The Pathways from a Behavior Change Communication Intervention to Infant and Young Child Feeding in Bangladesh Are Mediated and Potentiated by Maternal Self-Efficacy

Measuring the intangible resources caregivers need to provide nurturing care during the complementary feeding period: a scoping review in low- and lower-middle-income countries

OBJECTIVE

Caregivers require tangible (e.g. food and financial) and intangible resources to provide care to ensure child health, nutrition and development. Intangible resources include beliefs and knowledge, education, self-efficacy, perceived physical health, mental health, healthy stress levels, social support, empowerment, equitable gender attitudes, safety and security and time sufficiency. These intangible caregiver resources are included as intermediate outcomes in nutrition conceptual frameworks yet are rarely measured as part of maternal and child nutrition research or evaluations. To facilitate their measurement, this scoping review focused on understudied caregiver resources that have been measured during the complementary feeding period in low- and lower-middle-income countries.

DESIGN

We screened 9,232 abstracts, reviewed 277 full-text articles and included 163 articles that measured caregiver resources related to complementary feeding or the nutritional status of children 6 months to 2 years of age.

RESULTS

We identified measures of each caregiver resource, though the number of measures and quality of descriptions varied widely. Most articles (77 %) measured only one caregiver resource, mental health (n 83) and social support (n 54) most frequently. Psychometric properties were often reported for mental health measures, but less commonly for other constructs. Few studies reported adapting measures for specific contexts. Existing measures for mental health, equitable gender attitudes, safety and security and time sufficiency were commonly used; other constructs lacked standardised measures.

CONCLUSIONS

Measurement of caregiver resources during the complementary feeding period is limited. Measuring caregiver resources is essential for prioritising caregivers and understanding how resources influence child care, feeding and nutrition.

Opportunities and barriers for maternal nutrition behavior change: an in-depth qualitative analysis of pregnant women and their families in Uttar Pradesh, India

Background

Maternal undernutrition during pregnancy remains a critical public health issue in India. While evidence-based interventions exist, poor program implementation and limited uptake of behavior change interventions make addressing undernutrition complex. To address this challenge, Alive & Thrive implemented interventions to strengthen interpersonal counseling, micronutrient supplement provision, and community mobilization through the government antenatal care (ANC) platform in Uttar Pradesh, India.

Objective

This qualitative study aimed to: (1) examine pregnant women's experiences of key nutrition-related behaviors (ANC attendance, consuming a diverse diet, supplement intake, weight gain monitoring, and breastfeeding intentions); (2) examine the influence of family members on these behaviors; and (3) identify key facilitators and barriers that affect behavioral adoption.

Methods

We conducted a qualitative study with in-depth interviews with 24 pregnant women, 13 husbands, and 15 mothers-in-law (MIL). We analyzed data through a thematic approach using the Capability-Opportunity-Motivation-Behavior (COM-B) framework.

Results

For ANC checkups and maternal weight gain monitoring, key facilitators were frontline worker home visits, convenient transportation, and family support, while the primary barrier was low motivation and lack understanding of the importance of ANC checkups. For dietary diversity, there was high reported capability (knowledge related to the key behavior) and most family members were aware of key recommendations; however, structural opportunity barriers (financial strain, lack of food availability and accessibility) prevented behavioral change. Opportunity ranked high for iron and folic acid supplement (IFA) intake, but was not consistently consumed due to side effects. Conversely, lack of supply was the largest barrier for calcium supplement intake. For breastfeeding, there was low overall capability and several participants described receiving inaccurate counseling messages.

Conclusion

Key drivers of maternal nutrition behavior adoption were indicator specific and varied across the capability-opportunity-motivation behavior change spectrum. Findings from this study can help to strengthen future program effectiveness by identifying specific areas of program improvement.

