The Journal of Nutrition Community and International Nutrition

Enhancements to Nutrition Program in Indian Integrated Child Development Services Increased Growth and Energy Intake of Children1,2
Rasmi Avula,3* Edward A. Frongillo,3* Mandana Arabi,4 Sheel Sharma,5 and Werner Schultink4
3 Department of Health Promotion, Education, and Behavior, Arnold School of Public Health, University of South Carolina, Columbia, SC 29208; 4PD/Nutrition Section, UNICEF NYHQ, New York, NY 10017; and 5Department of Food Science and Nutrition, Banasthali University, P.O. Banasthali Vidyapith-304022, India

Abstract
The Indian Integrated Child Development Services (ICDS) provides supplemental food to children aged 6 mo to 6 y. This study assessed the impact of enhancements to the existing Supplemental Nutrition Program of local production of supplemental food, home fortication with a micronutrient powder, and improved program monitoring. A quasiexperimental longitudinal design was used. Data were collected from 15 Anganwadi centers randomly selected from the enhanced program and 15 from the usual program. Multilevel linear regression was used to examine changes over time between the 2 groups accounting for village-level variation in intent-to-treat analysis. Children in the enhanced program initially aged 1218 mo gained 0.72 (P = 0.02) greater height-for-age Z-score. Signicant differences were observed in gain in weight-for-age Z-score among those initially aged 911 (2.48; P = 0.01), 1218 (0.76; P = 0.01), and 1924 mo old (0.73; P = 0.01), and gain in weight-for-height Z-score among 911 (2.66; P = 0.04) and 1924 mo old (0.99; P = 0.01). For these age groups, the prevalence of stunting, underweight, or wasting averted ranged from 20.3 to 33.4%. Energy intake in the enhanced program was signicantly greater for boys initially aged 1218 mo (575.1 kJ/d; P = 0.03). Results from a qualitative substudy supported the plausibility of observed outcomes. ICDS would be more effective in improving child nutrition if it included these enhancements. The enhancements studied may be useful in improving program delivery and uptake of other similar programs. J. Nutr. 141: 680684, 2011.

Downloaded from jn.nutrition.org by guest on January 17, 2014

Introduction
Nearly 60 million children (43%) in India are underweight and 48% are stunted. From 1992 to 2005, the prevalence of underweight decreased from 53 to 40% and the prevalence of stunting decreased from 52 to 45% (1), progress insufcient to meet Millennium Development Goals. While the prevalence of exclusive breast-feeding for 6 mo improved from 37% in 2000 to 46% in 2005, delayed initiation of breast-feeding, giving water or other liquids to breast-fed children under 6 mo, and late introduction and low quality of complementary feeding continue to be problems (1,2). High anemia prevalence (74.3%) among preschool children also warrants integrated approaches to improve nutrition and reduce infectious diseases (3). In the 1970s the Indian government initiated the Integrated Child Development Services (ICDS)6 program to provide several services through a network of Anganwadi centers (AWC) delivered by Anganwadi workers (AWW): supplementary nutrition, immunization, health check-up, referral services, preschool nonformal education, and nutrition and health education. There is 1 AWC per village or for a population of ;1000 (4). A program focus has been the Supplemental Nutrition Program (SNP). The SNP provides supplementary food to children between 6 mo and 6 y of age and to pregnant and lactating women. A hot meal is served every day at the AWC to children 3 y and older. A take-home ration is given to pregnant and lactating women and children 636 mo old (4). The SNP performs poorly due to irregularities in the food supply, leakage to nontargeted individuals, lack of awareness among mothers about the food or eligibility of their children, and failure of the

Supported by the United Nations Childrens Fund. The authors acknowledge the support of the Global Alliance for Improved Nutrition, Micronutrient Initiative, UNICEF, and the Department of Food Science and Nutrition, Banasthali University, Rajasthan, India. 2 Author disclosures: R. Avula, E. A. Frongillo, M. Arabi, S. Sharma, and W. Schultink, no conicts of interest. * To whom correspondence should be addressed. E-mail: efrongillo@sc.edu.

