![[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
* Department of Food Technology and Nutrition, Faculty of Agriculture and Applied Biological Sciences, Ghent University, B-9000 Ghent-Belgium;
Nutrition and Child Health Unit, Institute of Tropical Medicine, B-2000 Antwerpen, Belgium; and
** Department of Food Science, Chalmers University of Technology, S-402 29, Goteborg, Sweden
2To whom correspondence should be addressed. E-mail: pkolsteren{at}itg.be.
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
KEY WORDS: • complementary feeding • iron deficiency • anemia
Growth and the chance of survival at birth are determined to a large extent by the intrauterine period (1). Exclusive breast-feeding for the first 4–6 mo protects the infant from nutritional deficiencies and decreases the stress of infection. Young infants are most vulnerable at the time complementary foods are introduced. Traditional complementary foods are often bulky, have a low energy density and contain insufficient amounts of micronutrients, in particular iron and zinc (2,3). Complementary foods are largely cereal based and contain considerable amounts of phytate, which negatively affects micronutrient bioavailability, thereby inducing deficiencies in minerals. They are also not without risk of contamination (4). Low infant feeding frequency further contributes to the undernutrition of children. Surveys carried out in Tanzania (5) indeed showed that most children are fed only 2 or 3 times a day.
Iron deficiency anemia is an important nutritional problem in Tanzania. It is estimated that 45% of the children < 5 y old are suffering from nutritional anemia (6). It is more prevalent in infants and pregnant women and is usually the result of low bioavailability of dietary iron (7). Although the effects of iron deficiency anemia are reversible, including impaired intellectual development, this will rarely be the case under the local suboptimal conditions. Thus, despite the fact that the diagnosis of anemia is fairly simple and treatment is inexpensive, the prevalence of anemia remains high (8).
The most frequently used strategies to correct iron deficiency are food fortification and/or iron supplementation (9). Local staples have been fortified with ferrous sulfate (10), zinc sulfate (11), vitamin A, and vitamin C (12–14). The effectiveness of large-scale fortification programs, however, is reduced due to factors such as cost, constant availability, timely distribution of fortificants, and compliance with the prescribed fortificant (15). Similar experiences were observed in many places with iron supplementation (16).
The above-mentioned limitations of fortification and supplementation underline the importance of preventing growth faltering and micronutrient deficiencies, such as iron deficiency anemia through a food-based approach (7,17). Food modification approaches that employ natural processes such as germination to combat micronutrient deficiencies and improve the growth of infants deserve more attention because they are likely to be more sustainable in the long run (4).
Based on the knowledge of the traditional complementary foods (CFs)3 in Tanzania collected by the authors earlier, a processed CF prepared from locally produced crops was formulated. The present study aimed to compare the growth and iron status of Tanzanian infants, from 6 to 11 mo, provided processed and unprocessed (placebo) CF. The main hypothesis was that a processed CF with higher energy density and lower concentration in phytates and tannins would improve growth of infants and decrease iron deficiency.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
300 km west of Dar-es-Salaam. Kilosa is a district with a population of 350,000 inhabitants. Kilosa was chosen for this study because it is among the districts in the Morogoro region that have a high prevalence of iron deficiency anemia (18). Design of the study. The study was set up as a double-blind, randomized, controlled trial in which the main investigator and the mothers had no knowledge of the type of food given to the infants (Fig. 1). Processed complementary food constituted the intervention group and the unprocessed food served as the control. Infants were continuously enrolled when they reached the age of 6 mo and assigned to the respective groups on the basis of their previously determined allocation. For this, all parents with infants < 6 mo old were contacted and invited to participate in the study. Parents of 364 infants agreed to participate and were randomly assigned to treatment or control. We expected a number of dropouts in the period between randomization and actual enrolment; therefore, we increased the number recruited to be sure to reach the minimal required number of infants for the study. Allocation to the treatment or control group was determined using a bloc randomization technique. The main investigator was informed of group assignments at the end of the data collection. The 2 types of CF were distributed until the infants reached the age of 12 mo. Mothers were free to give their infants any other food of their choice in addition to the CF provided during the study. Measurements were taken twice, at the age of 6 and 12 mo, with a malaria blood smear at 9 mo. Verbal consent was sought from mothers for their infants to participate in the trial. The ethics committee of the Tanzania Food and Nutrition Centre and University of Gent reviewed the protocol and gave approval for the trial.