Effect of educational intervention based on health belief model on mothers monitoring growth of 6-12 months child with growth disorders

Background

Maternal education is one of the main ways to improve children's nutritional behaviors and development. The purpose of this study is to investigate the effect of educational intervention based on Health Belief Model (HBM) on mothers monitoring growth of 6–12 months child with growth disorders in Ghirokarzin city, Fars Provonce, Iran.

Methods

This quasi-experimental study was conducted on mothers of 6–12 months children with growth disorders of Ghirokarzin city, Fars province, Iran in 2021–2022. One hundred twenty mothers of 6–12 months child with growth disorders in Ghirokarzin city were selected using random sampling method and were divided into two groups of intervention (60) and control (60). The experimental group received training on the HBM constructs. Both groups completed the questionnaire before and three months after.intervention. A questionnaire beased on Health Belief Model constructs were used to collect information. The data was analyzed with SPSS 22 software using paired t-tests, Chi-square tests, and independent t-tests, with a significance level of 0.05.

Results

Three months after the educational intervention, the experimental group showed a significant increase in terms of knowledge, HBM constructs, weight of the children and feeding behavior.

Conclusion

This study showed the educational intervention based on the HBM improved the knowledge and feeding behavior of mothers and improved Growth Disorders of child. Hence, this model can act as a framework for designing and implementing educational interventions for prevention of growth disorders in children.

The Roles of Nutrition Education and Food Vouchers in Improving Child Nutrition: Evidence from a Field Experiment in Ethiopia

Mothers' lack of knowledge about child nutrition and limited resources lead to poor diets among children in developing countries, increasing their risk of chronic undernutrition. We implemented a cluster randomized control trial that randomly provides four-month-long Behavior Change Communication (BCC) and food vouchers in Ethiopia. We find improvements in child-feeding practices and a reduction in chronic child undernutrition only when BCC and vouchers are provided together. BCC or voucher alone had limited impacts. Our results highlight the importance of adding an effective educational component to existing transfer programs.

Maternal self-efficacy is associated with mother-child feeding practices in middle childhood

OBJECTIVE

Maternal self-efficacy (MSE) is a well-established correlate of health outcomes in young children, though little is known about this association in older children. The purpose of this cross-sectional study was to examine how MSE relates to mother-child feeding practices in middle childhood.

METHODS

A total of 306 children ages 8-12 (Mean age = 9.72, SD = 0.91, 52.9% female) and their mothers participated in this study. Mothers completed the New General Self-Efficacy Scale and the Child Feeding Questionnaire. A series of hierarchical regression models were used to examine how MSE related to maternal food restriction, maternal pressure to eat, and maternal food monitoring.

RESULTS

After accounting for maternal Body Mass Index (BMI), education, and age, and child Body Mass Index Percentile (BMI%ile) and age within the first step of each regression model, MSE emerged as a significant predictor for maternal restriction of foods (p = .024) and maternal food monitoring (p = .048). MSE was not significantly related to maternal pressure to eat.

CONCLUSIONS

This study is the first to establish that higher MSE is significantly related to healthier mother-child feeding practices in middle childhood, providing preliminary justification for enhancing MSE within comprehensive family interventions to support child weight-related health.

Maternal Capabilities Are Associated with Child Caregiving Behaviors Among Women in Rural Zimbabwe

BACKGROUND

Young children require high-quality care for healthy growth and development. We defined "maternal capabilities" as factors that influence mothers' caregiving ability (physical and mental health, social support, time, decision-making autonomy, gender norm attitudes, and mothering self-efficacy), and developed survey tools to assess them.

OBJECTIVES

We hypothesized that mothers with stronger capabilities during pregnancy would be more likely to practice improved care behaviors after their child was born.

METHODS

We assessed maternal capabilities among 4667 pregnant women newly enrolled in the Sanitation Hygiene Infant Nutrition Efficacy (SHINE) trial. Several improved child-care practices were promoted until 18 mo postpartum, the trial endpoint. Care practices were assessed by survey, direct observation, or transcription from health records during postpartum research visits. We used logistic regression to determine the predictive association between maternal capabilities during pregnancy and child-care practices.

