Department of Epidemiology, German Institute of Human Nutrition Potsdam-Rehbrücke, Nuthetal, Germany. anne-kathrin.illner @ dife.de
Ann Nutr Metab. 2010;56(2):99-105. doi: 10.1159/000277667. Epub 2010 Jan 26.
Recent research has called into question the current practice to estimate individual usual food intake in large-scale studies. In such studies, usual food intake has been defined as diet over the past year. The aim of this review is to summarise the concepts of dietary assessment methods providing food intake data over this time period. A conceptualised framework is given to help researchers to understand the more recent developments to improve dietary assessment in large-scale prospective studies, and also to help to spot the gaps that need to be addressed in future methodological research. The conceptual framework illustrates the current options for the assessment of an individual's food consumption over 1 year. Ideally, a person's food intake on each day of this year should be assessed. Due to participants' burden, and organisational and financial constraints, however, the options are limited to directly requesting the long-term average (e.g. food frequency questionnaires), or selecting a few days with detailed food consumption measurements (e.g. 24-hour dietary recalls) or using snapshot techniques (e.g. barcode scanning of purchases). It seems necessary and important to further evaluate the performance of statistical modelling of the individual usual food intake from all available sources. Future dietary assessment might profit from the growing prominence of internet and telecommunication technologies to further enhance the available data on food consumption for each study participant. Research is crucial to investigate the performance of innovative assessment tools. However, the self-reported nature of the data itself will always lead to bias.
最近的研究对目前在大规模研究中估计个体日常食物摄入量的做法提出了质疑。在这些研究中，通常的食物摄入量被定义为过去一年的饮食。本综述的目的是总结在这段时间内提供食物摄入量数据的饮食评估方法的概念。给出了一个概念框架，以帮助研究人员了解在大规模前瞻性研究中改进饮食评估的最新进展，也有助于发现未来方法学研究中需要解决的差距。该概念框架说明了评估个体一年内食物消费的当前选择。理想情况下，应评估该年度中每个人的每日食物摄入量。但是，由于参与者的负担、组织和财务限制，选择仅限于直接请求长期平均值（例如食物频率问卷），或选择几天进行详细的食物消费测量（例如 24 小时饮食回忆）或使用快照技术（例如购买商品的条形码扫描）。似乎有必要和重要的是进一步评估从所有可用来源对个体日常食物摄入量进行统计建模的性能。未来的饮食评估可能会受益于互联网和电信技术的日益突出地位，以进一步增强每个研究参与者的食物消费可用数据。研究对于调查创新评估工具的性能至关重要。但是，数据本身的自我报告性质总会导致偏差。