Abbreviations used: AWC, Anganwadi center; AWW, Anganwadi worker; CDPO, Child Development Project Ofcer; HAZ, height-for-age Z-score; ICDS, Integrated Child Development Services; SNP, Supplemental Nutrition Program; WAZ, weight-for-age Z-score; WHZ, weight-for-height Z-score.

680

2011 American Society for Nutrition. Manuscript received October 2, 2009. Initial review completed November 24, 2009. Revision accepted January 17, 2011. First published online February 23, 2011; doi:10.3945/jn.109.116954.

AWW to contact mothers when food is available (4). Recent efforts have improved the quality of the food supplement by changing requirements for the nutrient content for various age groups. Other operational enhancements may be needed to increase program effectiveness. This study assessed if 3 enhancements to the SNP would benet growth and dietary outcomes of children 630 mo old after a period of 6 mo. The enhancements were local preparation of supplemental food, inclusion of a micronutrient supplement, and increased monitoring of program delivery.

children were sampled from the AWC; the difference in the number of children in the enhanced and usual program groups reects the number of children in the AWC. After conclusion of the quantitative data collection, qualitative data were collected. Anthropometric measures. Children who were 630 mo old at the beginning of the study were measured for height and weight at the initial and nal time points. Recumbent length was measured for children who were ,24 mo. Initially, there were 680 children in the enhanced group and 448 children in the usual program group. There were fewer children in the latter group because of physical access and transportation challenges for the data collection team in that location. At the nal time point, data were available for 640 children in the enhanced program group and for 399 children in the usual program group. Some caregivers left their villages and some refused to allow their children to be measured, resulting in attrition. The attrition rate was similar in the enhanced-program group (8%) compared with the usual program group (11%). In each group, weight and height (adjusted for age and sex) did not signicantly differ between children available for both initial and nal time points compared with those available for the initial time point only. The children were measured at the AWC by 2 trained data collectors from the evaluation team that was independent of the program, which minimized the potential for measurement bias even though the data collectors could not be unaware of the treatment allocation. When the children were not available at the center, measurements were taken at their homes. Height-for-age (HAZ), weight-for-age (WAZ), and weight-forheight (WHZ) Z-scores were computed based on the 2006 WHO Child Growth Standards (9). Change variables were computed for all the Z-scores to examine differences from initial to nal time points. The distributions of and changes in Z-scores were carefully examined to ensure that the analytic data set did not contain anomalies and outlying values. Dietary intake measures. Dietary data were collected from mothers of children 618 mo old using 24-h dietary recall with the help of trained interviewers who prompted responses. Dietary data were available for 371 and 229 children in the enhanced and usual program groups, respectively, at the initial time point and for 332 (89.4%) and 207 (90.4%), respectively, at the nal time point. The nutrient compositions of the food preparations were calculated based on the Nutritive Value of Indian Foods (10). Change variables from initial to nal time points were computed for energy, protein, and iron intakes. Because the quantity of breast milk intake by children could not be measured, contribution of breast milk to these nutrients was not included. Because boys are typically favored over girls in South Asian countries (11), we examined gender differences in growth and dietary intake. Qualitative data collection. The qualitative study was conducted in the enhanced program arm to understand possible mechanisms through which program enhancements might have affected children. Interviews were conducted by the rst author with AWW to understand their knowledge of how Anuka should be fed to the child, perceptions about distribution of supplementary food, and caregiver acceptance of these products. Interviews with caregivers were conducted to understand their perspectives on program enhancements regarding acceptability of supplementary food and usage of Anuka and to triangulate with information obtained from interviews with the AWW. Five AWC from the enhanced program were visited. Five AWW, 8 caregivers, and the ICDS Child Development Project Ofcer (CDPO) were interviewed using a semistructured interview format. Caregivers identied as enhanced program participants were randomly selected from an existing list at the AWC. They were approached in their households for interviews. Caregiver interviews were conducted until data saturation was attained; it was achieved after 8 interviews. Information obtained through these interviews complemented the information gathered from the AWW. Data were compiled to identify underlying themes, and overall ndings were summarized using standard methods (12). Analysis. Initial group differences were tested adjusting for age and sex. Multi-level linear regression tested differences in changes over time between the 2 groups taking into account village-level variation in Enhanced nutrition program 681

