![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
* Institute of Nutrition of Central America and Panama, Apartado Postal 1188, Guatemala,
Division of Nutritional Sciences, Cornell University, Ithaca, NY 14853-6301 and
** Department of Nutrition, University of Montréal, C. P. 6128, succursale A, Montréal, Québec, Canada H3C 3J7
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
KEY WORDS: • stunting • malnutrition • children • caregiver • behavior
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
model of care (Engle et al. 1997a
) (Fig. 1
).Care behaviors include breastfeeding and feeding of young children,
hygiene and health practices, food preparation and storage,
psychosocial stimulation and care for women. Caregiving behaviors are
difficult to measure because they vary substantially from one culture
to another (Engle et al. 1997a
). Moreover, child-care
behaviors are often neglected because it is thought that their
influence on child nutrition is minimal with respect to the influence
of household factors.
|
). The
resources that a caregiver draws on in giving care include: education,
knowledge and beliefs, physical health and nutritional status, mental
health and self-confidence, autonomy and control of resources,
reasonable workload and availability of time, and family and community
social support. As mentioned by Engle et al. (1997a),
a
caregiver who has the resources available is more likely to give
effective care and therefore maintain good child nutrition. Some
caregiver resources were extensively studied in developing countries,
like education and nutritional knowledge (Christian et al. 1988
, Guldan et al. 1993
, Kutty 1989
, Ruel et al. 1992
), nutritional status of
caregiver (Allen et al. 1990
, Christian et al., 1989
, Winkvist 1995
) and workload and time
constraints (Engle and Pedersen 1989
, Popkin and Solon 1976
, Rabiee and Geissler 1992
,
Tucker and Sanjur 1988
, Wandel and Holmboe-Ottesen 1992
). Others such as mental health or stress
(Engle and Ricciuti 1995
, Griffiths 1988
), autonomy (Bégin et al. 1997
,
Blumberg 1988
, Engle 1993
, Haddad and Hoddinott 1994
) and social support from community or family
members (Engle 1991
, Weisner and Gallimore 1977
) received much less attention. This paper attempts to identify caregiver characteristics that influence child nutritional status as measured by height-for-age in rural Chad and evaluates their importance while controlling for household socioeconomic factors. The caregiver characteristics that were studied correspond to caregiving behaviors and to resources for care such as caregiver nutritional status, mental health, autonomy, workload, and social support, as described in the extended UNICEF model of care. In our study, we hypothesized that the caregiver characteristics, in particular her autonomy, mental health and social support, which we will refer to as psychosocial characteristics of caregiver, are important predictors of child height-for-age even when controlling for household socioeconomic status. For this study, child height-for-age was used as the index of nutritional status, because the household socioeconomic and caregiver's variables that we studied were more likely to have a long-term nutritional influence on height than on weight. Throughout the paper, the term "caregiver" is used rather than "mother," although most of the time, the caregiver is the mother.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The data for this study were collected in Mao, Chad, in 1988 (October to December). Earlier in the same year, a food and nutrition survey was carried out in rural households involved in dune and wadi (fertile land depressions between sand dunes) agriculture in three zones of the Sahelian region of Kanem, including Mao. Only in Mao was this second study conducted. All households were eligible. Using a census list and a random digit table, 84 households (7% of the Mao population) were randomly chosen and 64 with preschoolers (136) were surveyed. In total, 16 infants and 98 children (12–71 mo) had no missing weight or height. Since we wanted to examine maternal influence on child care and nutrition, only the 98 children over 12 mo were selected for the present study as children above that age were more likely to be fed foods other than breast milk and more likely to be ill. Questions about child feeding, decisions on child feeding, and health practices were not informative for mothers with a child less than 12 mo. The population under study comprised two ethnic groups, the Kanembous and the Goranes, both being Muslim. Approximately 57% of the families lived in huts while the others lived in houses built with cement blocks. The average household size was seven to eight persons.
Due to wadi cultivation, agricultural activities are continued for most
of the year. Women are usually involved in sowing and in wadi
irrigation and harvesting, while men are responsible for the more
physically demanding tasks of cereal production, such as plowing and
harvesting on dunes. Households are highly dependent on cereal and
horticultural production for their food and as a source of income.