RESULTS

Mothers with more egalitarian gender norm attitudes were more likely to have an institutional delivery [adjusted OR (AOR), 2.06; 95% CI, 1.57-2.69], initiate breastfeeding within 1 h of delivery (AOR, 1.38; 95% CI, 1.03-1.84), exclusively breastfeed (EBF) from birth to 3 mo (AOR, 2.55; 95% CI, 1.95-3.35) and 3-6 mo (AOR, 1.75; 95% CI, 1.36-2.25), and, among households randomized to receive extra modules on sanitation and hygiene, have soap and water at a handwashing station (AOR, 1.76; 95% CI, 1.29-2.39). Mothers experiencing time stress were less likely to EBF from birth to 3 mo (AOR, 0.79; 95% CI, 0.66-0.93). Greater social support was associated with institutional delivery (AOR, 1.53; 95% CI, 1.37-1.98) and, among mothers randomized to receive extra complementary feeding modules, feeding children a minimally diverse diet (AOR, 1.18; 95% CI, 1.01-1.37). Depressed mothers were 37% and 33%, respectively, less likely to have an institutional delivery (AOR, 0.63; 95% CI, 0.44-0.88) and a fully immunized child (AOR, 0.67; 95% CI, 0.50-0.90).

CONCLUSIONS

Interventions to reduce maternal depression, time stress, inadequate social support, and inequitable gender norms may improve maternal child caregiving.

Yes, no, maybe so: the importance of cognitive interviewing to enhance structured surveys on respectful maternity care in northern India

Quantitative survey findings are important in measuring health-related phenomena, including on sensitive topics such as respectful maternity care (RMC). But how well do survey results truly capture respondent experiences and opinions? Quantitative tool development and piloting often involve translating questions from other settings and assessing the mechanics of implementation, which fails to deeply explore how respondents understand survey questions and response options. To address this gap, we conducted cognitive interviews on survey questions (n = 88) adapted from validated RMC instruments used in Ethiopia, Kenya and elsewhere in India. Cognitive interviews with rural women (n = 21) in Madhya Pradesh, India involved asking the respondent the survey question, recording her response, then interviewing her about what the question and response options meant to her. We analysed the interviews to revise the tool and identify question failures, which we grouped into six areas: issues with sequencing, length and sensitivity; problematic response options; inappropriate vocabulary; temporal and spatial confusion; accessing different cognitive domains; and failure to resonate with the respondent’s worldview and reality. Although women tended to provide initial answers to the survey questions, cognitive interviews revealed widespread mismatch between respondent interpretation and question intent. Likert scale response options were generally incomprehensible and questions involving hypothetical scenarios could be interpreted in unexpected ways. Many key terms and concepts from the international RMC literature did not translate well and showed low resonance with respondents, including consent and being involved in decisions about one’s care. This study highlights the threat to data quality and the validity of findings when translating quantitative surveys between languages and cultures and showcases the value of cognitive interviews in identifying question failures. While survey tool revision can address many of these issues, further critical discussion is needed on the use of standardized questions to assess the same domains across contexts.

A qualitative assessment of barriers and facilitators to implementing recommended infant nutrition practices in Mumbai, India

BACKGROUND

Childhood malnutrition has been a longstanding crisis in Mumbai, India. Despite national IYCF (Infant Young Child Feeding) guidelines to promote best practices for infant/toddler feeding, nearly one-third of children under age five are stunted or underweight. To improve child nutrition, interventions should address the cultural, social, and environmental influences on infant feeding practices. This study is an in-depth qualitative assessment of family barriers and facilitators to implementing recommended nutrition practices in two Mumbai slum communities, within the context of an existing nutrition education-based intervention by a local non-governmental non-profit organization.