Methods
Study setting. The setting was Rajasthan, 1 of 6 Indian states in which at least 1 in 2 children is underweight (4). The state is divided into 33 districts. Communities selected were in 2 blocks within 1 district ;82 km from the states capital city. The study was approved by ICDS and verbal informed consent was obtained from participating mothers and health workers. Program enhancements. There were 3 enhancements to the SNP program. First, ready-to-eat supplementary food (made of roasted and nely blended 40 g of wheat, 25 g of soybean, 30 g of sugar, and 5 g of edible oil) was prepared within the village instead of procuring it centrally. The suggested amount of supplementary food per meal was 75 g, which would provide 1285 kJ. Second, Anuka, a multiple micronutrient supplement [per sachet: 12 mg iron, 50 mg folic acid, 300 mg vitamin A (as retinol acetate), 40 mg vitamin C, and 5 mg zinc, topped up to 500 mg with dextrose to make it more palatable] was provided in addition to the supplementary food. Third, monitoring of the AWC was increased. The enhancements were based on theoretical consideration that motivating program implementers (i.e. AWW) and program recipients (i.e. caregivers) would induce effective program implementation and utilization, respectively, resulting in improved child outcomes. A locally produced supplementary food might result in better quality and create ownership for the product among caregivers, thus motivating them to use it. This model was facilitated by the ruling of the Supreme Court of India in 2004 for using decentralized food for the AWC instead of a centralized supply to alleviate the problems associated with quality and acceptability of the food (5). The second enhancement of providing Anuka to be fed to a child mixed in food might act as a motivator for caregivers to introduce complementary food. It might induce AWW to deliver the program better and caregivers to be receptive to the messages. Moreover, including a multiple micronutrient supplement has improved micronutrient status among children elsewhere (68). The 3rd enhancement of improved monitoring was intended to facilitate improved program delivery, given that little emphasis has been placed in ICDS on assessing quality of service delivery, and strengthening this activity has been recommended (4). Implementation of enhanced program. During a designated day of the week, the supplementary food and 5 sachets of Anuka were distributed to caregivers. The AWW demonstrated feeding Anuka to children by mixing it in the supplementary food and encouraged caregivers to feed 1 sachet while at the center. Caregivers were instructed to empty the entire sachet into the amount of food that the child would consume at one time. There were regular visits by female supervisors to monitor the AWWs work. For 6 mo, 1 block received the enhanced program while the other received the usual SNP program, i.e. the centrally procured ready-to-eat supplementary food with neither Anuka nor improved monitoring. Design and sample. The study used a quasi-experimental longitudinal design. Of the 186 AWC in the enhanced program, 15 were randomly selected and another 15 AWC that received the usual SNP program were randomly selected for study; program staff did not inuence the selection of the AWC for the study. Children were aged 630 mo at the beginning of the study and 1236 mo at the nal time point. Quantitative data were collected both prior to and after the end of program enhancement. The

Downloaded from jn.nutrition.org by guest on January 17, 2014

intent-to-treat analysis using STATA xtreg (STATA version 8.1). If village-level variation was estimated as zero, models were rerun using the reg procedure. Because the same children were measured at both time points, each child acted as his or her own control. Differences in changes in HAZ, WAZ, and WHZ scores were assessed between enhanced and usual program groups; differences in dietary intake were assessed as changes in energy, protein, and iron intakes. Analysis was conducted by grouping the 6- to 30-mo-old children into 5 groups corresponding to those with different infant-feeding recommendations. Growth outcomes were assessed after adjusting for age at both time points and for initial Z-score and were reported as unstandardized regression coefcients of corresponding Z-scores. For each age group, the benet attributable to the program (i.e. the prevalence of stunting, underweight, or wasting averted) was estimated using the observed prevalence of usual program group, the estimated program effect from the regression model, and the usual epidemiologic formula for attributable risk. For assessing change in energy intake, age and breastfeeding status at both time points and energy intake at the initial time point were adjusted. Additionally, energy intake was adjusted when assessing changes in protein and iron intakes. We tested for interaction effects between program groups and gender for both growth and dietary intake. For all tests, P , 0.05 was considered signicant.