|
The processed CF consisted of 65.2% finger millet, 19.1% kidney beans, 8% peanuts, and 7.7% mango purée. Finger millet and kidney beans were washed and soaked in preboiled water for 2 and 7 h, respectively, and germinated for 48 h at 30°C. Later the lot was autoclaved and then solar dried for 6 h. Peanuts were roasted in an oven at 150°C for 20 min. Mangoes were washed, peeled, sliced, and the purée extracted. Later the purée was dried in a solar drier for 12 h. The ingredients were mixed and milled to make a flour. The unprocessed CF was a blend of finger millet, kidney beans, peanuts and dried mango purée, mixed in the same proportion as the processed CF, and then milled to make a flour. The director at the production site did the packing and labeling. Packages and labels were identical and the 2 preparations were visually indistinguishable. The label contained preparation (cooking) instructions. Before the intervention, an acceptability trial for the CF was done involving 50 mothers with their infants.
Quality control was observed from the purchase of raw materials to processing and distribution. For qualitative and quick examination for aflatoxin contamination, a UV lamp (Type B-4 W Kurtz-und langwellig) was used. At various CF production stages, samples were checked for contamination with fumonisins (RIDASCREEN FAST Fumonisin kit, Biopharm), aflatoxins (RIDASCREEN FAST Aflatoxin kit, Biopharm), and cyanides (19). Fumonisins and aflatoxin concentrations were below the recommended limits of 1 mg/kg and 2 µg/kg, respectively, and cyanides were not detected. Outgrowth of pathogens including Bacillus cereus, Staphylococcus aureus and Clostridium perfringens was investigated for each batch of CF according to the HACCP plan (20) and found to be <100 colony forming units/g (21).
Every 2 wk, 1 and 1.6 kg of CF were allocated for infants 6–8 mo and 9–11 mo old, respectively. On a daily basis, each child was supplied with 69 and 113 g dry matter of CF, providing 1194 and 1956 kJ/d for infants aged 6–8 and 9–11 mo, respectively (22). The detailed formulation of the CF is described in Mbithi-Mwikya et al. (23).
Mothers came to collect the food every 2 wk. The MCH nurses supervising the mothers had a list of the infants and recorded every food collection. They demonstrated to the mothers how to prepare the CF (although the packets had a self-explanatory label) and reminded them how much in local measures (tablespoons) of the CF powder to use each day. The CF for 1 d was prepared once. The nurses also advised the mothers to add 1–2 teaspoonfuls (5–10 mL) of oil to the preparation. The goal was to administer 1151 kJ/d (275 kcal/d) and 1883 kJ/d (450 kcal/d) from CF. In case of absence, the nurse made certain that a message was sent to the responsible mother to collect her consignment on the same day. Nutrition officers from the center followed up the mothers in their homes at least twice each week to make sure the CF was prepared correctly.
Weight and length were recorded at enrollment (age 6 mo) and at the end of the trial (age 12 mo). Recumbent lengths of the infants were measured to the nearest 0.1 cm using an infant measuring board (Perspective Enterprises). Weight of infants was measured to the nearest 100 g using a Salter scale (Model 235 6S) with a capacity of measuring up to 25 kg. The infant’s birth weight was obtained from his/her clinic card. The MCH nurse is responsible for measuring the infant’s weight immediately after birth.
Hb concentration was measured from a finger prick blood sample by the Hemocue B-Hemoglobin System (Hemocue AB) (24,25). Zinc protoporphyrin (ZP) was determined on a drop of blood with a portable hematofluorometer (Aviv Biomedical). The instrument was standardized using control solutions from the same company (26,27). The analyses were performed at a later date.
Blood smears, stained with Giemsa, were analyzed within 10 h for quantitative determination of malarial parasites. Malaria parasite counts were made per 200 white blood cells. Infants found infected with malaria and those who became ill at any time during the trial were treated free of charge at the respective health centers.