Women contribute to household income by selling wadi's products, as
well as handicrafts. More detailed descriptions of the area and of the
households can be found elsewhere (Bégin et al. 1992
, Delisle et al. 1991
).
Data collection.
Household and caregiver data were obtained from personal structured
interviews. The survey questionnaire included open-ended and closed
questions and was developed based on findings of the previous survey in
the same community and on results of studies of caregiver and household
determinants of child nutritional status reported in the literature.
The questionnaire was pretested among 10 households living in an area
similar to the one surveyed. All interviews were conducted in the two
local languages by trained female and male enumerators. In each
household, the head of the household was interviewed about economic and
health resources, and all women were interviewed about childcare
variables and resources for care. Caregivers were interviewed by women,
and heads of households (usually men) by male interviewers. A summary
of the information collected is provided in Table 1,
presented according to the categories described in the expanded UNICEF
model of care.
|
At the household level, demographic, socioeconomic and environmental
variables were collected. Amount of cultivated land was measured using
a compass, strings and stakes, and the area was calculated with a
programmed pocket calculator. Household heads were asked to report
amount of harvested cereals for the past 6 mo and the amount of cereals
stored, which was verified by the enumerator whenever possible.
Household heads were asked about their sources of income and the number
of livestock and fruit trees. Household food situation was assessed by
measuring food consumption at the household level. Food consumed by the
household was weighed over two consecutive days. Household adequacy in
energy, proteins, vitamin A and iron was estimated according to the
FAO/WHO report (1988)
. Details on the methods used were
described in a previous paper (Bégin et al. 1992
).
Child recumbent length (<24 mo-old) or height was measured to 0.1 cm
following standard procedure (WHO 1983
). Age was
obtained from immunization cards or caregiver recall, using a
historical calendar of local events. Heights were converted to sex- and
age-specific Z-scores (Waterlow et al. 1977
) relative to
the NCHS/WHO distribution (WHO 1979
), using the CDC
Anthropometric Software Package (version 3.0, 1987).
Data analysis.
All variables related to height-for-age
(HA)4
with a P-value 
0.15 in univariate analyses were included
in ordinary least square regression analyses. Factors studied were
classified as caregiver or household variables, and separate models
were run for each of the two sets of variables, using stepwise multiple
regression procedures (Neter et al. 1990
). Caregiver and
household factors that were significant predictors of HA at
P 0.15 were then combined in the same model to
assess their relative importance. The final model included variables
that were significant at P 0.10. Because of missing
values, between 76 and 92 cases were used in the regression equations.
In eight households, the amount of harvested cereals could not be
obtained as the head of the household was absent, and in three
households, caregivers were absent during the interview. All the above
analyses were performed with the statistical package for social
sciences, version 4 (SPSS Inc. 1990
).
The individual child rather than the caregiver or the household was
used as the unit of analysis. The varcomp procedure of SAS
(1985)
was used to partition the variance of the residuals from
the regression analysis into mother and child variance components. The
variance component for mothers was very small (estimated to be zero),
indicating that there was no clustering effect of multiple children
within mothers (Snedecor 1980
); therefore, the
regression assumption of independence of observations was not violated.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
0.15) are listed in Table 2.
The age of the child was strongly and positively associated with his or
her HA. Caregivers who received more help to accomplish their domestic
or productive tasks had taller children. Caregivers' total income was
not related to child HA, maybe because of the difficulty in assessing
their income. However, reported use of caregiver income for household
purchases (like kitchen utensils) was positively associated with height
status of the child.
|
Regarding household socioeconomic and demographic factors, the amount
of harvested cereals and report of cereal sales as a source of income
were positively associated with greater HA of children. Variables that
are not listed in Table 2
did not show a significant relationship with
HA, very often because of homogeneity in the variables or difficulty in
measuring them. For example, none of the women had received formal
education. Also, none of the variables on child-weaning practices were
related to HA, possibly because of poor recall by the caregiver of
events that occurred several years prior to the interview for many of
the children.