METHODS

The population was purposively sampled to represent a variety of household demographics. Data were collected through 33 in-depth semi-structured interviews with caregivers (mothers and paternal grandmothers) of children age 0-2 years. Transcripts were translated and transcribed, and analyzed using qualitative analysis procedures and software.

RESULTS

A complex set of barriers and facilitators influence mothers'/caregivers' infant-toddler feeding practices. Most infants were fed complementary foods and non-nutritious processed snacks, counter to IYCF recommendations. Key barriers included: lack of nutrition knowledge and experience, receiving conflicting messages from different sources, limited social support, and poor self-efficacy for maternal decision-making. Key facilitators included: professional nutrition guidance, personal self-efficacy and empowerment, and family support. Interventions to improve child nutrition should address mothers'/caregivers' key barriers and facilitators to recommended infant-toddler feeding practices.

CONCLUSIONS

Nutrition interventions should prioritize standard messaging across healthcare providers, engage all family members, target prevention of early introduction of sugary and non-nutritious processed foods, and strengthen maternal self-efficacy for following IYCF recommended guidelines.

Interventions for increasing fruit and vegetable consumption in children aged five years and under

BACKGROUND

Insufficient consumption of fruits and vegetables in childhood increases the risk of future non-communicable diseases, including cardiovascular disease. Testing the effects of interventions to increase consumption of fruit and vegetables, including those focused on specific child-feeding strategies or broader multicomponent interventions targeting the home or childcare environment is required to assess the potential to reduce this disease burden.

OBJECTIVES

To assess the effectiveness, cost effectiveness and associated adverse events of interventions designed to increase the consumption of fruit, vegetables or both amongst children aged five years and under.

SEARCH METHODS

We searched CENTRAL, MEDLINE, Embase and two clinical trials registries to identify eligible trials on 25 January 2020. We searched Proquest Dissertations and Theses in November 2019. We reviewed reference lists of included trials and handsearched three international nutrition journals. We contacted authors of included trials to identify further potentially relevant trials.

SELECTION CRITERIA

We included randomised controlled trials, including cluster-randomised controlled trials and cross-over trials, of any intervention primarily targeting consumption of fruit, vegetables or both among children aged five years and under, and incorporating a dietary or biochemical assessment of fruit or vegetable consumption. Two review authors independently screened titles and abstracts of identified papers; a third review author resolved disagreements.

DATA COLLECTION AND ANALYSIS

Two review authors independently extracted data and assessed the risks of bias of included trials; a third review author resolved disagreements. Due to unexplained heterogeneity, we used random-effects models in meta-analyses for the primary review outcomes where we identified sufficient trials. We calculated standardised mean differences (SMDs) to account for the heterogeneity of fruit and vegetable consumption measures. We conducted assessments of risks of bias and evaluated the quality of evidence (GRADE approach) using Cochrane procedures.

MAIN RESULTS

We included 80 trials with 218 trial arms and 12,965 participants. Fifty trials examined the impact of child-feeding practices (e.g. repeated food exposure) in increasing child vegetable intake. Fifteen trials examined the impact of parent nutrition education only in increasing child fruit and vegetable intake. Fourteen trials examined the impact of multicomponent interventions (e.g. parent nutrition education and preschool policy changes) in increasing child fruit and vegetable intake. Two trials examined the effect of a nutrition education intervention delivered to children in increasing child fruit and vegetable intake. One trial examined the impact of a child-focused mindfulness intervention in increasing vegetable intake. We judged 23 of the 80 included trials as free from high risks of bias across all domains. Performance, detection and attrition bias were the most common domains judged at high risk of bias for the remaining trials. There is low-quality evidence that child-feeding practices versus no intervention may have a small positive effect on child vegetable consumption, equivalent to an increase of 5.30 grams as-desired consumption of vegetables (SMD 0.50, 95% CI 0.29 to 0.71; 19 trials, 2140 participants; mean post-intervention follow-up = 8.3 weeks). Multicomponent interventions versus no intervention has a small effect on child consumption of fruit and vegetables (SMD 0.32, 95% CI 0.09 to 0.55; 9 trials, 2961 participants; moderate-quality evidence; mean post-intervention follow-up = 5.4 weeks), equivalent to an increase of 0.34 cups of fruit and vegetables a day. It is uncertain whether there are any short-term differences in child consumption of fruit and vegetables in meta-analyses of trials examining parent nutrition education versus no intervention (SMD 0.13, 95% CI -0.02 to 0.28; 11 trials, 3050 participants; very low-quality evidence; mean post-intervention follow-up = 13.2 weeks). We were unable to pool child nutrition education interventions in meta-analysis; both trials reported a positive intervention effect on child consumption of fruit and vegetables (low-quality evidence). Very few trials reported long-term effectiveness (6 trials), cost effectiveness (1 trial) or unintended adverse consequences of interventions (2 trials), limiting our ability to assess these outcomes. Trials reported receiving governmental or charitable funds, except for four trials reporting industry funding.

