Different Combinations of Behavior Change Interventions and Frequencies of Interpersonal Contacts Are Associated With Infant and Young Child Feeding Practices in Bangladesh, Ethiopia, and Vietnam

Background: Social and behavior change communication interventions are integral to improving dietary and care practices, but evidence on the impact of the combination and intensity of these interventions in different contexts is scarce. Objectives: We examined the extent of and factors associated with intervention exposure: interpersonal communication (IPC) alone or with other interventions (i.e., mass media, community mobilization, or nutrition-sensitive agricultural activities), number of and factors associated with IPC contacts, and combinations of intervention components and number of contacts associated with infant and young child feeding (IYCF) practices. Methods: We used endline survey data from impact evaluations in Bangladesh, Ethiopia, and Vietnam ( = 1001, 1720, and 1001 mothers with children agedy, respectively). Multivariable regression models were used for analyses. Results: Exposure to the interventions varied in all 3 countries. On average, mothers received 8 visits in the last 6 mo in Bangladesh, 2 visits in the last 3 mo in Ethiopia, and 1 visit in the last 6 mo in Vietnam. Across countries, the factors associated with intervention exposure and number of IPC contacts differed. In Ethiopia, exposure to IPC with other interventions was associated with higher odds of achieving minimum meal frequency (OR: 1.6), minimum dietary diversity (OR: 1.8), and consumption of iron-rich foods (OR: 4.7). In Vietnam, exposure to IPC alone or with mass media was associated with higher odds of exclusive breastfeeding (EBF; OR: 2.8-3.7). Near-monthly visits were associated with 2-3 times higher odds of IYCF practices in Bangladesh and Ethiopia. In Vietnam, even 1 IPC visit was associated with 2 times higher odds of EBF. Conclusions: Exposure matters for impact, but the combination of behavior change interventions and number of IPC contacts required to support IYCF behavior change are context specific. This trial was registered at www.clinicaltrials.gov as NCT01678716 (Bangladesh), NCT02775552 (Ethiopia), and NCT01676623 (Vietnam)