Nutrition in the digital age - How digital tools can help to solve the personalized nutrition conundrum. (August 2019)

• Record Type: Journal Article
• Authors: Michel, M.; Burbidge, A.
• Is Part Of: Trends in food science & technology. Volume 90(2019)
• Page Start: 194
• Page End: 200
• Publication Date: 2019-08
• Subjects: Nutrition -- Personalization -- Life-style -- Non-communicable diseases -- Digital self-learning expert system -- Health; Food industry and trade -- Periodicals; Food -- Biotechnology -- Periodicals; 664.005
• DOI: 10.1016/j.tifs.2019.02.018
• Languages: English
• ISSNs: 0924-2244

Abstract

Abstract: Background: It has become increasingly clear that the current population averaged nutrition paradigm is not able to address the growing non-communicable disease (NCD) epidemics . Current approaches fail primarily for two reasons: firstly, poor adherence to public dietary advice and, secondly, individual health responses are not well reflected by population averages, as generic public health advice lacks relevance for individuals, personally and medically. Scope and approach: Personalized 'expert' systems are, potentially, powerful weapons against NCDs . Existing systems are, however, handicapped by both difficulty in measuring dietary intake reliably and that of tying population level nutritional knowledge to stochastic individual responses. In order to address these shortcomings, we propose an approach that no longer distinguishes between behavioural and physiological responses. Key findings and conclusions: We outline how a conceptual self-learning expert system could implement this approach, based on multifactorial lifestyle interventions, and give specific examples in the contexts of diabetes and obesity. Combining behaviour and physiological responses into a single entity removes the requirement to measure food intake, enabling users to map their individualised 'path of least resistance' to specific health outcomes. This new approach could be provided at minimal cost by leveraging users existing mobile devices, e.g. smart phones, watches, fitbits etc. The novelty in this concept is that the methodology purposely does not attempt to understand the complexity of the underlying physiological, metabolic and psychological responses. Despite requiring scientifically validated biomarkers, understanding the discrete influences of each of these factors is not required to drive improved individual-level outcomes. Highlights: Self-learning systems have the potential to improve an individual's health through diet management. Current apps focus on measuring food intake, which is extremely difficult to achieve. Combining behaviour and physiological responses into a single entity removes the requirement to measure food intake. The proposed system enables the user to painlessly map their individualised 'path of least resistance' to specific health outcomes. Deep scientific understanding, beyond scientifically validated biomarkers, is not necessary in order to optimise individual responses.