Results
The initial weight of children in the enhanced program group was signicantly lower than those in the usual program group by 0.30.7 kg for the 68, 911, 1218, and 1924 mo age groups (Table 1). At the initial time point, there was no signicant difference in the mean nutrient intake between the 2 groups (Table 2). For change in HAZ, the difference between enhanced and usual program groups was not signicant among age groups (P = 0.20.7) except those initially aged 1218 mo (0.72; P = 0.02) (Table 3). The benet attributable to the program for stunting for this group was 20.3%. For change in WAZ, the difference between the groups was signicant for those initially aged 911 (2.48; P = 0.01), 1218 (0.76; P = 0.01), and 1924 mo (0.73; P = 0.01) (Table 3). For these 3 age groups, the benet attributable to the program for underweight was 33.4, 24.4, and 24.5%, respectively. For change in WHZ, differences were observed between enhanced and usual groups for children initially aged 9 11 (2.66; P = 0.04) and 1924 mo (0.99; P = 0.01) (Table 3). For these 2 age groups, the benet attributable to the program for wasting was 26.5 and 26.4%, respectively. Two of the 18 differences for age groups and anthropometric measures examined in Table 3 had signicant interactions. For HAZ among 26- to 30-mo-old children, the enhanced program effect for boys was 0.42, whereas for girls it was 20.01. Among
TABLE 1 Initial anthropometric characteristics of children in enhanced and usual program groups1
Height Age mo 68 911 1218 1924 2530
1

Weight Usual group Enhanced group kg (93) (96) (189) (177) (125) Usual group

Enhanced group cm 65.7 69.1 72.4 78.6 81.7 6 3.3 6 4.5 6 3.7 6 4.7 6 4.1 (93) (96) (189) (177) (125) 66.9 70.2 72.7 78.9 81.7

6 3.6 6 3.9 6 4.1 6 5.1 6 4.1

(46) (41) (148) (96) (117)

6.5 7.2 7.8 8.8 9.7

6 1.0 6 1.1 6 1.0 6 1.3 6 1.2

7.2 7.7 8.1 9.3 10.0

6 1.0a (46) 6 1.2b (41) 6 1.1a (148) 6 1.2b (96) 6 1.2 (117)

Values are means 6 SD (n). Letters indicate different from the enhanced group: aP , 0.01, bP , 0.05.

