Integration of food intake biomarker data with minimal self-reported dietary information to assess eating behaviour and evaluate nutrient intake

为了清楚地了解食物暴露、营养状况或疾病风险之间的关系，需要准确评估膳食摄入量，并结合营养代谢型的测量，以解释食物成分x营养素和食物相互作用的个体代谢处理的变异性。饮食暴露传统上是使用自我报告的措施，如食物频率表（FFQ）。不幸的是，自我报告的方法是主观的，误报是常见的，这大大降低了营养研究的价值。此外，由于现代饮食的复杂性，捕获自我报告的信息通常需要受过训练的研究人员的大量输入以帮助研究/调查参与者，这增加了营养研究的成本和显著的参与者负担。为了帮助解决这些问题，并为数字化和扩大研究规模提供空间，阅读大学的团队开发并验证了在线FFQ（eNutri）。FFQ包括食物部分照片，并通过网络应用程序自动分析，大大减轻了研究参与者和研究人员的负担。阿伯里斯特威斯的研究小组已经证明，在家中收集的现场（第一个早晨的空隙）尿液样本中存在的食物中产生的代谢物，并张贴到分析实验室，为饮食暴露和个人营养代谢型提供生物标志物。虽然覆盖范围已经包括了广泛的具有高度公共卫生意义的食物（包括红肉，油性鱼，全麦，水果，叶菜），但尿液中不太可能存在暴露于普通淀粉或高脂肪食物的生物标志物。在一系列临床试验/饮食调查中与合作者合作（包括阅读）已经开发了一个尿样生物库，其中包括与食物摄入有关的大量元数据，这将是验证尿液生物标志物的性能和调查抽样人群评估习惯性饮食和营养代谢型的最有效方法的理想选择。在阿伯里斯特威斯和阅读的研究中，并验证目前基于尿液生物标志物和/或FFQ数据的英国人群的客观饮食质量指数（DQI：类似于美国开发的替代健康饮食指数）。我们假设，通过整合和分析来自尿液代谢组学和使用eNutri FFQ数据库开发的迷你FFQ的膳食暴露数据，可以开发客观的膳食暴露数据和反映真实营养状况和个体营养代谢型的“健康饮食指数”。

For clear understanding of the relationship between food exposure, nutrient status or disease risk there is a need for an accurate assessment of dietary intake combined with measures of nutritional metabotype that account for variability in food composition x individual metabolic handling of nutrients and foods interactions. Dietary exposure is recorded traditionally using self-reported measures such as Food Frequency Questionnaires (FFQs). Unfortunately, self-reporting methods are subjective, and misreporting is common which greatly reduces the value of nutritional research. In addition, capturing self-reported information generally requires substantial input from trained researchers to assist study/survey participants due to the complexity of modern diets which adds cost and significant participant burden to nutrition research. To help counteract these problems and provide scope for digitising and scaling up of research studies, the team in Reading University has developed and validated an online FFQ (eNutri). The FFQ includes food portion photographs and is automatically analysed via a web-application and greatly reduces the burden to both study participants and researchers. The team in Aberystwyth has demonstrated that metabolites derived from foods present in spot (First Morning Void) urine samples collected in the home and posted to an analytical laboratory provide biomarkers for dietary exposure and individual nutritional metabotype. Although coverage already includes a wide range of foods of high public health significance (including e.g. red meat, oily fish, wholegrain, fruits, leafy vegetables), it is unlikely that biomarkers for exposure to common starchy or high fat foods will be present in urine. Working with collaborators in a range of clinical trials/dietary surveys (including Reading) a biobank of urine samples has been developed, complete with substantial meta-data relating to food intake, that will be ideal for validating the performance of urine biomarkers and investigating the most effective way of sampling populations to assess habitual diet and nutritional metabotypeCommon to research in both Aberystwyth and Reading is the drive to develop and validate an objective Diet Quality Index (DQI: similar to the Alternative Healthy Eating Index developed in US) for the UK population based currently on urine biomarker and/or FFQ data. We hypothesise that objective dietary exposure data and a 'Healthy Eating Index' reflective of true nutritional status and individual nutritional metabotype can be developed by integration and analysis of dietary exposure data derived from both urine metabolome and a mini-FFQ developed using the eNutri FFQ database.