![[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}
|
|
|
|
|
||
,2
* Unité de Formation et de Recherche, Sciences de la Vie et de la Terre, Laboratoire de Biochimie-Microbiologie, Université de Ouagadougou, 01 BP 7021, Burkina Faso; Unité de Recherche 106: "Nutrition, Alimentation, Sociétés" de l'IRD (Institut de Recherche pour le Développement), 01 BP 182, Ouagadougou, Burkina Faso; and ** Unité de Recherche 106: "Nutrition, Alimentation, Sociétés", Centre IRD de Montpellier (WHO collaborating Center for Nutrition), BP 64501, 34394 Montpellier Cedex 5, France
2 To whom correspondence should be addressed. E-mail: prosper.sawadogo{at}ird.bf.
 |
ABSTRACT |
|---|
 TOP ABSTRACT SUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
KEY WORDS: • child feeding practices • dietary diversity • breast-feeding • Burkina Faso, West Africa
Breast-feeding and complementary feeding practices are fundamental to children's survival and development (1). In many developing countries, nutritional problems in infants and young children are closely linked to these practices. Among other things, feeding practices have an impact on physical growth, which is regarded as one of the best indicators of children's well-being (2). However, the relation between the quality of feeding practices during early age and nutritional status are difficult to establish, and, depending on the context and overall living conditions, the influence of feeding factors on children's nutritional status can vary considerably (3). In addition, feeding practices are often complex, change with a child's age, and are seldom all positive or all negative. It is therefore not easy to assess the global quality of feeding practices at the scale of the individual child. In 2002, Ruel and Menon (4) proposed an Infant and Child Feeding Index (ICFI)3 based on an age-specific scoring system that gives points for positive practices in terms of breast-feeding, bottle-feeding, meal frequency, and food diversification. This method takes into account the young child's main feeding practices and expresses them comprehensively through a single summary index. Such an approach has many potential advantages: it summarizes information, facilitates an exploratory diagnosis in a particular situation, and helps target and monitor specific interventions. It may also enable comparisons at an international scale. Using the Demographic and Health Surveys data of 7 Latin American countries, the authors showed that the ICFI was closely related to the mean height-for-age Z-score (HAZ) among children aged 12–36 mo (5). In a later study, the same authors showed that a dietary diversity index calculated over 1 wk, which was one of the components of ICFI, was also related to HAZ (6). This study again used the Demographic and Health Surveys data, but this time in 11 developing countries, including some African countries, and it examined a narrower age range of children, 6–23 mo. To our knowledge, only one published study has considered this kind of relation between ICFI or its components and the nutritional status of young children at a smaller-than-national scale, i.e., in a more homogenous sample (7). Conducted in an African rural area of Senegal, the study found that ICFI was not associated with either height-for-age or the linear growth of children aged 12–42 mo.
We examined the relation between feeding practices and nutritional status of infants and young children living in rural Burkina Faso, West Africa. The data are from a cross-sectional study conducted in 2002 and are based on a representative sample of over 2400 children aged 6–35 mo. We followed the principles of the method proposed by Ruel and Menon (5) to build a composite index of ICFI, which, in our study, was adapted to the characteristics of the context and to available data. The relation between this index and its different components with the children's nutritional status were explored separately in 3 age groups (6–11, 12–23, and 24–35 mo). The consistency between the index and its different components was also studied.
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
350,000 inhabitants unequally distributed in 278 villages. The Gourmantche is the main ethnic group. The mean population density is 41 inhabitants/km2. The main livelihood of the population is farming and stock breeding. Sampling. Eighty villages were randomly selected with a probability proportional to their population, based on data from the 1998 administrative census. Within each village, 12 collective housing units (or "compounds") were randomly selected from an updated list of all the heads of households currently living in the village. All children aged 6–35 mo and their mothers, living in the same compound as the selected heads of households, were included in the study.
Infant and child feeding index. Infant and child feeding practices were assessed through a qualitative 24-h recall of all foods consumed the previous day: breast-feeding, bottle-feeding, number of meals and snacks, and exact composition of the meals ingested. Based on this initial information the ICFI was constructed using the method proposed by Ruel and Menon, which was adapted to the context and to available information. To take into account the age limits of feeding recommendations, the ICFI was compiled separately for the 3 age groups: 6–11, 12–23, and 24–35 mo, respectively.
