Modified-Malay Eating Behavior and Pattern Questionnaire (Malay-EBPQ): translation and validation among Malaysian women.

ABSTRACT

BACKGROUND: Eating behavior primarily depends on eating patterns which are largely influenced by interactions between physiology, environment, psychology, culture and socio-economic status. This study was designed to translate and validate the Eating Behavior Pattern Questionnaire (EBPQ) among Malaysian women. METHODS: A cross-sectional study involving translation and validation of the English version of EBPQ. The original questionnaire, contained 51 items extracted into six domains was translated in Malay using forward and backward translation, pre-tested and validated among conveniently sampled female healthcare personnel. Vegetarians, pregnant ladies and women in confinement were excluded due to special daily dietary plans. Construct validity, reliability and feasibility were analyzed using Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA). RESULTS: During translation, item modifications were made and subjected to field testing among 394 women. The original questionnaire was used as a reference to identify the positioning of items in constructs. Fifteen items were removed due to poor correlation with items within constructs. Seven factors were extracted using Varimax rotation with Kaiser-Meyer-Olkin (KMO) value range from 0.725-0.872 and significant Bartlett's test of Sphericity (p < 0.001). The item-loading of the items within the constructs ranged between 0.415-0.812 (explained variation = 62.7%). Cultural and lifestyle behavior was relabeled to lifestyle and behavioral eating, and snacking on sweets was relabeled as snacking pattern. Emotional eating was divided into two sub-factors as snacking behavior and emotional influence. CFA resulted with an acceptable fit with no presence of floor and ceiling effects. Intra-class correlation coefficient (ICC) for all the constructs were reported good and excellent. The overall internal consistency was reported as good. CONCLUSION: The modified 36-item Malay-EBPQ had moderate internal consistency, reliable and fit with multi-dimensional measures of eating behaviors and dietary patterns among women in the multi-racial population with cultural diversity.