The Spanish version of the Home Environment Survey (HES) among families of children with overweight/obesity: a validation study.

PURPOSE

The aim of this article was to validate the Spanish version of the Home Environment Survey (HES-S) and was divided in two studies: (1) to assess the reliability, convergent validity of HES-S in a survey of 145 parents of children with overweight/obesity; (2) to study the magnitude of the association between children's BMI status with the latent scores theoretically defined by the HES model.

METHODS

To test the scale and the model, a confirmatory factor analysis (CFA) and a path analysis were carried out among a sample of 156 parents of preadolescents (106 overweight/obesity and 50 normal-weight children). No CFA or EFA were carried out in the validation of the original instrument.

RESULTS

Study 1, both the Physical Activity and the Eating Habits components of the scale showed adequate levels of internal consistency for the majority of the scales, except for two. One of them, Healthy Eating Parental Policies (HEP) subscale was reduced after excluded two items, although it did not improve substantially. This model indicated that there was a significant association between the two Eating Habits scales and the child's weight status, but child's weight was not associated with the Physical Activity components. Convergent validity was confirmed by correlations with related variables: family eating habits (F-EAT), parent's physical activity (IPAQ), and children's physical activity (assessed via accelerometers during one week). Study 2, our results replicated the original four factor structure proposed for physical activity (CFI = 0.99; RMSEA = 0.03), but the original factor structure of the eating habits component was not supported. In addition, the relationship of the child's weight status, the Physical Activity components, and the two scales of Eating Habits (Parental Modeling and Policies) was explored with a path analysis showing good fit indices (CFI = 0.95; RMSEA = 0.06). Child's BMI was negatively associated with Healthy Eating Parental Role Modeling (r =  - 0.21) and with Healthy Eating Parental Policies (r =  - 0.19), but not with the factors of Child's Physical Activity model.

CONCLUSION

To our knowledge, this is the first instrument to assess obesogenic family environment in Spanish speaking countries, which is a relevant dimension within a health perspective so as to implement new policies and strategies in obesity tertiary prevention. Overall, the confirmatory factor analysis of the HES-S has only provided additional support for one part related to Physical Activity. In addition, Child's BMI was correlated with scales of Eating Habits but not with Child's Physical Activity factor. These results clearly suggest that further research is warranted.

LEVEL III

Case-control analytic study.