Breast-feeding. Whatever the age, a score of +1 was attributed to the child who was breast-fed on the day preceding the survey.
Bottle-feeding. Whatever the age, a score of +1 was attributed to the child who was never bottle-fed.
Number of meals and snacks. The number of "true" meals (i.e., porridges, special meals, or family meals) consumed by the child was computed first and then compared with the current age-specific recommendations (8): a score of +2 was given if the recommended level was reached and a score of +1 was given if the number of meals was below the recommendation but different from zero. In addition, a score of +1 was given when the child had consumed additional snacks or leftovers at least twice on the day preceding the survey.
Food diversity. The method proposed by Ruel and Menon takes into account food diversity indices calculated both over the last 24 h and over the last 7 d. Unfortunately, the latter information was not available in our survey. However, we deemed it important to retain the original idea of putting some emphasis on the quality of complementary foods. Consequently, we decided to take into account food diversity in terms of both a food variety score (FVS) and a dietary diversity score (DDS) (9). The FVS corresponds to the number of different food items consumed over the recall period and the DDS refers to the number of different food groups. We took 8 food groups into account: cereals, roots and tubers, nuts and pulses, fruits and vegetables, meat and fish, eggs, milk and dairy products, and fats. For both the FVS and the DDS, the score was divided into terciles separately in each age group. In calculating the ICFI, scores of +2, +1, and 0 points were attributed to the high, medium, and low terciles, respectively.
Together, the ICFI theoretically ranged from 0 to 9 points. In the analyses the index was used after recoding in terciles (poor, average, and good levels of feeding practices) separately for each age-group. The low range and the shape of the population distribution sometimes led to a division into categories whose percentages differ from 33%, especially among the youngest children.
ICFI internal validity was evaluated in several ways: 1) by testing the associations between ICFI and each of its components through chi-square tests, and 2) by assessing the associations between the components themselves through T coefficients of Tschuprow and by calculating the Cronbach 
coefficient (10). An value higher than 0.7 is generally considered to be satisfactory (11).
Anthropometry. The children were weighed naked to the nearest 10 g on mechanical 2–16 kg capacity baby scales (Seca). The mothers were weighed to the nearest 100 g on electronic scales (Seca). The recumbent length (of children up to 2 y) or the standing height (of children 2 y and above and mothers) were measured to the nearest mm with locally made wooden boards equipped with height gauges. The measurements were standardized according to the WHO recommended method (2). Special care was taken to determine the children's age with accuracy from an official document when available, or by using a calendar of local events specifically designed for the province. The height-for-age and weight-for-height indices expressed in Z-scores were computed using the 1978 National Center for Health Statistics/WHO reference using Epinut software (Epi Info Version 6, Centers for Disease Control) (12).
Demographic, socio-economic and health context. Data for each child. A "health monitoring" composite score was computed using information on whether the child had a health card, current attendance at health center activities and at growth monitoring sessions, adequacy of the child's immunization status with respect to the regular Epinut schedule, participation in meningitides immunization campaigns, and in vitamin A distribution. Altogether the score ranged from 0 to 8 points, and its distribution enabled clear separation into "low" and "high" values. In addition, morbidity during the fortnight prior to the survey was recorded (sick yes/no, regardless of illness type).
Data for each mother. Age, matrimonial status, level of education, and personal sources of income were collected. In addition, a composite score of "care for women" was built from the following information: knowledge and use of family planning, obstetrical background (history of stillbirth or infant death), level of prenatal care (number of visits, malaria prophylaxis, and iron supplementation), beneficial practices during pregnancy (improved feeding, alleviation of physical burden, and postpartum rest time), declared ill treatment, power of decision, and autonomy. Altogether the score ranged from 0 to 12 points and was subsequently divided into terciles.
Data for each household. A composite index of the economic level was constructed through a correspondence analysis performed on the matrix of indicator variables that code the housing quality (walls, roof, and floor), possession of current assets (electric lamp, petrol lamp, radio, bicycle, or moped) and possession of cattle (13,14). For a given household, its value on the first principal component of the correspondence analysis gives a coordinate that is interpreted as a summary indicator of its economic level. This index was then divided into terciles. Other collected variables included the number of people living in the household and in the compound, religion, ethnic group, level of education of the head of household, and whether he had a secondary nonagricultural occupation. The hygiene practices within the household were assessed by a score composed from information on water (type and distance to source), latrines, promiscuity with animals, garbage disposal, and a spot-check of the cleanliness of the compound.
