Berlin, Daniel. Food consumption behavior of american adults nationally (food away from home chain frequency) and the implications for their health. Retrieved from https://doi.org/doi:10.7282/t3-fqkh-8k34
DescriptionThis research utilizes the National Household Food Acquisition and Purchase Survey (FoodAPS) dataset collected by the USDA/ERS to analyze the effect that food consumption behavior has on the health of American adults. The objectives of this study are to examine American adults’ chain frequency behavior, examine if differences exist in nutritional content and consumption of chain food regionally, and to examine the implications of choices made by various socio-demographic groups on consumers’ health. The sample was reduced to only include primary survey respondents who actually purchased and consumed food from chain outlets during the survey period.
I examine the associations and effects that nutritional awareness has on the frequency of chain usage using a uniquely constructed nutritional awareness scoring index created from various FoodAPS survey questions. I compare my findings on food consumption behavior across various regions of the country and across various consumer characteristics to examine if differences exist and where they may be. Means difference and association/correlation tests are used to identify significant differences by variables between groups. Ordinal regression is used to examine the effect of socio-demographic variables on chain usage. Finally, multinomial regressions and discriminant analysis models are used to examine the impact of socio-demographic variables on health outcome (BMI).
I find supporting evidence that socio-demographic variables such as age, race, income, education, ethnicity, and nutritional awareness have impacts on both chain usage behavior as well as resulting health outcomes. For example, Asian individuals had the lowest levels of chain consumption (and the best BMI outcomes) while Black individuals had the highest chain visits. In terms of education, we found that as education levels increased, individuals also had increased rates of chain usage. As age increased, we found a reduced amount of chain usage in the higher age categories. I also find that an increase in amount spent at chains per visit had a significant association with lower chain usage and better BMI outcome of individuals. Individuals who spent the most per chain visit had significantly lower chain usage, as well as better health outcomes. This may be due to purchasing better quality foods at chain outlets. Poorer, less educated, and black individuals were found to consistently have worse health outcomes from chain visits.
This thesis also provides policy implications by war of recommending an increased focus on nutritional education programs particularly for those in the lower socio-demographic groups. I suggest an expansion of the FDA menu labeling requirements to apply to restaurants and food establishments that are also non-chains (less than 20 locations) due finding that individuals did not make more healthful purchases in non-chain outlets as opposed to chain outlets. Instead, I find that individuals exhibited similar nutritional behavior at both outlet types and recommend that non-chain outlets also be required to post nutritional information.