Acceptability of multiple micronutrient powders and iron syrup in Bihar, India.

Abstract Nearly two thirds of young children are anaemic in Bihar, India. Paediatric iron and folic acid syrup (IFAS) and multiple micronutrient powders (MNPs) are two evidence‐based interventions to prevent anaemia. Using a randomized crossover design, we examined the acceptability of IFAS versus MNPs for children 6–23 months. In a catchment area of 2 health centres in Bihar, health front‐line workers (FLWs) delivered either (a) IFAS twice weekly or (b) MNPs for 1 month followed by the other supplementation strategy for 1 month to the same families (NCT02610881). Household surveys were conducted at baseline (N = 100), 1 month after receiving the first intervention (1 month; N = 95), and 1 month after the second intervention (2 months; N = 93). Focus group discussions (10 FLWs) and in‐depth interviews (20 mothers) were held at 1 and 2 months. We used chi‐square and Fisher exact tests to test mothers' product preferences. Qualitative data were analysed using MaxQDA and Excel employing a thematic analysis approach. There was high adherence and acceptability for both products (>80%). There was no significant difference in preference (p < .05) on perceived benefits (39% MNPs, 40% IFAS), side effects (30% MNPs, 30% IFAS), ease of use (42% IFAS, 31% MNPs), child preference (45% IFAS, 37% MNPs), and maternal preference (44% IFAS, 34% MNPs). Mothers and FLWs indicated that the direct administration of IFAS ensured that children consumed the full dose, and MNPs intake depended on the quantity of food consumed, especially among younger children, which emphasizes the need to integrate supplementation with the promotion of optimal child feeding practices.

42% of anaemia among young children in most low-middle-income countries is due to iron deficiency (Stevens et al., 2013).
Paediatric iron and folic acid syrup (IFAS) and multiple micronutrient powders (MNPs) are two evidence-based strategies to prevent anaemia (Ministry of Health and Family Welfare, 2013;World Health Organization [WHO], 2016). The Government of India currently recommends IFAS to be directly administered via a 1-ml dropper to all children (6-59 months) twice per week by front-line workers called the Accredited Social Health Activists (ASHAs; Ministry of Health and Family Welfare, 2013). Important limitations of traditional oral iron supplementation include reported unpalatable metallic taste, teeth staining, potential for overdose, limited shelf life, poor compliance, perception of the syrup as medicine, lack of supplies, and procurement-associated challenges (Mora, 2002;Nestel & Alnwick, 1997). Although a recent review has shown that direct administration of IFAS is an effective strategy for prevention and management of anaemia and improving compliance (Bairwa et al., 2017), additional information is needed to understand scalability in terms of the limitations of this intervention and burden on the health care system.
Home fortification with MNPs represents an innovative solution that addresses some of the challenges associated with the provision of paediatric iron syrup wherein the caregiver adds a daily dose of key micronutrients, including iron, directly to a child's meal within the home. MNPs are encapsulated with a thin vegetable-based hydrogenated lipid coating that prevents changes in appearance, texture, or taste of the meal, making it more palatable to children (Zlotkin et al., 2004). MNPs have similar efficacy as IFAS with the added benefit of easy administration and high adherence (Zlotkin, Antwi, Schauer, & Yeung, 2003;Zlotkin, Arthur, Antwi, & Yeung, 2001). Administration of MNP does not require the supervision of a front-line worker and can be provided in boxes with individually packaged sachets for daily household administration over several weeks. Currently, there is a strong global evidence base for MNPs and their impact on reducing anaemia De-Regil, Suchdev, Vist, Walleser, & Pena-Rosas, 2011;WHO, 2016). According to a Cochrane review, home fortification with micronutrient powders reduced anaemia in young children by 31% and iron deficiency by 51% (De-Regil et al., 2011). Given the strength of the evidence for the efficacy of this product, the WHO (2011WHO ( , 2016 released strong recommendations in support of home fortification of foods with MNPs to improve iron status and reduce anaemia among young children. However, there are also challenges associated with the use of MNPs including the need for significant interpersonal behaviour change communication and maintenance of the supply chain (Strengthening Partnerships, Results, and Innovations in Nutrition Globally, 2015).
Given the high prevalence of anaemia among Bihar's children, there is a critical need to identify and evaluate multiple delivery strategies to help meet the daily nutrient requirements for child health and development. Although there is global evidence of the impact for both iron syrup and MNP in improving anaemia in young children, evidence is lacking regarding which strategy is most acceptable and likely to achieve coverage in India. To ensure the success of India's anaemia prevention and reduction programmes, it is critical that we identify the most acceptable approach both for families and for front-line workers charged with delivering the supplement. half received MNPs followed by IFAS ( Figure S1). With this crossover design, each child had the opportunity to try both products and served as his or her own control. Analysis was conducted both stratified by HC and pooled. The trial was registered in ClincalTrials. gov (NCT02610881).

