INVITED COMMENTARY - SOME LIMITATIONS OF SEMIQUANTITATIVE FOOD FREQUENCY QUESTIONNAIRES
INVITED COMMENTARY - SOME LIMITATIONS OF SEMIQUANTITATIVE FOOD FREQUENCY QUESTIONNAIRES
复制标题
DOI：
10.1093/oxfordjournals.aje.a116212
复制
发表时间：
1992-05-15
影响因子：
5
通讯作者：
SEMPOS, CT
中科院分区：
文献类型：
--
作者：
SEMPOS, CT
根据Willett（1，第87页），”食物频率问卷已成为流行病学研究中测量饮食摄入量的主要方法。“半定量食物频率问卷是目前使用的最常见的食物频率工具。Willett小组进行了一些最严格的研究，以评估半定量食物频率问卷方法的有效性，该小组一直是流行病学研究中使用这种方法的主要支持者之一。在本期杂志中，Rimm et al. (2)目前的调查问卷的有效性的扩展版本的结果，以前开发的Willett等人。（三）、Rimm et al. (2)提供了一个很好的机会来重新思考为什么食物频率问卷变得如此受欢迎，并且，鉴于它们的受欢迎程度，回顾它们的一些局限性，我相信，随着时间的推移，这些局限性已经变得模糊。Willett（1，第70页）指出：”食物频率方法的基本原则是，平均长期饮食，例如，数周、数月或数年的摄入量，是概念上重要的暴露量，而不是几天的摄入量。“因此，在许多研究中使用食物频率方法来估计一年内的日常摄入量，而不是使用每日食物记录或24小时回忆。食物频率仪器通常是预编码的形式，由几种选定的核心食物组成，这些核心食物被分成几个选定的关键营养素、添加剂、污染物或其他食物成分的营养成分相似的食物组。根据食物份量信息的收集方式，有三种不同类型的食物频率问卷。定量食物频率问卷中没有收集份量信息。对于半定量食物频率问卷，表格上注明了标准份量，或由研究者假设列出的每种食物的标准份量，并要求受访者报告标准份量的摄入频率。定量食物频率问卷要求被调查者提供关于食物份量的信息。有几个原因可以解释为什么半定量食物频率问卷是这种方法最流行的版本。首先，营养素可以从他们估计，但不能估计从定性的食物频率仪器没有插补部分大小。第二，半定量食物频率问卷可以在15-20分钟内自我管理，而定量食物频率问卷通常需要使用食物模型或食物图片进行长时间的访谈，以评估份量。第三，可以将自我填写的问卷邮寄给抽样人员，这使得可以对分布在广泛地理区域的抽样人员进行大型队列研究。第四，因为问卷是预编码的，所以它们可以相对快速、容易和便宜地进行光学扫描或处理。事实上，半定量食物频率问卷是预编码的，这导致了它们的许多局限性。预编码形式对营养领域来说并不新鲜。这个概念是
According to Willett (1, p. 87)," Food-frequency questionnaires have become the primary method for measuring dietary intake in epidemiologic studies." The semiquantitative food frequency questionnaire is by far the most common food frequency instrument presently in use. The Willett group has conducted some of the most rigorous studies to evaluate the validity of the semiquantitative food frequency questionnaire methodology, and the group has been one of the leading proponents of the use of this methodology in epidemiologic studies. In this issue of the Journal, Rimm et al.(2) present results on the validity of an expanded version of a questionnaire previously developed by Willett et al.(3). The paper by Rimm et al.(2) offers a good opportunity to reconsider why food frequency questionnaires have become so popular, and, in light of their popularity, to review some of their limitations which, I believe, have become blurred with the passage of time. Willett (1, p. 70) states that" The underlying principle of the food frequency approach is that average long-term diet, for example, intake over weeks, months, or years, is the conceptually important exposure rather than intake on a few specific days." Food frequency methodology is used, therefore, in many studies to estimate usual intake over, say, a year instead of using daily food records or 24-hour recalls. Food frequency instruments are usually precoded forms that consist of a few selected core foods which are grouped into food groups of similar nutrient composition for a few selected key nutrients, additives, contaminants, or other food constituents. Based on the way portion size information is collected, there are three distinct types of food frequency questionnaires. No portion size information is collected as part of qualitative food frequency questionnaires. For semiquantitative food frequency questionnaires, a standard portion size is stated on the form or assumed by the investigator for each food listed, and the respondent is asked to report frequency of intake in terms of the standard portion sizes. Quantitative food frequency questionnaires ask the respondent to supply information about portion size.There are several reasons why semiquantitative food frequency questionnaires are the most popular version of this methodology. First, nutrients can be estimated from them but cannot be estimated from qualitative food frequency instruments without imputing portion sizes. Second, semiquantitative food frequency questionnaires can be self-administered in 15-20 minutes while quantitative food frequency questionnaires usually require a lengthy interview using food models or pictures of food to assess portion size. Third, self-administered questionnaires can be mailed to sample persons, and this permits large cohort studies with sample persons spread over a wide geographic area. Fourth, because the questionnaires are precoded they can be optically scanned or processed relatively quickly, easily, and cheaply. The fact that semiquantitative food frequency questionnaires are precoded leads to many of their limitations. Precoded forms are not new to the nutrition field. The concept was