A 24-h dietary recall was conducted among mothers of 137 randomly selected infants aged between 6 and 11 mo; this was 50% of the total number of infants and they were equally distributed among the 2 intervention groups. A new appointment was fixed for others who were not found at home at the time of the visit. Random numbers were used to select infants from each group. The dietary recall interviews were conducted in the home by a nutritionist accompanied by a village health worker. The mother was requested to show the type and amounts of foods the infant had actually consumed over the last 24 h. The amount of food consumed by the infant was weighed using a digital weighing scale (Tefal scales) with an accuracy of 0.5 g or measured by a measuring cylinder (Pyrex-UK) with an accuracy of 0.5 mL in the case of volumes. Spilled food was not estimated. Macro- and micronutrient contents of the foods were calculated using the FAO food composition table (28) with Excel Office 2000.
Samples of CF weighing 250 g from each production batch were analyzed for protein [Kjedahl AOAC method 920.87 (19)], fat [Weibull method (29)], and iron [atomic absorption spectrophotometry AOAC method 970.12 (19)]. The nitrogen protein conversion factor for millet was 5.83, and 5.3 for beans and peanuts. The solubility of iron was determined by the method of Svanberg and Sandberg (30) and phytate using the technique of Haugh and Lantzsch (31).
Infants were excluded from the trial if they had received blood transfusions or presented a health condition that required further treatment as assessed by a medical doctor at the time of enrollment. These excluded infants were referred to the local health facilities for appropriate treatment. All infants were examined for malaria at 6 mo of age and treated accordingly. Because all infants were breast-fed and Kilosa is a malaria endemic area, anemia was considered to be due to malaria in the standard protocols. At the end of the trial, the infants were reevaluated and treated for most likely etiology of anemia by the supervising medical doctor.
Statistical analysis. Data were entered in EPI-INFO (version 6.04d; Centers for Disease Control and Prevention, WHO, 1996), and analysis was done using the Stata 8.0 package (Stata version 8.0; STATA). Weight-for-length and length-for-age Z-scores were computed using EPINUT according to National Centre for Health Statistics Standards of 1977.
Descriptive statistics were done on each variable to identify outliers and assess the normal distribution of continuous variables. Outliers were defined from the box plot as values more extreme than the 3-interquatile range of the box. In the presence of outliers, a new variable was created excluding these values. However, in case of doubt, the dataset was cross-checked with the original data in the rosters. All tests were done first with the original variable, and then redone with the new variable to assess the influence of such outliers. Normal distribution of continuous variables was appraised by a Kolmogorov Smirnoff test. In case of severe departure from normality, the variables were log-transformed.
The 
error was set at 5% in all tests. The strategy of data analysis was set in 2 steps. First, a difference at 12 mo of age between the 2 intervention groups was assessed for each primary outcome. These primary outcomes were mean ZP and mean Hb. Differences in anthropometric indicators, i.e., mean weight-for-length Z-score (WLZ) and length-for-age Z-score (LAZ) at 12 mo of age were also examined. A standard t test was used for continuous variables, and a 
2 test for categorical variables. Similarly, the general trend in main outcomes, between the beginning and the end of the trial, were assessed by applying a paired t test or a McNemar test for categorical variables.
Second, a logistic regression analysis was applied using the EVW model (32). Continuous variables were transformed into categorical variables as indicated below. The dependent variables were as follows: 1) high ZP at 12 mo of age (ZP > 5 µg/g Hb, coded 0/1) for the first set of models, 2) anemia (Hb < 110 g/L, coded 0/1) for the second set of models, and 3) LAZ less than –2 SD, coded 0/1 for the third set of models.4
The exposure variable was the type of CF received. The following covariates were inserted into the initial model because they were potential confounding factors or effect modifiers (biological or environmental): 1) child parameters: ZP > 5 µg/g Hb at baseline (0 = no, 1 = yes), Hb < 110 g/L at baseline (0 = no, 1 = yes), blood smear positive for malaria at age 9 mo (0 = no, 1 = yes), birth weight < 2500 g (0 = no, 1 = yes), LAZ less than –2 Z-score at baseline (0 = no, 1 = yes), sex (0 = girl, 1 = boy), season at entry in the study (0 = harvest season, 1 = other season); 2) mother’s parameters: mother’s education lower than primary school (0 = no, 1 = yes), mother living alone (0 = no, 1 = yes), mother’s age < 20 y (0 = no, 1 = yes), parity 
3 (0 = no, 1 = yes), BMI (0 = 18.5, 1 = < 18.5), income < 10,000 shillings (0 = no, 1 = yes).
