![[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}
|
|
|
|
|
||
Departments of International Health and * Biostatistics, Emory University, Rollins School of Public Health, Atlanta, GA 30322
1To whom correspondence should be addressed.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
KEY WORDS: • anemia • stunting • vitamin A deficiency • Honduras • children
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). There is also
clustering at the subnational level; in Brazil, for example, nutrition
problems are concentrated in the northeast (Monteiro et al. 1995
). Throughout the world, much effort goes into identifying
regions, districts or communities with more problems and hence greater
need for resources (Jennings et al. 1991
).
We would expect nutrition problems to cluster in individuals because we
assume that these problems share causal factors. The web of causes
depicted in the UNICEF conceptual framework on malnutrition
(UNICEF 1998
) includes poverty, low levels of education
and other social factors at a basic level. These factors influence the
manifestations of malnutrition through the following three underlying
causes: insufficient household food security, inadequate maternal and
child care, and insufficient health services and an unhealthy
environment. Underlying causes in turn lead to inadequate dietary
intakes and disease, the immediate causes of malnutrition. For any two
nutrition problems, for example, anemia and vitamin A deficiency (VAD),
the overlap in causes, particularly at the underlying level, would be
great. Even though the nutrients involved in causing one deficiency
differ from those of the other, in settings of poverty and disease, we
would expect deficiencies such as anemia and VAD to co-occur or
cluster at the individual level.
Surprisingly, there is very little information in the literature concerning whether nutrition problems do cluster at the individual level. In this report, we investigated the co-occurrence of three common and important nutrition problems among young Honduran children, i.e., VAD, anemia and stunting. Our hypothesis was that any two nutrition problems and all three problems would co-occur at a greater frequency than would be expected by chance.
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We used data from the 1996 Honduras National Micronutrient Survey. All
departments of Honduras were included except for Islas de la
Bahía and Gracias a Dios, which are relatively isolated and
unpopulated. Multilevel cluster sampling, followed by random sampling
of households in each cluster, was used. Anthropometric measurements,
blood samples and sociodemographic data were collected for a total of
1734 children, 12–71 mo of age. Length (children <2 y) or height
(children >2 y) was measured using standardized anthropometric
techniques. Z-scores were computed by comparing the length and
height measurements with the WHO/NCHS reference population using ANTHRO
(ANTHRO 1990
). Hemoglobin concentrations were determined
in the field from finger-prick blood using the Hemocue
ß-hemoglobin test (Ängelholm, Sweden). Serum retinol was
analyzed from capillary blood (Secretaría de Salud 1997
). Because this was a secondary data analysis, human/animal
ethical treatment clearance was not required.
For this analysis, the age range was restricted to children 12–60 mo of age to increase comparability with other surveys and publications. Of 1343 eligible children, 1243 (93%) had data on all three variables of interest (i.e., hemoglobin, serum retinol and length/height).
Anemia (a proxy for iron deficiency) was defined as hemoglobin <11 g/L
(WHO 1968
). Using the cut-off value of <20
µg/L (0.7 µmol/L) for serum retinol
(WHO 1994
), the prevalence of VAD in the sample was
14.2%. To increase power for this analysis, VAD was defined as <30
µg/dL (1.05 µmol/L), which indicates
risk of low vitamin A stores (Olson 1994
) and has been
used to define populations at risk of VAD (Ahmed et al. 1997
, Solano et al. 1997
). Stunting was defined
as <-2 SD of height for age of the WHO/NCHS reference
population (Waterlow et al. 1977
).
Analytic strategy.
Analyses were carried out for two age groups separately, i.e., 12–35.9
mo (n = 633) and 36–59.9 mo (n
= 610). This was done because children <36 mo are often the main
target for nutrition intervention programs (Jennings et al. 1991
); in addition, the prevalence of nutrition problems
differed by age group. Sexes were pooled because results were similar
for boys and girls.