AUTHORS' CONCLUSIONS

Despite identifying 80 eligible trials of various intervention approaches, the evidence for how to increase children's fruit and vegetable consumption remains limited in terms of quality of evidence and magnitude of effect. Of the types of interventions identified, there was moderate-quality evidence that multicomponent interventions probably lead to, and low-quality evidence that child-feeding practice may lead to, only small increases in fruit and vegetable consumption in children aged five years and under. It is uncertain whether parent nutrition education or child nutrition education interventions alone are effective in increasing fruit and vegetable consumption in children aged five years and under. Our confidence in effect estimates for all intervention approaches, with the exception of multicomponent interventions, is limited on the basis of the very low to low-quality evidence. Long-term follow-up of at least 12 months is required and future research should adopt more rigorous methods to advance the field. This is a living systematic review. Living systematic reviews offer a new approach to review updating, in which the review is continually updated, incorporating relevant new evidence as it becomes available. Please refer to the Cochrane Database of Systematic Reviews for the current status of this review.

"If there is no water, we cannot feed our children": The far-reaching consequences of water insecurity on infant feeding practices and infant health across 16 low- and middle-income countries

OBJECTIVES

Infant feeding plays a critical role in child health and development. Few studies to date have examined the link between household water insecurity and infant feeding, and none in a cross-cultural context. Therefore, we examined the perceived impact of household water insecurity in four domains: breastfeeding, non-breastmilk feeding, caregiver capabilities, and infant health. Our research was conducted as part of the Household Water Insecurity Experiences (HWISE) study.

METHODS

We interviewed respondents from 19 sites in 16 low- and middle-income countries (N = 3303) about the link between water insecurity and infant feeding. We then thematically analyzed their open-ended textual responses. In each of the four domains (breastfeeding, non-breastmilk feeding, caregiver capabilities, infant health), we inductively identified cross-cultural metathemes. We analyzed the distribution of themes across sites quantitatively and qualitatively.

RESULTS

Water was perceived to directly affect breastfeeding and non-breastmilk feeding via numerous pathways, including timing and frequency of feeding, unclean foods, and reduced dietary diversity. Water was perceived to indirectly affect infant feeding through caregiver capabilities by increasing time demands, exacerbating disease, undernutrition, and mortality, and requiring greater efficacy of caregivers. Respondents made connections between water challenges and infant health, for example, increased risk of infectious diseases, undernutrition, and mortality.

CONCLUSIONS

These findings suggest that water presents many, and sometimes unexpected, challenges to infant feeding. By systematically investigating biocultural pathways by which water impacts infant and young child feeding, it will be possible to understand if, and how, water security can be leveraged to improve child nutrition and health.

