Premium
Context Matters: Insights from Two Randomized Evaluations of Behavior Change Interventions on Factors Influencing Infant and Young Child Feeding Practices in Bangladesh and Vietnam
Author(s) -
Me Purnima,
Nguyen Phuong,
Kim Sunny,
Tran Lan,
Frongillo Edward,
Ruel Marie,
Rawat Rahul
Publication year - 2017
Publication title -
the faseb journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.709
H-Index - 277
eISSN - 1530-6860
pISSN - 0892-6638
DOI - 10.1096/fasebj.31.1_supplement.165.6
Subject(s) - breastfeeding , medicine , context (archaeology) , psychological intervention , behavior change communication , environmental health , breast feeding , logistic regression , confounding , demography , population , pediatrics , nursing , geography , archaeology , pathology , sociology , research methodology
Background Appropriate infant and young child feeding (IYCF) practices contribute to optimal child growth and development. Behavior change interventions (BCI) can improve IYCF, but lessons on effectiveness at large‐scale are limited. In two evaluations of large‐scale BCI in Bangladesh and Vietnam, we showed impact on breastfeeding and complementary feeding practices. Not all mothers practiced recommended behaviors; therefore, we examined factors associated with IYCF practices promoted by the BCI. Methods We used cross‐sectional endline data from cluster‐randomized impact evaluations in Bangladesh and Vietnam. For this analysis, survey data from mothers with children < 2y in intensive intervention areas who had been exposed to interpersonal counselling (IPC) by health workers in Bangladesh (n=969) and in Vietnam (n=553) were used. Multivariable logistic regression analyses were used to examine factors associated with IYCF practices, adjusting for confounders and geographic clustering. Results Exposure to IPC was high in Bangladesh (96%) and moderate in Vietnam (55%). In Bangladesh, reported practice was 95% for early initiation of breastfeeding (EIBF), 88% for exclusive breastfeeding (EBF), and 64% for minimum dietary diversity (MDD). In Vietnam, reported practice was 60% for EIBF, 71% for EBF, and 93% for MDD. In both countries, Caesarian (C)‐section births, lack of breastfeeding support at birth, and poor knowledge were associated with lower EIBF. Mothers who had C‐sections were 85–93% less likely to practice EIBF (OR: 0.16: 95% CI: 0.07–0.39 in Bangladesh and OR: 0.07, 95% CI: 0.03–0.19 in Vietnam). Return to work before the child was 6 mo old (OR: 0.40, 95% CI: 0.18–0.90) and maternal stress (OR: 0.36, 95% CI: 0.17–0.78) were associated with lower EBF in Vietnam; in both countries, mothers with low knowledge were 69–84% less likely to practice EBF. In Bangladesh, compared to those in the highest SES tertile, babies of mothers in the lowest tertile of SES were 62% less likely to achieve MDD (OR: 0.38, 95% CI: 0.27–0.53). Conclusions Our results emphasize that contextual determinants are highly specific to individual behaviors promoted by BCI, and therefore, require specific accompanying measures to support each set of IYCF practices. For example, improving EIBF in Bangladesh and Vietnam may require more supportive health facility birth environments, particularly for C‐section births. Similarly, where women's participation in the work‐force is high, policies that support later return to work after delivery may help enhance BCI impact on EBF. Finally, in poor environments, cash or food transfer programs accompanying BCI may help families more easily increase diet diversity. Thus, high quality BCI interventions should be coupled with policies, programs or interventions that address behaviour‐specific constraints that lie beyond lack of knowledge, skills and motivation. Support or Funding Information Bill & Melinda Gates Foundation, through Alive & Thrive, managed by FHI 360; CGIAR Research Program on Agriculture for Nutrition and Health, led by IFPRI.