All covariates were considered to be potential effect modifiers and introduced in the initial model as product terms involving E (exposure to processed food). The presence of multicollinearity and other numerical problems in regression analyses was appraised by verifying the presence of a high estimated SE for the regression estimates (33). Then a hierarchical backwards elimination procedure was applied to eliminate nonsignificant variables.
Removal of variables was at P > 0.05 for the likelihood ratio test. First effect modifications were tested using a likelihood ratio test for the entire collection of interaction terms (32). Then variables were removed one by one according to the likelihood ratio test. The model including the remaining covariates was considered the gold standard. Confounding was then assessed by monitoring changes in the effect measure (odds ratio) for subsets of covariates. The subset of covariates included in the final model was the one allowing the best gain in precision.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
To verify compliance, nutrition officers from the center followed up the mothers at home to make sure the CF was prepared correctly. Compliance was considered to be good if the mother collected the CF, correctly prepared it, and fed it to her infant according to instructions. Surprise visits to the home were made once or twice each week. However, we were not able to fully monitor compliance in the strict sense.
Baseline characteristics.
The characteristics of infants, mothers, and household variables in both groups were similar at enrollment of the infants, aged 6 mo (Table 1) with the exception of WLZ. Mothers’ education (P = 0.11), marital status (P = 0.36), and family income (P = 0.43) were not significantly different between the intervention and control group (2 test).
|
|
|
|
No significant interaction between covariates and the type of CF was detected (likelihood ratio test for set 1, P-value 0.20; likelihood ratio test for set 2, P-value 0.11; likelihood ratio test for set 3, P-value 0.11) with the logistic regression.
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
We cannot exclude, however, that under less supportive conditions, the energy denser food would yield different results. Indirect evidence for this can be found in the considerably higher feeding frequency for the unprocessed food. Two meals were sufficient to provide the necessary amount of processed CF, whereas 5–6 were required for the unprocessed CF. We introduced a bias into the study by continuously emphasizing the amount of CF to be eaten each day. In settings in which time constraints of mothers limit feeding frequency, differences in total intake can be expected. The study also has certain limits because we cannot exclude with certainty that the CF was shared with other family members. However, there are other studies that also documented negative results. In Ghana, the effects of feeding weanimix and 3 other locally formulated, centrally processed CFs on nutritional status of breast-fed infants (6–12 mo of age), were not different between intervention groups in weight or length gain, Hb, and hematocrit values (3).
Stevens and Nelson (35) investigated the effect of feeding 6-mo-old infants a milk formula with no iron compared with one with 12.0 mg Fe/L for 12 mo and found no differences in mean Hb, ferritin concentrations, and growth between the 2 groups of infants. Furthermore, a multicountry study, investigating the effect of supplementation with high energy density CF fortified with minerals and vitamins, on weight and linear growth of 4- to 7-mo-old infants in 4 developing countries, found no significant difference in overall linear growth, although there was an effect of the micronutrients in stunted but not in nonstunted children (36).
Some studies did find a difference, but they differed in a number of important ways. In the study of Chinamma and Golpaldas (37), 2 groups of children aged 6–24 mo were fed either a high-energy, low-viscosity complementary food or a high-energy, high-viscosity food for 6 mo. The former group increased significantly more in weight and length than the latter. The difference between that study and ours is that the children were restricted to 1 experimental meal/d, although they were allowed to consume that single meal ad libitum. In a sense, therefore, 1 group was consuming less experimental food each day. In our study the mothers were encouraged to feed their infants the entire daily amount of CF, which means that they could have increased the feeding frequency up to 5–6 times/d.