Program Impact Pathways and Contexts: A Commentary on Theoretical Issues and Research Applications to Support the EsIAN Component of Mexico's Conditional Cash Transfer Program

This commentary on the Integrated Strategy for Attention to Nutrition (EsIAN) journal supplement begins with a discussion about the challenges that implementation researchers confront with respect to analyzing complex impact pathways. We note that the research on the implementation of the EsIAN component of Mexico's conditional cash transfer program was based implicitly or explicitly on a program impact pathway approach, which used both quantitative and qualitative methods to examine bottlenecks in program implementation. We then identify 5 categories of contexts that affect the impact, implementation, and survival of intervention programs: 1) biological, 2) social-cultural, 3) delivery modalities and platforms, 4) bureaucratic, and 5) political. Each of these contexts presents theoretical and methodological challenges for investigators. In this commentary, we focus primarily on biological and social-cultural contexts, discussing the theoretical and methodological challenges the investigators faced and the research strategies they used to address them, which have produced a unique compilation of "learning by doing" studies. We also touch briefly on the political context in which the Prospera program research was conducted. We conclude with statements that highlight the exceptional value of the journal supplement, not only with respect to the analysis of the interventions the studies cover and the sustained examination of a long-term program but also as a major contribution to the literature in implementation science in nutrition.

Interventions for increasing fruit and vegetable consumption in children aged five years and under

BACKGROUND

Insufficient consumption of fruits and vegetables in childhood increases the risk of future non-communicable diseases, including cardiovascular disease. Interventions to increase consumption of fruit and vegetables, such as those focused on specific child-feeding strategies and parent nutrition education interventions in early childhood may therefore be an effective strategy in reducing this disease burden.

OBJECTIVES

To assess the effectiveness, cost effectiveness and associated adverse events of interventions designed to increase the consumption of fruit, vegetables or both amongst children aged five years and under.

SEARCH METHODS

We searched CENTRAL, MEDLINE, Embase and two clinical trials registries to identify eligible trials on 25 August 2019. We searched Proquest Dissertations and Theses in May 2019. We reviewed reference lists of included trials and handsearched three international nutrition journals. We contacted authors of included trials to identify further potentially relevant trials.

SELECTION CRITERIA

We included randomised controlled trials, including cluster-randomised controlled trials and cross-over trials, of any intervention primarily targeting consumption of fruit, vegetables or both among children aged five years and under, and incorporating a dietary or biochemical assessment of fruit or vegetable consumption. Two review authors independently screened titles and abstracts of identified papers; a third review author resolved disagreements.

DATA COLLECTION AND ANALYSIS

Two review authors independently extracted data and assessed the risks of bias of included trials; a third review author resolved disagreements. Due to unexplained heterogeneity, we used random-effects models in meta-analyses for the primary review outcomes where we identified sufficient trials. We calculated standardised mean differences (SMDs) to account for the heterogeneity of fruit and vegetable consumption measures. We conducted assessments of risks of bias and evaluated the quality of evidence (GRADE approach) using Cochrane procedures.

MAIN RESULTS

We included 78 trials with 214 trial arms and 13,746 participants. Forty-eight trials examined the impact of child-feeding practices (e.g. repeated food exposure) in increasing child vegetable intake. Fifteen trials examined the impact of parent nutrition education in increasing child fruit and vegetable intake. Fourteen trials examined the impact of multicomponent interventions (e.g. parent nutrition education and preschool policy changes) in increasing child fruit and vegetable intake. Two trials examined the effect of a nutrition education intervention delivered to children in increasing child fruit and vegetable intake. One trial examined the impact of a child-focused mindfulness intervention in increasing vegetable intake. We judged 20 of the 78 included trials as free from high risks of bias across all domains. Performance, detection and attrition bias were the most common domains judged at high risk of bias for the remaining trials. There is very low-quality evidence that child-feeding practices versus no intervention may have a small positive effect on child vegetable consumption equivalent to an increase of 4.45 g as-desired consumption of vegetables (SMD 0.42, 95% CI 0.23 to 0.60; 18 trials, 2004 participants; mean post-intervention follow-up = 8.2 weeks). Multicomponent interventions versus no intervention has a small effect on child consumption of fruit and vegetables (SMD 0.34, 95% CI 0.10 to 0.57; 9 trials, 3022 participants; moderate-quality evidence; mean post-intervention follow-up = 5.4 weeks), equivalent to an increase of 0.36 cups of fruit and vegetables per day. It is uncertain whether there are any short-term differences in child consumption of fruit and vegetables in meta-analyses of trials examining parent nutrition education versus no intervention (SMD 0.12, 95% CI -0.03 to 0.28; 11 trials, 3078 participants; very low-quality evidence; mean post-intervention follow-up = 13.2 weeks). We were unable to pool child nutrition education interventions in meta-analysis; both trials reported a positive intervention effect on child consumption of fruit and vegetables (low-quality evidence). Very few trials reported long-term effectiveness (6 trials), cost effectiveness (1 trial) and unintended adverse consequences of interventions (2 trials), limiting their assessment. Trials reported receiving governmental or charitable funds, except for four trials reporting industry funding.