| Intervention
ASHAs from our selected HCs were trained on the study protocol, product use, and infant and young child feeding (IYCF) practices before each round of product delivery. Each ASHA delivered either IFAS or MNP to 10 children within their catchment areas over the course of 1 month. Following a 2-week washout period, ASHAs delivered the alternative product to the same children. While delivering the products, ASHAs counselled mothers on how to use the product and on IYCF practices and gave a pamphlet with information on IYCF (ageappropriate complementary feeding messages on meal quantity, frequency, consistency, dietary diversity, and hand washing).
The ASHAs were asked to administer directly the IFAS using a 1ml dropper to each child in their catchment on Wednesday and Saturday over the span of 1 month for a total of eight visits from the ASHA (as per government policy). The iron syrup contained 20 mg elemental iron and 100 mcg folic acid per millilitre with caramel flavouring in a 100-ml bottle (a separate bottle was used for each child).
In case of MNPs, ASHAs delivered 30 sachets (dose for 1 month) during their first visit and conducted a second follow-up visit at 2 weeks to track product use and answer any questions pertaining to the product. Similarly, for IFAS, ASHAs counselled mothers on the product and gave an information pamphlet on IYCF and product use.
The MNP product was administered by the caregiver and did not require the presence of the ASHA. The MNP contained 12.5 mg iron (ferrous fumarate), 5 mg zinc, 0.16 mg folic acid, 0.3 mg vitamin A, 30 mg vitamin C, 0.9 mcg vitamin B-12, and 90 mcg iodine. The composition and packaging (local name Jeevan Jyoti, "light of life") were produced specially for this context through earlier formative research (Young et al., 2014).

| Data collection
We conducted household surveys at baseline (N = 100), after 1 month of use of Product 1 (1 month; N = 95) and after 1 month of use of Product 2 (2 months; N = 93). Surveys collected data on household demographics; child feeding practices; and product receipt, use, and acceptability. Acceptability was determined using a Likert scale (agree, no opinion, and disagree). Seven children were lost to follow-up due to travelling for festivals or marriages. Focus group discussions were conducted with ASHAs (N = 5) at baseline, after 1 month, and after 2 months to understand their perceptions of the two products, ease of delivering them, workload, and incentives. In addition, we conducted semistructured interviews with 10 mothers per HC at the end of Month 1 and Month 2 to understand their perceptions of the two products, reasons for preference of one product over the other, and overall preference for the supplements.

| Quality control
Extensive training was provided to enumerators prior to each data collection period. A team of nine enumerators collected quantitative data.
Three field supervisors accompanied the data collection team to ensure quality and perform spot checks and back-checks. Issues were immediately reported to the study coordinator. All surveys were edited prior to data entry, and a 10% back-check was done on all surveys to ensure no errors were made during the entry process.
Qualitative data collection was conducted by three trained and experienced facilitators at baseline and four facilitators after 1 and 2 months. All qualitative data were translated and transcribed verbatim, and quality checks were conducted to ensure accuracy.