Stunting is a very complex multicausal problem, and it is probable that other nutrients were limiting or that the local conditions did not allow a reversal of the trend or stabilization. Similar trends were observed by Martorell et al. (1), whereby 25% of infants at baseline (4 mo old) were already stunted, presumably due to intrauterine growth retardation (IUGR). They concluded that infants who experience IUGR usually never completely catch up in size to their normal-birth-weight peers even when raised under optimal conditions. Similar observations were made in Indonesia (38,39).
Despite an overall significant increase in mean Hb, a large proportion of the infants were still anemic at the end of the study even when low birth weight was controlled for in the analysis. An important cause of anemia in the present population is malaria. At 6 mo, we found malaria parasites in 50% of samples. The Hb concentration was 1.29 g/L lower at mo 12 in infants with a positive blood smear at mo 6 and mo 12 (t test P < 0.0001), whereas ZP concentration was 1.84 µg/g Hb higher (t test P < 0.0001). Mortality in the study population was also very high with 19 deaths in the 309 supplemented infants (61 per 1000). Failure to restore Hb concentrations to normal could be explained by the continuous malaria reinfection among infants despite the treatment given when they became ill. Msuya and Curtis (38) indeed showed that in Tanzania, almost all children became reinfected within 2–4 wk after effective malaria parasite clearance with sulfadoxine (fansidar).
Iron status, as defined by erythrocyte ZP, was not significantly different between the processed and unprocessed CF groups, although iron status improved in general. This was rather unexpected because the groups consumed equal amounts of CF, and the laboratory results showed a substantial increase in solubility, a parameter commonly used as a proxy for bioavailability. The iron content in both processed and unprocessed CFs was low, and a minimal bioavailability of 10% is required to provide the recommended daily amount of 0.6–0.7 mg absorbed iron. The reduction in phytates in the processed CF (34%), which improved iron solubility to 19%, might not have been enough to make a clinical difference. For a metabolically significant improvement in bioavailability, Hurrell (39) postulated that phytates must be reduced by >95%. Finally, infants might have met their basic iron needs, but the food content was insufficient to meet increased needs and to permit recovery from iron deficiency (40).
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
3 Abbreviations used: CF, complementary food; LAZ, length-for-age Z-score, WLZ, weight-for-length Z-score; Hb, hemoglobin; ZP, zinc protoporphyrin.
4 The weight-for-height indicator was not considered because 99% of the children were in the normal range (mean ± SD).
Manuscript received 2 September 2003. Initial review completed 28 October 2003. Revision accepted 5 February 2004.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
1. Martorell, R., Ramakrishnan, U., Schroeder, D. G., Melgar, P. & Neufeld, L. (1998) Intrauterine growth retardation, body size, body composition and physical performance in adolescence. Eur. J. Clin. Nutr. 52(suppl. 1):S43-S52.
2. Bennett, V. A., Morales, E., Gonzalez, J., Peerson, J. M., Lopez de Romaña, G. & Brown, K. H. (1999) Effects of dietary viscosity and energy density on total daily energy consumption by young Peruvian children. Am. J. Clin. Nutr. 70:285-291.
3. Lartey, A., Manu, A., Brown, B. H., Peerson, J. M. & Dewey, K. G. (1999) A randomised community-based trial of the effects of improved, centrally processed complementary foods on growth and micronutrient status of Ghanaian infants from 6 to 12 months of age. Am. J. Clin. Nutr. 70:391-404.
4. World Health Organization (1998) Complementary Feeding of Young Children in Developing Countries: A Review of Current Scientific Knowledge 1998 WHO Geneva, Switzerland.
5. UNICEF (1998) Household Food Security Situation in Selected Districts in Tanzania: Knowledge, Attitude and Practices (KAP) of Community Leadership. UNICEF study report 1998:1-20 .