Separate multivariate regression models were run for the
caregiver and household variables listed in Table 2
. Two caregiver
models were developed including either number of hours worked per day
or mean hours devoted to domestic tasks, because these variables were
correlated (r = 0.40). Similarly, two household models
were developed including either the amounts of harvested cereals per
adult-equivalent or men's cereal sales (r = 0.30
between the two variables). We selected the one caregiver and the one
household model that showed the highest coefficient of determination
(R2). All factors found to be significant
predictors of HA at P 0.15 were then combined in a
single model to assess the importance of caregiver variables while
controlling for household socioeconomic status. Surprisingly, maternal
height and caregiver workload were not found to be significant
predictors of HA and therefore were not included in the models.
Means and SD of the child, caregiver and household variables included
in the final combined regression model are given in Table 3.
As noted earlier, children were particularly malnourished; more than
50% of the children were stunted (HA <-2.0 SD) and 21.4% were
severely stunted (HA <-3.0 SD). Seventy-four percent of the
caregivers were involved (alone or with another member of the family,
usually the father) in decisions regarding child's feeding. We
interpreted this ability to make decisions as indicating a greater
autonomy for those caregivers. When children were sick, 37% of
caregivers reported seeking advice from older women. Also, 66% of
caregivers reported being satisfied with life. The main reasons for
dissatisfaction were illness of a child (33% of dissatisfied mothers),
lack of income (27%) and heavy workload (23%). In 17% of
dissatisfied caregivers, no particular reason could be given for this
attitude. About 45% of mothers did not receive assistance in their
daily chores, or were helped for only one task, usually for preparation
of meals. The other activity for which mothers reported receiving some
help was fetching water or wood.
|
Results of multiple regression analyses are shown in Table 4.
Significant caregiver and household predictors of child HA are shown in
models 1 and 2, respectively. The variance in HA explained by model 2
was slightly higher than that for model 1. Child age was a strong
predictor of HA even when controlling for other factors, as observed in
univariate analyses. In model 1, consultation of older women was a
significant predictor of child height status. Caregiver's input in
child-feeding decisions and assistance received in her activities were
also positively associated with child HA. Dissatisfied mothers tended
to have shorter children. Model 2 shows that all three household
socioeconomic variables were significant (P < 0.05)
predictors of child HA. The higher the amount of cereals harvested, the
greater the child's height. Bigamy and household dependence on
women's handicrafts for income showed a negative association with
child HA. Model 3 shows the combined effects of caregiver and household
socioeconomic factors. Except for consultation of older women, all
caregiver variables remained significant predictors of HA, even when
household socioeconomic factors were simultaneously entered into the
equation. Moreover, the regression coefficients remained quite stable
when the separate and combined models were compared. The combined model
explained 54% of the variance in the height of children, compared to
36 and 41% for the separate models.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Caregivers who reported a positive outlook on life had higher HA than
children of dissatisfied caregivers. Despite the fact that some
caregivers had specific reasons to justify their dissatisfaction, 17%
of them said that they had no particular reason for having a fatalistic
attitude. According to the review of Zeitlin et al. (1990)
, psychological adjustment and a positive attitude on the
part of caregivers appear to be important for satisfactory child
growth, especially in low-income families living in deprived
environments, although it is possible that the relation also operates
in the reverse direction, with healthier children inspiring more
positive outlooks for mothers. As emphasized in the same review,
caregivers with a positive attitude may have more initiative and may be
more prone to seek help when needed. It may be correct to assume that
caregivers with a negative attitude will have less initiative and be
more difficult for programs to reach.
On the other hand, many caregivers had specific reasons to justify
their dissatisfaction such as illness of a child, lack of income or a
heavy workload. Efforts to foster positive caregiver attitudes could
thus begin by improving women's quality of life. Some caregivers
suggested, for example, that the use of a water pump would lessen their
workload. Providing households with safe drinking water would in turn
contribute to improving children's health and nutritional status.
Prevalence of diarrhea at the time of the survey was found to be high
in the study area (15%), and the presence of diarrhea or fever during
the survey was found to be negatively associated with children's HA in
previous analyses (Bégin and Delisle 1993
).
