The DQQ consists of a general introduction followed by 29 yes/no questions along the lines of ‘Did you eat such-and-such yesterday?’ The questionnaire administrator has to read out loud exactly what the form says and is not allowed to ask follow-up questions. Photo Shutterstock If you use a standardized yes/no questionnaire to ask a large, random group of people what they ate the day before, that gives a perfectly good picture of the quality of the diet in the population to which the respondents belong. This conclusion comes from Betül Uyar (Global Nutrition), who is due to obtain her PhD at the end of September for her research on the Diet Quality Questionnaire.
We often don’t know much about people’s diets in low-income and middle-income countries. Advanced nutrition research methods — such as food diaries that require you to have scales or a 24-hour recall survey where a dietician asks you what you ate in the past 24 hours — are too expensive and time-consuming, so a more straightforward method was needed. That new method is the Diet Quality Questionnaire (see inset), developed in an initiative by the Global Diet Quality Project.
Yes or no
The Diet Quality Questionnaire is designed so that anyone can ask the questions, explains Uyar. ‘It starts with a general introduction, followed by 29 yes/no questions along the lines of ‘Did you eat such-and-such yesterday?’ ‘Such-and-such’ refers to examples of a type of food, such as oranges, mandarins and grapefruits. So that question would be about the intake of citrus fruits. ‘The questionnaire administrator has to read out loud exactly what the form says and is not allowed to ask follow-up questions. That means the questionnaire can’t go wrong.’
Uyar’s PhD research is part of the Global Diet Quality Project, which has now developed DQQ questionnaires for at least 140 countries and for various subcategories such as women and children. The product groups — for example citrus fruits or refined grains — are the same in all the questionnaires, but the chosen examples differ per country and are tailored to reflect the local cuisine. ‘When asking about refined grains, in the Netherlands we ask whether they ate bread and in Indonesia whether they ate rice.’
Diet score
The researchers use the yes and no answers to calculate various dietary scores that give an impression of how healthy a population’s diet is, or that of a subgroup. For example, they calculate the proportion of the population who ate at least some vegetables, fruit or nuts the day before. That information can be used as the basis for targeted dietary interventions and outreach.
The Diet Quality Questionnaire gives a broad picture of what people eat, but it doesn’t tell you anything about the quantities. ‘That means we can’t calculate what people’s nutrient intake was. The method is specifically designed for populations, not individuals. If you ask a large number of people what they ate yesterday, you will get a good picture of the average diet.’
Overestimation and underestimation
Uyar compared the outcomes using the Diet Quality Questionnaire with the outcomes from the 24-hour recall method and the food diary plus scales method to assess whether the DQQ does what it promises, namely give an accurate measurement of what people eat. ‘We decided beforehand that we would be satisfied with anything within a margin of a 10 per cent overestimation or underestimation. The results were very consistent. Among adults, our questionnaire generally gives a slight overestimation of the healthiness of their diet, regardless of which method we used as a reference. We saw the opposite in adolescents: the DQQ gives a slight underestimate of the quality of their diet. Even so, this is a good first step in assessing the dietary quality in countries for which we have little or no nutritional data, or only outdated data.’
The Diet Quality Questionnaire
The Diet Quality Questionnaire is divided into three categories: plant-based foodstuffs, animal-based foodstuffs and food products you shouldn’t eat too much of, such as cakes, deep-fried products and soft drinks. These categories are subdivided into groups such as vegetables, dairy produce and sweet drinks. The vegetable group is broken down into orange vegetables with a lot of vitamin A, green leafy vegetables and so on.
The questionnaire for the United States contains the following questions, for example:
Yesterday, did you eat any of the following foods:
– Bread, rice, pasta, tortilla, or cereal? YES or NO
– Fresh corn, popcorn, oats, granola, brown rice, or quinoa? YES or NO
– Potato? YES or NO
– Beans, refried beans, peas, lentils, hummus, chickpeas, tofu, or lima beans? YES or NO
Yesterday, did you eat any of the following vegetables:
– Carrots, orange squash, pumpkin, sweet potato, or red bell pepper? YES or NO
– Broccoli, spinach, arugula, kale, collards, turnip greens, or mustard greens? YES or NO
– Lettuce, tomatoes, green beans, celery, green peppers, cabbage, or cucumber? YES or NO
– Zucchini, mushrooms, eggplant, cauliflower, okra, asparagus, or radish? YES or NO
Yesterday, did you eat any of the following fruits:
– Cantaloupe, mango, papaya, apricots, or dried apricots? YES or NO
– Orange, clementine, mandarin, tangerine, or grapefruit? YES or NO
– Banana, apple, watermelon, grapes, avocado, berries, or cherries? YES or NO
– Pineapple, pear, kiwi, plums, prunes, peaches, or nectarines? YES or NO
Yesterday, did you eat any of the following sweets:
– Cakes, cookies, brownies, donuts, pastries, or pie? YES or NO