12- to 18-mo-old children, for WAZ the enhanced program effect for boys was 0.97, whereas for girls it was 0.53. Analysis of energy intake included regular food and supplementary food but neither breast milk nor Anuka. For change in energy intake, there was a signicant interaction between gender and program group among children initially aged 1218 mo (Table 4). Boys in the enhanced program group had 575 kJ/d higher energy intake compared with those in the usual program. The enhanced program effect on girls, however, was smaller (285 kJ/d) (P = 0.07). Assuming no interaction, change in energy intake in the enhanced program was (not signicantly) 380, 74, and 408 kJ higher for children initially aged 68, 911, and 12 18 mo, respectively. For protein and iron intakes, as expected, there were no signicant interaction effects. There were no signicant differences for changes in protein and iron intakes after adjusting for energy intake. From the qualitative data, the enhanced program implementation was facilitated by decentralization of supplementary food, training of AWW, and improved monitoring. All the key informants stated that there had been an improvement in the quality of supplementary food resulting in increased acceptance of the product. The centrally procured supplementary food was irregularly supplied, was of poor quality and palatability, and had low acceptance. The product would be at least be 1 mo old when delivered to the AWC. Furthermore, its texture was coarse and had lower palatability, because it was prepared from raw ingredients. Two AWW, the CDPO, and 1 mother indicated that it would be consumed by family members other than the target child or fed to the cattle. Two mothers indicated that it used to be stale and smelled bad. In contrast, when supplementary food was prepared within the village, it was made weekly using roasted ingredients, thus improving the freshness and palatability of the product. Moreover, there was a sense of ownership and recognition of the product by the caregivers, resulting in feeding it to the child. All ICDS personnel in the enhanced program received training on feeding Anuka to children and counseling mothers on complementary feeding. The CDPO and all AWW interviewed described how they communicated how to use Anuka to mothers. Two AWW said that there were some complaints from mothers regarding instances of diarrhea; when Anuka was given to their children, they started losing weight. The AWW could accurately counsel them that diarrhea was due to inadequate hygiene practices and not due to Anuka. In the enhanced program, the AWC were regularly monitored through several visits by ICDS and UNICEF personnel and meticulous documentation. One key informant said that the success of the enhanced program was mainly due to local preparation of supplementary food and increased monitoring. Visits by ofcials to the AWC positively affected AWW performance. All the AWW indicated that they had worked hard during this period and mobilized mothers to bring their children and take supplementary food and Anuka. One AWW said that after the end of the enhanced program she has discontinued her rigorous campaign to sensitize mothers to feed their children supplementary food. During the study period, regular monitoring was feasible, because all the staff positions were occupied. The CDPO in charge of operations of the 186 AWC had 1 assistant CDPO and 6 female supervisors to monitor the AWC. At the time of the interview, however, there were only 3 supervisors available and the assistant CDPO post was vacant, resulting in decreased monitoring; the CDPO was not completely aware of the situations in the AWC.

Downloaded from jn.nutrition.org by guest on January 17, 2014

682

Avula et al.

TABLE 2

Initial nutrient intake among children in enhanced and usual program groups1
Energy Protein2 Usual group Enhanced group g/d 2.67 6 2.96 (90) 5.35 6 4.37 (96) 7.34 6 4.40 (185) 3.07 6 2.74 (46) 4.73 6 3.12 (41) 6.68 6 4.30 (142) Usual group Iron2 Enhanced group Usual group

Age mo 68 911 1218

1 2

Enhanced group

kJ/d 379 6 447 (90) 435 6 379 (46) 681 6 557 (96) 694 6 494 (41) 1023 6 597 (185) 976 6 593 (142)

mg/d 0.49 6 0.65 (90) 0.73 6 0.82 (46) 1.23 6 1.03 (96) 1.40 6 1.03 (41) 1.90 6 1.43 (183) 1.74 6 1.10 (142)

Values are means 6 SD (n). Difference between program groups assessed at P , 0.05. Differences in protein and iron intakes adjusted for energy intake.

Discussion
In rural northwestern India, through multiple enhancements to the existing SNP program, signicant impacts on growth and nutrient intake of children were achieved. The enhanced program signicantly increased growth in WAZ and HAZ for children who were initially aged 1218 mo and in WAZ and WHZ for children initially aged 911 and 1924 mo. The biggest effect for WAZ was 2.48 among children initially aged 911 mo, which corresponds to a difference in growth of 3 kg. The benet attributable to the program for stunting, underweight, and wasting in these age groups ranged from 20.3 to 33.4%, benets that correspond to averting between 63 and 99% of the prevalence. Similar results were observed in an evaluation study of a National Food Nutrition Program in Ecuador, wherein a fortied complementary food delivered through public health services had benecial effects on linear growth and weight among children who were older (1214 mo) at the beginning of the program (13). These results are in contrast to those observed in another large-scale government program in Mexico that benetted children younger than 6 mo at baseline who belonged to the poorest households (14). In Peru, no difference was observed in growth among children who received fortied complementary food compared with those who did not, even though there was improvement in nutrient intake (15). One possible reason we observed benets for older children is that older children were more likely to have been introduced to other foods at the beginning of the study and hence were offered complementary food more than the younger children (13). This is consistent with national data from India. Recent Demographic and Health Survey data (20052006) show that only ;53% of children ages 68 mo are given timely complementary feeding. Only 2 interactions in growth measures were signicant; both favored boys. We observed signicant modication by gender for program differences in energy intake from dietary and supplementary food among 12- to 18-mo-old children, with a