We found that the number of tasks for which caregivers received
assistance was more important than the task itself. For example, it was
observed that children were taller only when caregivers reported
receiving help for at least two activities, usually for meal
preparation and fetching water, which are two physically demanding and
time-consuming tasks. Caregivers were more likely to be assisted by
another adult for meal preparation, while children were more involved
in fetching water (or wood). Assistance received was highly correlated
with the age of the caregiver (r = 0.41,
P < 0.01), but its association with child nutritional
status was nevertheless independent of caregiver's age. Indeed, when
age of the caregiver was added to the model, it did not change the
relationship between child HA and domestic help available (data not
shown). In Jamaica (Kerr et al. 1978
) and Nigeria
(Morley et al. 1968
), caregivers who received less
support from the family had more malnourished children than caregivers
who received assistance. This suggests that family or social support
may provide some buffer mechanisms to adverse conditions usually
associated with child malnutrition.
Consulting older women in cases of child illness was no longer a significant variable when household variables were included in the model. This caregiver variable was found to be negatively associated with the household variable bigamy. In other words, women in bigamous households were less likely to consult older women than caregivers from monogamous households, and therefore the effect of advice-seeking was probably accounted for by the household variable.
The two variables used as proxies for household income were important predictors of child nutritional status. The amount of harvested cereals was significantly correlated with men's sales of cereals, so the two variables were not included in the same model. Households who could sell part of the harvest were likely more self-sufficient in cereal than those not selling. When women's handicrafts were reported by household heads as a source of family income, it was negatively associated with child growth. Households who reported crafts as an income-generating activity were mainly those who had no cereal sales. Income generated from handicrafts was probably not sufficient to compensate for lack of cereal sales. Also, women involved in handicrafts spent less time on domestic activities (P = 0.07) and had less leisure time (P < 0.01), which was often a moment devoted to child care as frequently mentioned by the mothers during the interviews.
In the study area, bigamy may reflect socioeconomic status, as it is
usually practiced by wealthier households. Bigamy may also be seen as
an advantage if the presence of additional adult females means greater
assistance with child care and domestic chores. However, we found that
bigamy was negatively associated with child nutritional status. This
contradicts a study in Kenya that showed no difference in quality of
care between married women in monogamous and polygamous households
(Borgerhoff-Mulder 1985
). It is possible that in our
study other factors related to bigamy counteracted the positive effect
of the higher socioeconomic status. This issue deserves further
investigation.
The strong positive association between child age and HA indicates that
stunting was less prevalent in older children than in younger ones;
this association was not different with gender. This age pattern of
stunting differs from what is usually observed in deprived
environments. Instead, the observed pattern reflects a secular
deterioration of child nutritional status associated with worsening
socioeconomic and environmental conditions in the area over the
previous 5 y (1982–1987) (Bégin et al. 1997
).
The proximate determinants of child nutritional status are dietary
intake and morbidity status. However, their use as indicators is
limited owing to the difficulty, unreliability and time cost of
measuring them. Therefore, the identification of alternative predictors
of child nutritional status is of interest. The socioeconomic,
demographic and psycho-social factors that we examined at caregiver and
household levels may represent simple tools to identify children at
risk of stunting in the community studied. Indicators pertaining to
caregiver psychosocial aspects or autonomy are particularly
interesting, as their influence was found to be even greater in
situations of severe economic stress (Bégin et al. 1997
).
Although fostering positive caregiver attitudes and empowering poor
families, in particular mothers, may be very beneficial, it may not
necessarily lead to sustained improvement of nutrition. It must be kept
in mind that in most instances, political and economic constraints
remain the major underlying factors influencing child malnutrition. A
caregiver, although autonomous and mentally healthy, may find some
difficulty to provide high-quality child care if there are major
economic and social constraints. In these circumstances, the lack of
good-quality care should not be interpreted as personal flaws or
deficiencies of the caregiver (Engle and Ricciuti 1995
).
As discussed by Pellet et al. (1993),
positive caregiver
attitudes and behaviors may help to alleviate malnutrition, but food
security of households must also be ensured. Moreover, nutrition and
health interventions with components to improve household food security
and health care are more likely to succeed if they are conducted along
with strategies to enhance the caregiver's resources and increase
social support in family and community. Finally, this study was done in
a particular context, i.e., it was preceded by 5 y of worsening
socioeconomic and environmental conditions which contributed to the
secular deterioration of child HA. These unusual conditions might have
affected some of the caregiver behaviors and resources that were
studied here, either by reducing their importance or even eliminating
it. This particular context may reduce the generalizability of our
findings to other developing countries.