| Statistical analyses
We summarize the descriptive data from surveys on household characteristics and experiences with the iron products using means/medians and frequencies and examine differences in caregiver product preferences using chi-square and Fisher exact tests. All results were also stratified by HC and child age groups (6-15 months and 16-24 months). A p-value less than .05 was considered significant.
Polytomous logistic regression was conducted to model predictors of product preference. All quantitative data analyses were performed using SAS version 9.3 (SAS Institute, Cary, NC, USA). Qualitative data from semistructured interviews and focus group discussions were transcribed and translated. Interviews were analysed using MaxQDA and Excel employing a thematic analysis approach. Deductive themes defined a priori were coded and described, and inductive themes emerging from the data were also noted and coded.

| RESULTS
A majority of households were Hindu (83%) and from deprived social strata belonging to ordinary backward caste (67%; Table 1). Product MNP and IFAS product acceptability was assessed after each round of exposure to the products at the end of month of product trial period per product (Table 3). The majority of mothers reported that their children consumed the iron product in the past month and expressed interest in continuing to use the product in the future (91-100%). Mothers perceived that both products provided positive benefits to their children, which included increased energy (82-88%), increased food consumption (78-88%), increased mental activity (82-88%), and improved child health (74-90%). Over 90% of mothers acknowledged the importance of giving IFAS and MNP to their child.
One mother who used MNP said, "… so many benefits have occurred …. She became more fat," and another mother who used IFAS said, "Cold and fever got over after taking medicine." MNP and IFAS product preference was assessed after mothers had the opportunity to try both products ( Mothers expressed nearly equal preference to continued using of IFAS (38%) and MNPs (38%), with only one mother not wanting either product and 24% mothers preferring both products. Product preference did not significantly vary by child age group.
The role of child feeding practices and necessity to provide MNP with a child's meal had both a positive and negative influence on maternal product preference. A mother who used MNP in the second round said, "Powder is preferable because the child who may not be eating rice will also start eating …. I prefer feeding baby myself." Another mother who used MNP in the first round said, "If powder is there, sometimes they will eat or sometimes they will throw/scatter …. It is to be fed with rice ... but if it is from the bottle [referring to IFA], it can be directly fed, so child will neither throw nor litter." Mothers in both HCs noted that MNP was better suited for older children who had "become smart" and were able to eat on their own.
The ASHAs expressed a similar opinion: "When we give powder for children aged 2-3 years it is easy and they will eat happily, but not babies. Babies and smaller children do not eat food properly and there are lots of problems. Mothers complain that even though they mix the powder with food, sometimes their child does not eat all the food and leaves something behind." Both ASHAs and mothers stated that using the iron syrup was easier and allowed assurance that the child was receiving the full dose of the supplement, as compared with MNP where the amount consumed depended on the quantity of food the child ate. As illustrated by a mother who commented that she liked the syrup as her child, "was not wasting … use to drink (the iron syrup) easily." Some mothers and ASHAs perceived that fidelity to iron syrup was higher because the ASHAs ensured the child was consuming the product twice  weekly; administration by the ASHAs was believed to reduce the burden on mothers as well. One ASHA commented that both products were equally good, "For us (ASHAs) both (IFAS and MNPs) are fine … now it depends on the caregiver because we do not know whether she is feeding or not, but iron syrup we are feeding on our own … so we prefer it." ASHAs were particularly happy with the iron syrup product citing that children were responding very well to it and that it was commonly referred to as a "medicine." ASHAs reported, "Mothers say this [IFAS] is very beneficial medicine, can you bring two more and give me?
Everyone is praising this medicine." ASHAs reported that the distribution of the iron-containing products enhanced their relationship with the community but had challenges with distributing them only for a few selected children.
ASHAs reported that both other children and mothers themselves requested the iron products: "they [mothers] see that we are feeding other children and not feeding their child … so we have become bad in their eyes." Although ASHAs were willing to distribute either product, they expressed an overall preference for the iron syrup in the qualitative interviews that related to the ease of use: "Syrup can be given easily in one time only, so they will also feel that child is having something, while the powder has to be given several times and they forget to give it, but syrup can be given in one time." When asked about scale up, an ASHA commented that the delivery of the supplement may depend on households: "if they  Health Centre 1 was randomized to first IFAS then MNPs, n = 50; Health Centre 2 was randomized to first MNP then IFAS, n = 50.
b Preference data obtained at 2 months from 93 households after households had tried both products; overall, there were no statistical differences in product preference (p > .05).
YOUNG ET AL.
understand we will give them one dose and then hand it over to the beneficiary"; in other homes, more direct support might be needed.
ASHAs in one HC attributed seasonality to maternal preference of IFAS to MNPs: "Jeevan Jyoti (MNP) was distributed in winter. So if anyone fell ill then they felt it may be due to Jeevan Jyoti … while we gave syrup it was spring, it was neither too hot nor too cold, and it has suited children as well … so we came to understand that this syrup is good." The role of season was not mentioned in the other HC.
When asked how the product should be distributed to communities in the future, one ASHA responded, "If ASHA is given incentive per day then it should be done through ASHA only." As ASHAs are not salaried workers and are paid based on the services they deliver, ASHAs expressed a need for incentives to justify taking on a new task of delivering either IFAS or MNPs: "We are satisfied but unhappy because of salary, which we have not received. We feel demotivated.
We do a lot of work and every department is proud of us, that our ASHAs are field workers but the government not giving salary demotivates us." ASHAs were then asked to state for which product they should require greater incentives. Although some answered, "We should get more incentives in both," others expressed that higher incentives would be needed with IFAS distribution compared with MNP due to increased work load (eight visits per month with IFAS; two visits per month with MNP).