greater effect for boys. The effect of enhanced program on WAZ within this age group also was greater for boys, supporting the plausibility of the growth differences. These results are consistent with previous studies and are in the expected direction. Bhandari et al. (16) observed increased energy intake among children aged 412 mo who received milk-based cereal and nutrition counseling. Gender differences were observed in food allocations in households in Latin America (17) and South Asia (18). Our results are consistent with long-standing and prevailing preferential treatment for boys in South Asian countries (11), including neglect of girls during infancy and early childhood (19). There were no signicant differences observed in protein and iron intakes from diet between the 2 groups. This should not, however, be interpreted as no effect on iron status or anemia in the enhanced program group, because we did not measure these outcomes. Home fortication with micronutrient supplements is known to be effective in reducing anemia among infants (68). During the past 30 y of the ICDS program, universalization has been emphasized rather than strengthening the quality of implementation (4). Inadequate worker skills, lack of preservice training with little in-service training, poor supervision, and weak monitoring and evaluation have been identied as affecting the performance of the program (4). Our study shows that training AWW and monitoring their performance facilitated program implementation. Furthermore, decentralization of the preparation of supplementary food was benecial. Although the composition of both local and centrally procured supplementary food was similar, the quality and acceptability of locally made supplementary food were higher. Similar results were observed in another study that showed local preparation of food for the AWC by self-help groups improved acceptability, quality, and availability of the food compared with a vendor-based system (5). Adding a multiple micronutrient supplementation component to a more comprehensive infant and young child feeding program may also have incentivized caregiver participation and facilitated delivery of program messages.

Downloaded from jn.nutrition.org by guest on January 17, 2014

TABLE 3

Change in Z-scores between enhanced and usual program groups from initial to nal time points controlling for gender, initial and nal age, and the corresponding initial Z-score1
n 135 128 317 237 221 HAZ 0.71 0.38 0.72a 0.1 0.29 WAZ 1.1 2.48a 0.76a 0.73a 0.11 WHZ 1.09 2.66a 0.6 0.99a 20.22

TABLE 4

Interaction effect between gender and enhanced compared with usual program groups for change in energy intake based on 24-h dietary recall1
Boys 294 (75) 68 (64) 575 (179)* Girls 408 (51) 111 (59) 285 (112)

Age, mo 68 911 1218 1924 2530

1

Age, mo 68 911 1218

Unstandardized coefcients in Z-score units. aP , 0.05.

1 Values are coefcients in kJ/d (n) for the difference between the enhanced and usual program groups in the change from the initial to the nal time point in energy intake, *P , 0.05.

Enhanced nutrition program

683

This study took place in a community utilizing the services of existing staff and the infrastructure of a national-level program. Multiple factors could have been at play that facilitated or hindered intended program implementation. We gained an understanding of some factors through the qualitative data and assessed the effect of the program on growth and nutrient intake of children through the quantitative measures. Although we cannot separate effects of individual program enhancements on growth and nutrient intake, together the 3 enhancements were effective. Interpreting these results through a programmatic lens may be useful, because in community settings multiple parameters are simultaneously at play when programs are implemented. It is important to examine how such factors inuence each other and the intended outcomes together rather than how individual components affect the outcomes. These results have important implications for ICDS in identifying and improving the aspects of the program to maximize impact on child outcomes. Given the propensity to favor male children, it is important to mitigate gender bias in food allocation and caregiving through complementary feeding counseling and to create an environment where boys and girls receive adequate and equal attention to achieve potential growth and development. Acknowledgments R.A., E.A.F., W.S., and M.A. designed the research; S.S. provided essential materials for the research; R.A. and E.A.F. analyzed the data; R.A., E.A.F., M.A., and W.S. wrote the paper; and R.A., E.A.F., and M.A. had primary responsibility for nal content. All authors read and approved the nal manuscript.

5.

6.

7.

8.

9.

10.

11. 12. 13.

14.

