Assessing snacking and drinking behavior in Real-Life Settings : Validation of the SnackBox technology

Summary

Background: Understanding snacking behavior and its determinants is crucial in addressing modern dietary patterns. However, existing methods, such as self-reports and controlled laboratory studies, exhibit limitations in reliability capturing snacking and drinking behavior in real-life settings. This study aimed to validate the SnackBox, a novel technology designed to objectively assess snacking behavior with minimal user effort and researcher interference. Methods: The SnackBox, capable of storing three different snacks or beverages, automatically measured the quantity consumed (in grams) and the time of consumption. Forty-eight office workers (29 female, mean age = 28.8, SD = 4.1) used the SnackBox at their workplace and home over a two-week period, recording their snack consumption. On measurement days, participants completed 5-hour dietary recalls through a mobile application. Ad libitum portions of snacks and beverages were provided, and post-study measurements of leftovers determined true consumption. Results: The SnackBox demonstrated higher reliability (ICC = 0.80, 95% CI = 0.72 – 1.85) in estimating consumed snack quantities compared to the 5-hour dietary recall application (ICC = 0.60, 95% CI = 0.45 – 0.70). Participants rated the SnackBox's usability (M = 0.84, SD = 0.88) significantly higher than that of the self-report app (M = -0.58, SD = 0.98) (p <.001). Conclusion: The SnackBox is a valid and low-burden tool for assessing snacking and drinking behavior in real-life settings. Its objective data collection surpasses self-reports, offering insights into snacking quantities. This has the potential to advance our understanding and interventions for healthier dietary habits.