The prevalence of the two-way combinations of VAD, anemia and
stunting (i.e., VAD and anemia, VAD and stunting, and stunting and
anemia) and the three-way combination (i.e., VAD, anemia and
stunting) was computed by age group from contingency tables. A test of
independence among the three variables was carried out by fitting a
log-linear model to the data using the SAS procedure PROC CATMOD
(Version 6.12, SAS Institute, Cary, NC). The observed prevalence of the
two- and three-way combinations was compared to the expected (i.e.,
by chance) prevalence of each, which was computed as the product of the
prevalence of each component nutrition problem. For the comparison, the
adjusted residuals for each combination were examined by subtracting
the expected from the observed value and dividing by the square root of
the expected value using an adjustment factor to allow for the
assumption of a standard normal distribution. To test for significance,
this value was treated as a Z-statistic (Agresti 1990
). The prevalence of stunting in anemic and nonanemic
children, in VAD children and non-VAD children, and in children who
had both VAD and anemia and those who had none or one of the nutrition
problems was also computed. Differences between deficient and
nondeficient children were compared using a 
2 test.
P < 0.05 was considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Not shown is that in younger children, 77% had at least one problem;
of these, 51.4% had more than one problem. In the older children,
64.5% had at least one problem; of these, 44.1% had more than one
problem. Results from the log-linear model showed a lack of
independence among the three variables for both the older and younger
age group [likelihood ratio statistic (df = 4) for younger age
group: 11.91, P < 0.05; likelihood ratio statistic (df
= 4) for older age group: 31.55, P < 0.0001].
The observed and expected proportions of children with two-way and
three-way co-occurrence of VAD, anemia and stunting are shown in
Table 2
for both age groups. The observed proportion was higher than the
expected proportion for all comparisons, and all differences were
significant except for that of stunting and anemia in younger children.
|
|
. For all comparisons, the prevalence of stunting was higher in the
deficient compared with the nondeficient children. For the younger age
group, the comparison of stunting prevalence in children with and
without VAD was significant. For the older children, all three
comparisons were significant. Not shown in the figure is that for the
younger age group, the prevalence of stunting in those who had both
deficiencies was 39% compared with 36% in those who had only anemia,
and 41% in those who had only VAD. Neither comparison was significant.
In the 36- to 59.9-mo age group, the prevalence of stunting in those
who had both deficiencies was 51% compared with 38% in those who had
only anemia, and 48% in those who had only VAD; neither of these
differences was significant.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Although the statistical analysis showed evidence for
co-occurrence, the actual differences between observed and expected
co-occurrence of nutrition problems were very small for most
comparisons. As shown in Table 2
, the differences were 
2% or lower
for all comparisons with the exception of VAD and stunting in older
children. The differences in the prevalence of stunting in deficient
and nondeficient children were also relatively small, with the
exception of older children who had VAD or both VAD and anemia (Fig. 1)
.
These findings have important programmatic implications for nutrition
interventions in Honduras. VAD, anemia and stunting affected large
proportions of Honduran children, 48, 40 and 35%, respectively, among
those 12–35.9 mo old; only about a fourth of children in this age
group were free from all three problems. This suggests that preventive
efforts must be aimed broadly at the population through multiple
strategies, which in combination address these and any other problems
recognized as important. Preventive efforts must be put in place early
in the life cycle to be most effective. Indeed, growth failure may be
irreversible in Central American children >3 y of age
(Martorell et al. 1995
, Schroeder et al. 1995
). Although VAD and anemia can be corrected at any age,
their functional consequences are great during early childhood and may
be long lasting (Lozoff et al. 1991
).
Because co-occurrence of nutrition problems is modest in this population, the efficiency of nutrition interventions would not be improved by targeting children with any one of the conditions. For example, in younger children, the positive predictive value for anemia among stunted children, that is, the proportion of stunted children who are anemic, was 43% compared with an overall prevalence of 40%. Similarly, the positive predictive value for VAD among stunted children was 54% compared with an overall prevalence of 48%. Targeting young children with stunting for nutritional intervention would, therefore, leave out a large proportion of the population with a prevalence of VAD and anemia nearly equal to that of the stunted children selected. In short, our findings suggest that for the nutrition problems considered in this analysis, having one or two problems does not appreciably increase the probability of having another, especially among children 12–35.9 mo of age. From a programmatic perspective, the nutrition problems should therefore be considered virtually independent. These conclusions do not appear to be related to the choice of cut-off point for defining VAD; although power was lower, the analysis of expected and observed frequencies using a cut-off value for VAD of <20 µg/dL (0.7 µmol/L) serum retinol led to exactly the same conclusions as when <30 µg/dL (1.05 µmol/L) was used.