Literature Cited
1. International Institute for Population Sciences (IIPS) and Macro International. 2007. National Family Health Survey (NFHS-3), 2005 06, India: key ndings [cited 16 Feb 2011]. Available from: http://www. measuredhs.com/pubs/pdf/SR128/SR128.pdf. WHO. UNICEF and partners. Indicators for assessing infant and young child feeding practices, part 3: country proles. Geneva: WHO; 2010. de Benoist B, McLean E, Egli I, Cogswell M, editors. Worldwide prevalence of anaemia 19932005: WHO global database on anaemia. Geneva: WHO Press; 2008. Gragnolati M, Bredenkamp C, Gupta MD, Lee Y, Shekar M. ICDS and persistent undernutrition: strategies to enhance the impact. Econ Polit Wkly. 2006:11931201.

15. 16.

2. 3.

17. 18. 19.

4.

Huff-Rousselle M, Purushothaman S, Tirupathaiah N, Fiedler JL. Assessment of decentralized food models in Indias integrated child development services program. Washington, DC: Food and Nutrition Technical Assistance (FANTA) Project, Academy of Educational Development; 2007 Oct [cited 16 Feb 2011]. Available from: http://www. fantaproject.org/downloads/pdfs/Food_Models_Oct07.pdf. Zlotkin S, Arthur P, Schauer C, Antwi KY, Yeung G, Piekarz A. Homefortication with iron and zinc sprinkles or iron sprinkles alone successfully treats anemia in infants and young children. J Nutr. 2003;133:107580. Smuts CM, Dhansay MA, Faber M, van Stuijvenberg ME, Gross R, Benade AJS. Efcacy of multiple micronutrient supplementation for improving anemia, micronutrient status, and growth in South African infants. J Nutr. 2005;135:S6539. Menon P, Ruel MT, Loechl CU, Arimond M, Habicht JP, Pelto G, Michaud L. Micronutrient sprinkles reduce anemia among 9- to 24-moold children when delivered through an integrated health and nutrition program in rural Haiti. J Nutr. 2007;137:102330. WHO. WHO multicenter growth reference study group. WHO child growth standards: length/height-for-age, weight-for-age, weight-forlength, weight-for-height and body mass index-for-age: methods and development. Geneva: WHO; 2006. Gopalan C, Rama Sastri BV, Balasubramanian SC, Narasinga Rao BS, Deosthale YS. Nutritive value of Indian foods. Hyderabad: National Institute of Nutrition; 2007. Watts C, Zimmerman C. Violence against women: global scope and magnitude. Lancet. 2002;359:12327. Miles MB, Huberman AM. Qualitative data analysis: a sourcebook of new methods. 2nd ed. Thousand Oaks (CA): Sage; 1994. Lutter CK, Rodriguez A, Fuenmayor G, Avila L, Sempertegui F, Escobar J. Growth and micronutrient status in children receiving a fortied complementary food. J Nutr. 2008;138:37988. Rivera JA, Sotres-Alvarez D, Habicht JP, Shamah T, Villalpando S. Impact of the Mexican program for education, health, and nutrition (Progresa) on rates of growth and anemia in infants and young children: a randomized effectiveness study. JAMA. 2004;291:256370. Lo pez de Roman a G. Experience with complementary feeding in the FONCODES project. Food Nutr Bull. 2000;21:438. Bhandari N, Bahl R, Nayyar B, Khokhar P, Rohde JE, Bhan MK. Food supplementation with encouragement to feed it to infants from 4 to 12 months of age has a small impact on weight gain. J Nutr. 2001;131: 194651. Frongillo EA, Begin F. Gender bias in food intake favors male preschool Guatemalan children. J Nutr. 1993;123:18996. Miller BD. Social class, gender and intrahousehold food allocations to children in South Asia. Soc Sci Med. 1997;44:168595. Gupta MD, Zhenghua J, Bohua L, Zhenming X, Chung W, Hwa-Ok B. Why is son preference so persistent in east and south Asia? A crosscountry study of China, India, and the Republic of Korea. J Dev Stud. 2003;40:15387.

Downloaded from jn.nutrition.org by guest on January 17, 2014

684

Avula et al.