AUTHORS' CONCLUSIONS

Despite identifying 78 eligible trials of various intervention approaches, the evidence for how to increase children's fruit and vegetable consumption remains limited. There was very low-quality evidence that child-feeding practice may lead to, and moderate-quality evidence that multicomponent interventions probably lead to small increases in fruit and vegetable consumption in children aged five years and younger. It is uncertain whether parent nutrition education interventions are effective in increasing fruit and vegetable consumption in children aged five years and younger. Given that the quality of the evidence is very low or low, future research will likely change estimates and conclusions. Long-term follow-up of at least 12 months is required and future research should adopt more rigorous methods to advance the field. This is a living systematic review. Living systematic reviews offer a new approach to review updating, in which the review is continually updated, incorporating relevant new evidence as it becomes available. Please refer to the Cochrane Database of Systematic Reviews for the current status of this review.

Information Diffusion and Social Norms Are Associated with Infant and Young Child Feeding Practices in Bangladesh

BACKGROUND

Interaction within mothers' social networks can theoretically diffuse messages from interventions and campaigns into norms and practices for infant and young child feeding (IYCF).

OBJECTIVES

We hypothesized that mothers' social networks, diffusion of information, and social norms differed in intensive [intensive interpersonal counseling (IPC), community mobilization (CM), and mass media (MM)] compared with nonintensive (standard IPC and less-intensive CM and MM) intervention areas, were associated with IYCF practices, and partly explained practice improvement.

METHODS

We conducted household surveys at endline in 2014 and follow-up in 2016 (n = ∼2000 each round). We used multiple regression to test differences and changes in networks, diffusion, and norms within intervention areas. We analyzed paths from intervention exposure to IYCF practices through networks, diffusion, and norms.

RESULTS

Mothers' networks were larger in intensive than in nonintensive areas in 2014 and increased in both areas over time [25-38 percentage points (pp)]. The prevalence of receipt of IYCF information was high, with no changes over time in intensive areas but an increase in nonintensive areas (8-16 pp). In both areas, more family members and health workers provided IYCF information over time. Sharing of information increased 17-23 pp in intensive and 11-41 pp in nonintensive areas over time. Perceived descriptive norms improved 8-16 pp in intensive and 17-28 pp in nonintensive areas. Perceived injunctive norms were high in both areas. Breastfeeding practices were associated with networks, diffusion, and norms (OR: 1.6-4.4 times larger comparing highest with lowest quartile). Minimum dietary diversity was associated with larger networks and diffusion (OR: 1.5-2.2) but not with social norms. Indirect paths from intervention exposure to practices explained 34-78% of total effects.

CONCLUSIONS

Diffusion of IYCF information through social networks, reinforced by positive social norms for messages promoted over time, will contribute to positive changes in IYCF practices that may be achieved and sustained through large-scale social and behavior change interventions. This trial was registered at clinicaltrials.gov as NCT0274084.