A major strength of this analysis is the use of nationally representative data and large sample sizes. Sample sizes exceeded 600 for each of the age groups examined, and each of the conditions of interest occurred in a fifth or more of the population. To our knowledge, this is the first analysis of co-occurrence of stunting and micronutrient deficiencies at the individual level. Replication of these findings in other settings is required to assess their relevance to other countries. In addition to cross-sectional studies such as this one, it may be useful to carry out longitudinal studies assessing, for example, the age of onset of the deficiencies and how the outcomes relate to dietary intakes and to the underlying conditions associated with poverty.
Manuscript received October 14, 1999. Revision accepted May 12, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1. ACC/SCN Second Report on the World Nutrition Situation. Vol. I, Global and Regional Results 1992 ACC/SCN Geneva, Switzerland.
2. Agresti A. Categorical Data Analysis 1990 John Wiley and Sons New York, NY.
3. Ahmed, F., Hasan, N. & Kabir, Y. (1997) Prevalence of vitamin A deficiency among adolescent female garment factory workers in Bangladesh. In: Report of the XVIII International Vitamin A Consultative Group Meeting, Cairo, Egypt (abs.).
4. ANTHRO (1990) Version 1.01. CDC/WHO, Atlanta, GA.
5.
Habicht J. P. Some characteristics of indicators of nutritional status for use in screening and surveillance. Am. J. Clin. Nutr. 1980;33:531-535
6. Jennings, J., Gillespie, S., Mason, J., Lotfi, M. & Scialfa, T. (1991) Managing Successful Nutrition Programmes. ACC/SCN State-of-the-Art Series Nutrition Policy Discussion Paper no. 8. A report based on an ACC/SCN Workshop at the 14th IUNS International Congress on Nutrition, Seoul, Korea, August 20–25, 1989. ACC/SCN, Geneva, Switzerland.
7. Lozoff B., Jimenez E., Wolf A. W. Long-term developmental outcome of infants with iron deficiency. N. Engl. J. Med. 1991;325:687-694[Abstract]
8. Martorell R., Schroeder D. G., Rivera J. A., Kaplowitz H. J. Patterns of linear growth in rural Guatemalan adolescents and children. J. Nutr. 1995;125(suppl.):1060S-1067S
9. Monteiro C. A., Mondini L., Medeiros de Souza A. L., Popkin B. M. The nutrition transition in Brazil. Eur. J. Clin. Nutr. 1995;49:105-113[Medline]
10. Olsen J. A. Vitamin A, retinoids, and carotenoids. Shils M. E. Olson J. A. Shike M. eds. 8th ed. Modern Nutrition in Health and Disease 1994;1:287-307 Lea and Febiger Philadelphia, PA.
11. Schroeder D. S., Martorell R., Rivera J. A., Ruel M. T., Habicht J. P. Age differences in the impact of nutritional supplementation on growth. J. Nutr. 1995;125(suppl.):1051S-1059S
12. Secretaría de Salud Encuesta Nacional Sobre Micronutrientes Honduras, 1996 1997 Tegucigalpa Honduras
13. Solano, L., Paez, M., Sanchez, A., Ortiz, L., Portillo, Z., Ramos, G. & Callegari, C. (1997) Vitamin A status of preschool children from a community at nutritional risk. In: Report of the XVIII International Vitamin A Consultative Group Meeting, Cairo, Egypt (abs.).
14. UNICEF The State of the World’s Children 1998 Oxford University Press New York, NY.
15. 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. WHO 1977;55:489-498[Medline]
16. World Health Organization Nutritional Anemias: Report of a WHO Scientific Group. WHO Technical Report Series 405 1968:1-37 WHO Geneva, Switzerland
17. World Health Organization Indicators for Assessing Vitamin A Deficiency and Their Application in Monitoring and Evaluating Intervention Programmes. Report of a Joint WHO/UNICEF Consultation 1994 WHO Geneva, Switzerland.
This article has been cited by other articles:
|
|
 |
|
  C. Lopriore, Y. Guidoum, A. Briend, and F. Branca Spread fortified with vitamins and minerals induces catch-up growth and eradicates severe anemia in stunted refugee children aged 3-6 y Am. J. Clinical Nutrition, October 1, 2004; 80(4): 973 - 981. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