| DISCUSSION
We found overall high adherence and acceptability for both IFAS and MNPs among mothers and ASHAs in Bihar, India, as supplements for 6-to 23-month-old children. This research aligns with a recent systematic review of MNP use that reported high acceptability and good adherence (50-90%; de Barros & Cardoso, 2016). The higher rates of adherence in our study (>80% reported taking the MNP and IFAS as instructed) may have been due to the short duration of follow-up and/or more intensive counselling.
One of the key strengths of our study is the randomized crossover design where both products were provided to the same mother-child pair. This study design reduces confounding factors at the mother and child level. At the end of the study, mothers were evenly split on preference for which product to continue, and some ASHAs preferred IFAS due to perceived lower "waste" of product.
Few studies have carefully compared different delivery options. We found one study that was conducted in a low-income U.S. population, and although adherence was poor for both MNPs (31-46%) and iron drops (32-63%), similar to our study, there was no significant difference in which product the families would use again (Geltman et al., 2009). IFAS and MNP also had similar perceived benefits and side effects.
Overall, mothers reported numerous benefits such as increased energy, greater food consumption, and improved child health and development for both products while reporting only minor side effects for either product. This aligns with prior reports on household acceptability and side effects such as diarrhoea, vomiting, and constipation (de Barros & Cardoso, 2016).
Two factors that appeared to influence preference were (a) the order in which families received the product (with families reporting to prefer whichever product they had last) and (b) child's age. Prior research has also reported an influence of product order in acceptability trials. Novelty of products and recall bias can influence the way respondents evaluate the first versus the second product tried (Dibari et al., 2013;Gunaratna, Bosha, Belayneh, Fekadu, & De Groote, 2016;Schlich, 1993), thus, emphasizing the need for a randomized crossover design to evaluate product acceptability. Although product preference by child's age did not vary significantly in our quantitative data, the findings from the qualitative research suggested a slight preference for iron syrup for younger children who would "waste less". MNP was perceived as better suited for older children who had "become smart" and were able to eat on their own.
Delayed introduction of complementary foods is a major concern in Bihar with less than a third of children In conclusion, both iron products are acceptable policy options for preventing paediatric anaemia. This work emphasizes the need to integrate supplementation with the promotion of optimal IYCF practices.
Given the high prevalence of suboptimal feeding practices in Bihar, any iron supplementation strategy, whether IFAS or MNPs, should be supported with strong implementation of age-appropriate counselling on appropriate complementary feeding practices to maximize impacts on child health.