Statistical analysis. All analyses were done separately for each age-group. First, bivariate analyses were used to describe the ICFI distribution as a function of the different socio-economic and demographic characteristics of individuals or households. To study the association between ICFI categories and mean anthropometric indices, potential confounding factors were identified on the basis of the internationally recognized Unicef conceptual model of the causes of malnutrition. Factors that can influence both the ICFI value and the children's nutritional status were considered as potential confounders. They were identified through bivariate analyses carried out separately for HAZ and WHZ, with a significance limit of 0.20. These analyses were adjusted for the child's sex and age and also for the mother's height, which are considered the basic factors that determine the child's nutritional status. Two-way interaction terms between the ICFI and each of the potential confounders were systematically tested. As none was significant, all the potential confounders were introduced in the final multivariate models of regression on the mean nutritional indices, together with the basic factors. Adjustment variables for a given nutritional index remained identical within the different age groups and were similar for HAZ and WHZ.
All statistical analyses were performed taking into account the design effect of the study (cluster effect at the village level) using the SAS (statistical analysis system), version 8.02 (15).The generalized linear model was used to test the association of the ICFI with categorical variables (using the Proc GENMOD procedure, option REPEATED = village) and the linear model was used for the association with the mean anthropometric indices (using the Proc MIXED model, option RANDOM = village). The adjusted means were calculated by considering the observed marginal distribution for the independent variables. Associations were considered significant at P < 0.05.
Ethics. The study protocol was approved by the Ministry of Health and by the Ministry of Research of Burkina Faso. Information about the objectives and principles of the study was given to participants in their own language. All individuals surveyed gave oral consent to participate in the study.
|
RESULTS |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Distribution of the ICFI and its components. The distribution of the ICFI components by children's age groups is presented in Table 1. Breast-feeding was almost ubiquitous and was prolonged (36% of children aged 24–35 mo were still breast-fed). Bottle-feeding was almost never resorted to. As expected, the number of meals and snacks increased with age, but among children aged 6–11 mo, 1 out of 2 had no meal on the day preceding the survey. Only 16 different complementary food items were identified for all ages. Animal products (i.e., fish, meat, milk, or eggs), fats, and fruit were rarely consumed by 29%, 15%, and 1.3% of the sample, respectively. On the whole, dietary diversity and variety increased with age. The ICFI distribution per age group and the division into terciles are presented in Figure 1.
|
|
coefficient (Table 2) proved to be good among children aged 6–11 mo ( = 0.79) but it was lower among children aged 12–23 mo ( = 0.63), and rather weak among children aged 24–35 mo ( = 0.38). In all age groups, removing breast-feeding and bottle-feeding from the index improved the value of the Cronbach coefficient.
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Comparison with other studies. As a whole, our results tend to confirm the HAZ results obtained by other authors based on national or multinational samples, even if there were slight differences in the indices used and in the age ranges. Ruel and Menon (5) showed a relationship between ICFI and HAZ among 6–35 mo-old children, and Arimond and Ruel (6) showed a relationship between food diversity (measured over 7 d) and HAZ among 12–23 mo-old children. On the other hand, we did not reach the same conclusions as Ntab et al. (7), who conducted a study in a comparable area (homogenous sample in a rural African region) but found no relationship between ICFI and HAZ or linear growth. The relatively small size of the sample (n = 500) and the wider age group (12–42 mo) could explain this lack of association.
To our knowledge, our study is the first to show that mean WHZ decreased as children aged 12–23 mo moved from the highest to the lowest tercile of ICFI. This is an important finding because this is the age when children display a clear drop in WHZ, which is probably associated with high rates of infection. The result suggests that better feeding practices are particularly important in preventing excessive loss in WHZ at this age. However, we also observed a higher mean WHZ among infants aged 6–11 mo belonging to the lower ICFI category, which included >50% of the infants who did not receive complementary food. This could be due to the strong effect of breast milk displacement in this age group.