6. UNICEF (1999) The State of the World’s Children 1999 Oxford University Press Oxford, UK.
7. Fox, T. E., Eagles, J. & Fairweather-Tait, S. J. (1998) Bioavailability of iron glycine as a fortificant in infant foods. Am. J. Clin. Nutr. 67:664-665.[Abstract]
8. Kolsteren, P. W., Rahman, S. R., Hilderbrand, K. & Diniz, A. S. (1999) Treatment for iron deficiency anaemia with a combined supplementation of iron, vitamin A and zinc in women of Dinajpur, Bangladesh. Eur. J. Clin. Nutr. 53:102-106.[Medline]
9. Stoltzfus, R. J., Albonico, M., Chawaya, H. M., Tielsch, J. M., Schulze, K. J. & Savioli, L. (1998) Effects of the Zanzibar school-based deworming programme on iron status of children. Am. J. Clin. Nutr. 67:179-180.
10. Mendoza, C., Viteri, F. E., Lonnerdal, B., Raboy, V., Young, K. A. & Brown, K. H. (2001) Absorption of iron from unmodified maize and genetically altered, low phytate maize fortified with ferrous sulfate or sodium iron EDTA. Am. J. Clin. Nutr. 73:80-85.
11. Fairweather-Tait, S. J., Wharf, S. G. & Fox, T. E. (1995) Zinc absorption in infants fed iron-fortified weaning food. Am. J. Clin. Nutr. 62:785-789.
12. Davidsson, L., Walczyk, T., Zavaleta, N. & Hurrell, R. (2001) Improving iron absorption from a Peruvian school breakfast meal adding ascorbic acid or Na2EDTA. Am. J. Clin. Nutr. 131:1879-1880.
13. Zlotkin, S., Arthur, P., Antwi, Y. K. & Yeung, G. (2001) Treatment of anemia with microencapsulated ferrous fumarate plus ascorbic acid supplied as sprinkles to complementary (weaning) foods. Am. J. Clin. Nutr. 74:791-795.
14. McLaren, D. S. & Frigg, M. (1997) Sight and life manual on vitamin A deficiency disorders (VADD) 1997 Task Force Sight and Life Basel, Switzerland.
15. Thu, B. D., Schultink, W. J., Dillon, D., Gross, R., Leswara, N. D. & Khoo, C. S. (1999) Effect of daily and weekly micronutrient supplementation on micronutrient deficiencies and growth in young Vietnamese children. Am. J. Clin. Nutr. 69:80-86.
16. Schultink, W. & Gross, R. (1996) Iron deficiency alleviation in developing countries. Nutr. Res. Rev. 9:281-293.[Medline]
17. Hallberg, L. (1999) The paradox of iron deficiency. Speech by Professor Leif Hallberg. Danone International Prize for Nutrition 1999:5 Danone Paris.
18. Kilosa Hospital Annual Report. Annual report and other activities 2000:1-10.
19. Association of Official Analytical Chemists (1995) Official Methods of Analysis 1995 AOAC Washington, DC.
20. National advisory committee on microbiology criteria for foods (1998) Hazard analysis and critical control point principles and application guidelines (HACCP). J. Food Prot. 61:762-775.[Medline]
21. Kimanya, M. E., Mamiro, P.R.S., Van Camp, J., Devlieghere, F., Opsomer, A., Kolsteren, P. W. & Debevere, J. (2003) Growth of Staphylococcus aureus and Bacillus cereus during germination and drying of finger millet and kidney beans. Int. J. Food Sci. Technol. 38:119-125.
22. Dop, M. C., Benbouzid, D., Trèche, S., de benoist, B., Verster, A. & Delpeuch, F. (1999) Complementary Feeding of Young Children in Africa and the Middle East 1999 WHO Geneva, Switzerland.
23. Mbithi-Mwikya, S., Van Camp, J., Mamiro, P.R.S., Ooghe, W., Kolsteren, P. W. & Huyghebaert, A. (2002) Evaluation of the nutritional characteristics of a finger millet based complementary food. J. Agric. Food Chem. 50:3030-3036.[Medline]
24. van den Broek, N. R. & Letsky, E. A. (2000) Etiology of anemia in pregnancy in south Malawi. Am. J. Clin. Nutr. 72:247S-256S.