In conclusion, this study in a Sahelian rural area shows that, among all the variables examined based on the extended UNICEF model of care, those reflecting psychosocial characteristics of caregivers and socioeconomic status of households were the best predictors of child HA. Maternal height, caregiver workload and income, although showing some association, were not significant predictors of HA when the other factors were controlled. However, caregiver autonomy and satisfaction with life, as well as social support in family, influenced child HA independently from household socioeconomic factors. The existing body of literature demonstrates the importance of various components of child care that improve the nutrition and health of the child. Our paper enhances this growing literature by demonstrating the role of variables illustrated in the UNICEF model of care, such as the resources required by the caregiver, that are generally not assessed when studying the determinants of child nutritional status. Hopefully this will bring more attention on the whole concept of child care when designing interventions to improve child health and nutrition.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Support for this research was provided by a
grant from IDRC, Ottawa. Acknowledgment is also made to the Canadian
Public Health Association, Ottawa, and the Centre Sahel, Québec,
for covering travel expenses.
2 The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 USC section 1734 solely to indicate this fact.
4 Abbreviation used: HA, height-for-age.
Manuscript received May 13, 1998. Initial review completed July 30, 1998. Revision accepted December 15, 1998.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1. Allen L. H., Pelto G. H., Chavez A., Martinez H., Ely R. D., Capacchione C. M. Maternal correlates of infant growth in rural Mexico. Atkinson S. A. Hanson L. A. Chandra R. K. eds. Breastfeeding, nutrition, infection and infant growth in developed and emerging countries 1990:299-306 ARTS biomedical publishers and distributors St. John's Newfoundland, Canada.
2. Bégin F., Alladoumgué M., Nandjingar K., Delisle H. Household dietary adequacy and individual nutritional status: Relationship and seasonal effect in a Sahelian community of Chad. Food Nutr. Bull. 1992;14:304-313
3. Bégin F., Delisle H. Importance de la morbidité et des facteurs maternels et familiaux dans la prédiction de l'état nutritionnel d'enfants en zone rurale, Tchad. Annales de l'ACFAS; 1993;60:212(abs.)
4. Bégin F., Habicht J. P., Frongillo E. A., Jr & Delisle H. The deterioration in children's nutritional status in rural Chad: The effect of mother's influence on feeding. Am. J. Public Hlth. 1997;87:1356-1359
5. Blumberg R. L. Income under female versus male control. J. Family Issues 1988;9:51-84
6. Borgerhoff-Mulder M., Milton M. Factors affecting infant care in the Kipsigis. J. Anthropol. Res 1985;41:231-262
7. Christian P., Abbi R., Gujral S., Gopaldas T. The role of maternal literacy and nutrition knowledge in determining children's nutritional status. Food Nutr. Bull. 1988;10:35-40
8. Christian P., Abbi R., Gujral S., Gopaldas T. Relationship between maternal and infant nutritional status. J. Trop. Pediatr. 1989;5:71-76
9. Delisle H., Alladoumgué M., Bégin F., Nandjingar , Lasorsa C. Household food consumption and nutritional adequacy in wadi zones of Chad, Central Africa. Ecol. Food Nutr. 1991;25:229-248
10. Engle P. L. Maternal work for earnings and child care strategies: Nutritional effects. Child Development 1991;62:954-965[Medline]
11. Engle P. L. Influences of mother's and father's income on children's nutritional status in Guatemala. Social Sci. Medicine 1993;37(11):1303-1312
12. Engle, P. L., Menon, P., Garrett, J. L. & Slack, A. (1997) Developing a research and action agenda for examining urbanization and caregiving: Examples from Southern and eastern Africa. FCND Discussion Paper No. 28.
13. Engle P. L., Menon P., Haddad L. Care and Nutrition: Concepts and Measurement 1997 International Food Policy Research Institute (IFPRI) Washington, D.C.