Dietary diversity. Several studies, some of which were carried out in sub-Saharan Africa, showed that improved dietary diversity or variety was a key point for the improvement of global feeding practices, and that a food diversity index was associated with children's nutritional status (3,6,16,17). We found similar results characterized by three points: 1) both dietary diversity and variety scores were included in the ICFI; 2) these scores were measured only over 24 h; 3) we did not find a relationship between a particular food group and children's nutritional status. This latter point may mean that dietary diversity in itself matters more for the quality of the diet than the consumption of a particular food group. The main justification for including both FVS and DDS in the ICFI, which essentially capture the same construct, was to emphasize the quality of complementary foods in the scoring system. In addition, we assert that this captures slight differences in the quality of the diet, i.e., when several foods of the same group are consumed. We verified this point by examining the results for ICFI calculated with and without FVS. When FVS was removed, the associations between ICFI and mean WHZ were lowered (among both age groups of 6–11 and 12–23 mo), whereas they remained almost identical between ICFI and mean HAZ (results not shown). It has also been pointed out that a 24-h recall cannot appropriately reveal the actual quality of the diet, mainly because of possible variations from one day to another (18). In our context, very few different food items were identified in the overall sample; thus, the chosen indicators may reflect the poverty of the diet, which hardly varies over time. As a matter of fact, we already showed that the nutritional quality of the complementary foods was particularly poor in this area (19). All of these factors lead us to the conclusion that nutritional status is associated with diversification of the daily diet.
Breast-feeding and children >2 y. It could be considered as an anomalous finding in our study that the ICFI was negatively associated with nutritional status in children >24 mo of age. In fact, this is related to our scoring system, which continues to add 1 positive point for breast-feeding in this age group. This association has already been identified as "reverse causality" by other authors (20,21). The hypothesis is that mothers tend to prolong the breast-feeding of malnourished children. It is worthwhile to point out that, beyond the age of 24 mo, there is currently no scientific basis for recommending breast-feeding nor are there recommendations for an optimal number of meals. Nevertheless, in our study the phenomenon was observed among children aged 24–35 mo and also among those aged 12–23 mo, despite the low rate of nonbreast-fed children in the latter age group. In addition, among the 24–35-mo age group, a lower mean HAZ was observed in the class of children consuming a higher number of meals (see Table 4) in which there were also a higher proportion of breast-fed children (data not shown), suggesting that the reverse causality hypothesis may also apply to the number of meals. In any case, taken together, these results stress that the duration of breast-feeding as well as the number of meals can be contextual.
Challenges in constructing a composite ICFI. The first consideration in constructing a composite ICFI is the cross-sectional nature of data collection, whereas infant growth results from long-term processes. We therefore have to assume that the feeding practices measured at a given time adequately assess previous practices. We tried to add to the ICFI scoring system some retrospective information about past feeding practices that were available in our study. These were practices whose impact on child nutritional status is well known, such as initiation of breast-feeding (elimination of colostrum, and delaying breast-feeding) (22), or the timing of introducing complementary foods (23,24). Although this led to changes in the distribution of ICFI, it did not modify its association with anthropometric indices.
Furthermore, we faced additional difficulty with respect to the choice of the cutpoints to define the ICFI categories, owing to the low range of the index values (from 1 to 9) and to the shape of its distribution. This was especially true for children aged 6–11 mo, among whom more than half did not consume complementary foods. Yet, similar results were found when only 2 categories for the index in this age group were considered (results not shown). Nevertheless, any difference in the construction of the index can change its characteristics and thus its association with child nutritional status (25).
Among the many difficulties creating a universally applicable ICFI is the lack of consensus in defining positive and negative practices in order to attribute age-specific points in the scoring system. Our results highlight the need for additional research on optimal feeding practices beyond the age of 2 y. We also decided to separate the number of true meals and the consumption of snacks. However, there is no clear definition of what should be considered a real meal or a snack, or how the latter should be accounted for. In the case of food diversity, the option chosen in our study, as in most previous studies (5,6,26), was to divide the variety and diversity scores for each age group into terciles. Yet it is obvious that the limits of the categories cannot be the same in all contexts; this also applies to the food groups and their exact definition (26). In this respect there is no mention of an objective threshold referred to as the necessary diversity level to ensure optimal growth. As far as we know, no such information is available, even if efforts are currently underway to harmonize methods of constructing diversity indices and to validate them with regard to real intakes of micronutrients (27).