25. Pehrsson, P. R., Moser-Veillon, P. B., Sims, L. S., Suitor, C. W. & Russek-Cohen, E. (2001) Postpartum iron status in nonlactating participants and nonparticipants in the special supplemental nutrition program for women, infants, and children. Am. J. Clin. Nutr. 73:86-92.
26. Jackson, R. T. & Al Mousa, Z. (2000) Iron deficiency is a more important cause of anemia than hemoglobinopathies in Kuwaiti adolescent girls. J. Nutr. 130:1212-1216.
27. Asobayire, F. S., Adou, P., Davidsson, L., Cook, J. D. & Hurrell, R. F. (2001) Prevalence of iron deficiency with and without concurrent anemia in population groups with high prevalences of malaria and other infections: a study in Cote d’Ivoire. Am. J. Clin. Nutr. 74:776-782.
28. FAO (1984) Cereals and grain products. Food Composition Table for Use in Africa 1984:98-101 FAO Rome, Italy. U.S. Department of Health, Education, and Welfare and FAO.
29. Egan, H., Kirk, R. & Sawyer, R. (1981) Pearson’s Chemical Analysis of Foods 1981:20-23 Churchill Livingstone Edinburgh, UK.
30. Svanberg, U. & Sandberg, A. S. (1993) Lactic fermentation of non-tannin and high tannin cereal: effects on in vitro estimation of iron bioavailability and phytate hydrolysis. J. Food Sci. 58:408-412.
31. Haugh, W. & Lantzsch, H. (1983) Sensitive method for the determination of phytate in cereals and cereal products. J. Sci. Food Agric. 34:1423-1426.
32. Kleinbaum, D. G. (1994) Logistic regression: a self-learning text 1994 Springer-Verlag New York, NY.
33. Hosmer, D. W. & Lemeshow, S. (1989) Applied Logistic Regression 1989 John Wiley and Sons New York, NY.
34. Watzke, H. (1998) Impact of processing on bioavailability examples in foods. Trends Food Sci. Technol. 9:320-327.
35. Stevens, D. & Nelson, A. (1995) The effect of iron in formula milk after 6 months of age. Arch. Dis. Child 73:216-220.
36. Simondon, K. B., Gartner, A., Berger, J., Cornu, A., Massamba, J. P., San Miguel, J. L., Ly, C., Missotte, I., Simondon, F., Traissac, P., Delpeuch, F. & Maire, B. (1996) Effect of early, short-term supplementation on weight and linear growth of 4–7-mo-old infants in developing countries: a four-country randomized trial. Am. J. Clin. Nutr. 64:537-545.
37. Chinnamma, J. & Gopaldas, T. (1993) Evaluation of the impact on growth of a controlled 6-months feeding trial on children (6–24 months) fed on a complementary feed of a high energy low bulk gruel versus high energy high bulk gruel in addition to their habitual home diet. J. Trop. Pediatr. 39:16-22.
38. Msuya, F.H.M. & Curtis, C. F. (1991) Trial of pyrethroid bed-nets in an area of Tanzania holoendemic for malaria. Part 4. Effects of incidence of malaria infection. Acta Trop. 49:165-171.[Medline]
39. Hurrell, R. (1992) Soy protein, phytate and iron absorption in humans. Am. J. Clin. Nutr. 56:573-578.
40. Domellof, M., Cohen, R. J. & Dewey, K. G. (2001) Iron supplementation of breast-fed Honduran and Swedish infants from 4 to 9 months of age. J. Pediatr. 138:679-687.[Medline]
This article has been cited by other articles:
|
|
 |
|
  K. H. Paul, K. L. Dickin, N. S. Ali, E. C. Monterrosa, and R. J. Stoltzfus Soy- and Rice-Based Processed Complementary Food Increases Nutrient Intakes in Infants and Is Equally Acceptable with or without Added Milk Powder J. Nutr., October 1, 2008; 138(10): 1963 - 1968. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Hotz and R. S. Gibson Traditional Food-Processing and Preparation Practices to Enhance the Bioavailability of Micronutrients in Plant-Based Diets J. Nutr., April 1, 2007; 137(4): 1097 - 1100. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
