing Design: What and Why?
Bulletin de Methodologie Socioloque, 1992; 36: 59-70

9. Kaplan CD Korf D and C Sterk Temporal and Social Contexts of Heroin-Using Populations: An Illustration of the Snowball Sampling Technique. Journal of Nervous and Mental Disorders, 1987; 175: 566-574.

10. Thompson SK Adaptive Sampling in Behavioural Survey in the Validity of Self-reported Drug Use.

11. Heckathorn D Respondent – driven Sampling: A New Approach to the Study of Hidden Populations, Social Problems. 1997; 44: 174-199.

12. Windle M Reasseing the Dimensions of Temperamental Individuality Across the Life Span: The Revised Dimensions of Temperament Survey. Journal of Adolescent Research, 1986; 1: 213-230.

13. DeVries MW Temperament and Infant Mortality among the Masai of East Africa. Amer. J. Psychiatry, 1984; 141: 1189-1194.

14. Gibson R Principles Of Nutritional Status Assessment. Oxford. Oxford University press 1990.

15. Deurenberg P ,Westrstrate JA and JC Seidel Body Mass Index as a Measure of Body Fatness Age and Sex Specific Prediction Formulary. Br. J. Nut. 1991; 65: 105-114.

16. Beyene Y and Y Berhane Characteristics of Street Children in Nazareth, Ethiopia, East Africa Medical Journal, 1997; 74: 85-88.

17. Ayaya SO and FO Esamai Health Problems of Street Children in Eldoret, Kenya. East Africa Medical Journal, 2001; 78: 624-629.

18. Wainaina J The “Parking Boys” of Nairobi. African Journal of Sociology, 1981; 1: 7-45.

19. Nabarro D, Howard P, Cassels C, Pant M, Wijiga A and N Padfield The Importance of Infections and Environmental Factors as Possible Determinants of Growth Retardation in Children In: Linear Growth Retardation In: Less Developed Countries. Waterlow JC (Eds) , Raven Press, New York, 1988.

20. Aptekar L Street Children of Cali. Duke University Press, Durham, NC, 1988.

21. UNICEF. United Nations Children’s Fund. Strategy for Improving Nutrition of Children and Women
In: Developing Countries. UNICEF, New York, 1990.

22. FAO/WHO Food and Agriculture Organisation of the United Nations/ World Health Organization. Major Issues for Nutritional Strategies. International Conference on Nutrition. FAO/WHO, Rome, 1992.

23. Weisner TS Support for Children and the African Family Crisis, African Families and the Crisis of Social Change. BERGIN &GARVEY Westport Connecticut, 1997.

24. Longhurst R and A Tompkins The Role of Care in Nutrition - A Neglected Essential Ingredient. SCN News, 1995; 12:1.

25. James WPT The Functional Significance of Low Body Mass Index. (IDECG 1992) European Journal of Clinical Nutrition, 1994; 48: 3.

26. Desai ID Nutritional Status and Physical Work Performance of Agricultural Migrants in Southern Brazil. Proceedings of XIVth International Congress of Nutrition Seoul Korea: Korean Nutrition Society, 1989, 297-301.

27. Campos R, Antunes CM, Raffaeli M, Halsay N, Ude W, Greco M, Greco D, Ruff A, Rolf J and SYS Group Social Networks and Daily Activities of Street Youth in Belo Horizonte, Brazil., Child Development, 1994; 65: 319-330.

28. Jansen l, Richter R and R Griesel Glue sniffing: A Comparison Study of Sniffers and Non sniffers, Journal of Adolescence, 1992; 15: 29-37.

29. Ennew J Street and Working Children-A Guide to Planning. Development Manual 4.
Save the Children, London, 1994.

*Corresponding author Email:ayuku@africaonline.co.ke
¹Faculty of Health Sciences, Moi University, P.O. Box 4606, Eldoret, Kenya