Finally, within our sample, almost all the children were breast-fed up to 2 y, whereas bottle-feeding was rare. Consequently, as shown in the index-reliability analysis, these 2 ICFI components lacked internal consistency with the global index because there was almost no variability in these factors for children up to 2 y, and because the association was in the reverse direction beyond 24 mo. Consequently, the only reason to include these variables in an ICFI are: 1) to retain the conceptual objective of capturing the multidimensional nature of infant feeding, and 2) to allow for international comparisons, because in other contexts these variables may be more meaningful. Nevertheless, for this purpose our results indicate that it is better to use single variables rather than composite indices that can mask the different practices they include. However, locally developed and adapted indices are relevant for other purposes such as monitoring progress and evaluating the impact of interventions within a specific context.
In conclusion, the contextual characteristics of our study, the differences in the measurement of the feeding practices, and the real meaning of each practice could be sources of difficulty for building and interpreting a composite feeding index, and perhaps this is also a source of confusion for defining its relationship within nutritional indices. Despite the lack of a standard definition, and despite the variations in the methods used to collect data and to construct diversity and variety scores, our results confirm the existing literature and suggest that dietary diversity is positively associated with anthropometric indicators of nutritional status among young children (6,26). Thus, dietary diversity can be considered a good proxy in assessing the global quality of infant and child feeding practices.
|
FOOTNOTES |
|---|
3 Abbreviations used: DDS, dietary diversity score; FVS, food variety score; HAZ, height-for-age Z-scores; ICFI, infant and child feeding index; WHZ, weight-for-height Z-scores.
Manuscript received 26 September 2005. Initial review completed 1 November 2005. Revision accepted 20 December 2005.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
1. WHO. Complementary feeding of young children in developing countries: a review of current scientific knowledge. Geneva: WHO; 1998.
2. WHO. Physical status: the use and interpretation of anthropometry. WHO technical report series no 854. Geneva: WHO; 1995.
3. Hatloy A, Hallund J, Diarra MM, Oshaug A. Food variety, socioeconomic status and nutritional status in urban and rural areas in Koutiala (Mali). Public Health Nutr. 2000;3:57–65.[Medline]
4. Ruel MT, Menon P. Creating a child feeding index using the demographic and health surveys: an example from Latin America. FCND discussion paper no 130. Washington, DC: IFPRI; 2002.
5. Ruel MT, Menon P. Child feeding practices are associated with child nutritional status in Latin America: innovative uses of the demographic and health surveys. J Nutr. 2002;132:1180–7.
6. Arimond M, Ruel MT. Dietary diversity is associated with child nutritional status: evidence from 11 demographic and health surveys. J Nutr. 2004;134:2579–85.
7. Ntab B, Simondon KB, Milet J, Cisse B, Sokhna C, Boulanger D, Simondon F. A young child feeding index is not associated with either height-for-age or height velocity in rural Senegalese children. J Nutr. 2005;135:457–64.
8. Dewey K. Guiding principles for complementary feeding of the breast-fed child. Washington, DC: PAHO/WHO; 2003.
9. Hatloy A, Torheim LE, Oshaug A. Food variety–a good indicator of nutritional adequacy of the diet? a case study from an urban area in Mali, West Africa. Eur J Clin Nutr. 1998;52:891–8.[Medline]
10. Moret L, Mesbah M, Chwalow J, Lellouch J. [Internal validation of a measurement scale: relation between principal component analysis, Cronbach's alpha coefficient and intra-class correlation coefficient] Rev Epidemiol Sante Publique. 1993;41:179–86.[Medline]
11. Bland JM, Altman DG. Statistics notes: Cronbach alpha. BMJ. 1997;314:572.
12. Dean AG, Dean JA, Coulombier D, Brendel KA, Smith DC, Burton AH, Dicker RC, Sullivan K, Fagan RF, Arner TG. Epi Info, version 6: a word processing, database, and statistics program for epidemiology on microcomputers. Atlanta, GA: Centers for Disease Control and Prevention; 1995.
