Using the Theory of Planned Behavior to examine effectiveness of an educational intervention on infant feeding in China

Abstract

Objective

To determine the effectiveness of an educational intervention on infant feeding behaviors and mothers' psychosocial mediating variables based on the Theory of Planned Behavior (TPB), and to assess the predictive effect of mediating variables on mothers' intention and feeding practices.

Methods

We did a cluster-randomized trial in 8 paired townships of Laishui County, China during 2006–2007. 599 mothers with infants aged 2–4-mo were enrolled at the baseline survey, of whom 485 were followed up after 11-mo intervention.

Results

The intervention group had significantly higher scores than controls in knowledge, attitudes, self-efficacy, intention, norm beliefs, as well as feeding behaviors (Hotellings T-square = 143.96, P < 0.01). Mothers' intention towards recommended feeding behaviors was positively associated with mothers' attitudes, subjective norms and self-efficacy at baseline, and was associated with their attitudes, self-efficacy, and knowledge at post-intervention evaluation. Intervention, mothers' knowledge, intention and subjective norm of villagers were independent predictors of their feeding behaviors after intervention.

Conclusions

Findings from this study suggested that TPB is an appropriate theory for explaining the effect of psychosocial factors such as knowledge, attitudes, self-efficacy, subjective norms, and intentions on infant feeding behaviors, and a useful guideline to design the targets and key approaches for infant feeding interventions.

Introduction

Safe and adequate complementary feeding has been identified as a critical factor in child malnutrition prevention. While general principles have been developed for guiding complementary feeding practices, in reality the provision of safe and adequate complementary foods is universally problematic in the developing countries because of social, economic, cultural, and behavioral barriers. Complementary foods are often inadequate in quality and quantity, and are often prepared, stored, and fed to children in unhygienic ways that may increase the risk of illness of young children (Caulfield et al., 1999). Therefore effective interventions are urgently needed to conquer the barriers to appropriate complementary feeding.

A number of nutrition education intervention programs have been carried out in developing world and showed improvement in complementary feeding for young children (Penny et al., 2005, Santos et al., 2001, Bhandari et al., 2004). But few of these studies were randomized controlled trials. Even fewer were based on theory of behavior change and examined psychosocial mediating variables which play a pivotal role in behavior change (Fatos, 2002, Maticsh and Sims, 1992). It is important to understand the mechanism by which the caretakers change their behaviors. For example, a study in China found that the intervention promoted complementary feeding messages, but parents could not always put the knowledge into practice, in part due to the influence of widespread traditional beliefs (Guldan et al., 2000). Therefore distributing knowledge and messages alone is not enough to promote behavior change. It is important to understand potential mediating factors that might contribute to the achievement of behavioral changes (Baranowski et al., 1997).

The Theory of Planned Behavior (TPB) is one of the most widely recognized and employed theories in behavior change studies. According to the theory, people's behavior is driven by three major factors: a favorable or unfavorable attitude toward the behavior, beliefs that significant individuals will approve or disapprove the intended behavior (subjective norm), and perception of behavioral control (Ajzen, 2008). In this study, we used TPB as the fundamental framework to evaluate whether an educational intervention changed mothers' knowledge, attitudes, subjective norm, self-efficacy and intentions related to infant feeding. One of the major purposes of this study was to determine whether the psychosocial mediating variables had predictive effects on feeding behaviors.