14. Engle P. L., Pedersen M. E. Maternal work for earnings and children's nutritional status in urban Guatemala. Ecol. Food Nutr. 1989;22:211-223
15. Engle P. L., Ricciuti H. N. Psychosocial aspects of care and nutrition. Food Nutr. Bull. 1995;16:356-377
16. Food and Agriculture Organization (FAO)/World Health Organization (WHO) (1989) Requirements of vitamin A, iron, folate and vitamin B12: report of a joint FAO/WHO expert consultation. FAO, Rome, Italy.
17. Griffiths, M. (1988) Maternal self-confidence and child well-being. Paper presented at the annual meeting of the Society for Applied Anthropology (20–24 April), Tampa, Florida.
18. Guldan G. S., Zeitlin M. F., Beiser A. S., Super C. M., Gershoff S. N., Datta S. Maternal education and child feeding practices in rural Bangladesh. Soc. Sci. Med. 1993;36:925-935
19. Haddad L., Hoddinott J. Women's income and boy-girl anthropometric status in the Côte d'Ivoire. World Development 1994;22:543-553
20.
Kerr M. J., Bogues J., Kerr D. Psychological functioning of mothers of malnourished children. Pediatr 1978;62:778-784
21. Kutty V. R. Women's education and its influence on attitudes to aspects of child care in a village community in Kerala. Soc. Sci. Med. 1989;29:1299-1303
22. Morley D., Bicknell J., Woodland M. Factors influencing the growth and nutritional status of infants and young children in a Nigerian village. Transac. Royal Soc. Trop. Med. Hyg. 1968;62:164-199[Medline]
23. Neter J., Wasserman W., Kutner M. H. Applied linear statistical models: regression, analysis of variance, and experimental designs 3rd edn. 1990 Homewood Irwin, IL.
24. Pellet P., Harrison G., Shuftan C. Positive deviance in child nutrition: a discussion. Ecol. Food Nutr. 1993;30:79-87
25. Popkin B. M., Solon F. S. Income, time, the working mother and child nutriture. J. Trop. Pediatr. Env. Child Hlth. 1976;22:156-166
26. Rabiee F., Geissler C. The impact of maternal workload on child nutrition in rural Iran. Food Nutr. Bull. 1992;14:43-48
27.
Ruel M. T., Habicht J. P., Pinstrup-Andersen P., Grohn Y. The mediating effect of maternal nutrition knowledge on the association between maternal schooling and child nutritional status in Lesotho. Am. J. Epidemiol. 1992;135:904-914
28. SAS Institute Inc. (1985) SAS User's Guide: Statistics, Version 5 Edition. SAS Institute Inc., Cary, N.C.
29. Snedecor G. W., Cochran W. G. Statistical Methods 7th edn. 1980 Iowa State University Press Ames, Iowa.
30. SPSS Inc. (1990) SPSS Reference Guide. SPSS statistical data analysis.
31. Tucker K., Sanjur D. Maternal employment and child nutrition in Panama. Soc. Sci. Med. 1988;26:605-612
32. UNICEF (United Nations Children's Fund) (1990) Strategy for improved nutrition of children and women in developing countries. New York, NY.
33. Wandel M., Holmboe-Ottesen G. Maternal work, child feeding, and nutrition in rural Tanzania. Food Nutr. Bull. 1992;14:49-54
34. Waterlow J. C., Buzina R., Keller W., Lane J. M., Nichaman M. Z., Tanner J. M. The presentation and use of height and weight data for comparing the nutritional status of groups of children under the age of 10 years. Bull. World Health Organization 1977;55:489-498
35. Weisner T. S., Gallimore R. My brother's keeper: Child and sibling caretaking. Current Anthropology 1977;18:169-190
36. Winkvist A. Health and nutrition status of the caregiver: Effect on caregiving capacity. Food Nut. Bull. 1995;16:389-397
37. World Health Organization Measurement of nutritional impact 1979 WHO Geneva, Switzerland.
38. World Health Organization Measuring change in nutritional status 1983 WHO Geneva, Switzerland.
39. Zeitlin M. F., Ghassemi H., Mansour M. Positive deviance in child nutrition 1990 United Nations University Press Tokyo, Japan.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