13. Traissac P, Delpeuch F, Maire B, Martin-Prevel Y, Cornu A, Trèche S. [Building a synthetic index of the households' economic level in nutritional surveys. Application examples in the Congo] Rev Epidemiol Sante Publique. 1997;45 Suppl 1:S114–5.
14. Sahn DE, Stifel DC. Assets as a measure of household welfare in developing countries. In: Sherraden M, Morris L, editors. Inclusion in asset building: research and policy symposium. St. Louis, MO: Center for Social Development; 2002. p. 33.
15. SAS Institute Inc. The SAS system, version 8. Cary, NC: SAS Institute Inc.; 1999.
16. Onyango A, Koski KG, Tucker KL. Food diversity versus breastfeeding choice in determining anthropometric status in rural Kenyan toddlers. Int J Epidemiol. 1998;27:484–9.
17. Tarini A, Bakari S, Delisle H. [The overall nutritional quality of the diet is reflected in the growth of Nigerian children] Sante. 1999;9:23–31.[Medline]
18. Piwoz EG. Improving feeding practices during childhood illness and convalescence: lessons learned in Africa. Washington, DC: Academy for Educational Development; 1994.
19. Sawadogo SP, Martin-Prével Y, Savy M, Kameli Y, Traoré SA. [Infant feeding practices in a rural area in Burkina Faso (Gnagna Province): description and nutritional consequences]. In: Brouwer ID, Traoré SA, Trèche S, editors. Food-based approaches for a healthy nutrition in West Africa (Proceedings of the second international workshop). Ouagadougou, Burkina Faso: Presses Université de Ouagadougou; 2003. p. 317–27.
20. Marquis GS, Habicht JP, Lanata CF, Black RE, Rasmussen KM. Association of breastfeeding and stunting in Peruvian toddlers: an example of reverse causality. Int J Epidemiol. 1997;26:349–56.
21. Simondon KB, Simondon F. Mothers prolong breastfeeding of undernourished children in rural Senegal. Int J Epidemiol. 1998;27:490–4.
22. WHO. HIV and infant feeding: a guide for health care managers and supervisors. Report no WC 503.2. Geneva: WHO; 2003.
23. Simondon KB, Simondon F. Age at introduction of complementary food and physical growth from 2 to 9 months in rural Senegal. Eur J Clin Nutr. 1997;51:703–7.[Medline]
24. Hop LT, Gross R, Giay T, Sastroamidjojo S, Schultink W, Lang NT. Premature complementary feeding is associated with poorer growth of Vietnamese children. J Nutr. 2000;130:2683–90.
25. Ruel MT, Brown KH, Caulfield LE. Moving forward with complementary feeding: indicators and research priorities. FCND discussion paper no 146. Washington, DC: IFPRI; 2003.
26. Onyango AW. Dietary diversity, child nutrition and health in contemporary African communities. Comp Biochem Physiol A Mol Integr Physiol. 2003;136:61–9.[Medline]
27. FAO, WHO, IFPRI. Workshop on dietary diversity and dietary quality. Rome: FAO; 11–13 October 2004.
This article has been cited by other articles:
|
|
 |
|
  N. Mpontshane, J. Van den Broeck, M. Chhagan, K. K. A. Luabeya, A. Johnson, and M. L. Bennish HIV Infection Is Associated with Decreased Dietary Diversity in South African Children J. Nutr., September 1, 2008; 138(9): 1705 - 1711. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. K. Saha, E. A. Frongillo, D. S. Alam, S. E. Arifeen, L. A. Persson, and K. M. Rasmussen Household Food Security Is Associated with Infant Feeding Practices in Rural Bangladesh J. Nutr., July 1, 2008; 138(7): 1383 - 1390. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. M Moursi, Y. Martin-Prevel, S. Eymard-Duvernay, G. Capon, S. Treche, B. Maire, and F. Delpeuch Assessment of child feeding practices using a summary index: stability over time and association with child growth in urban Madagascar Am. J. Clinical Nutrition, May 1, 2008; 87(5): 1472 - 1479. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Savy, Y. Martin-Prevel, P. Traissac, S. Eymard-Duvernay, and F. Delpeuch Dietary Diversity Scores and Nutritional Status of Women Change during the Seasonal Food Shortage in Rural Burkina Faso J. Nutr., October 1, 2006; 136(10): 2625 - 2632. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
