Choosing the optimal dietary assessment method

1. What is the purpose of the dietary assessment?

The most important step is to determine the exact purpose of the dietary assessment. Do you want to assess the intake of a specific food group, food or nutrient, or do you want to assess a complete dietary pattern? For example, if your aim is to assess the current dietary intake, a food record or one 24-hour dietary recall can be the best choice, but if your aim is to assess a habitual dietary pattern, multiple 24-hour dietary recalls or a food frequency questionnaire is better suited.

Also, think about how the information will be used: is it a primary outcome of your study, a secondary outcome, a compliance measure, or is it explorative?

2. What type of information do you want to assess?

Will the results be analyzed at an individual or group level? For example, if you want to check if a population has an adequate intake, a group mean intake and distribution of intake is needed. For epidemiological studies, you want to rank individuals or classified them according to their intake to determine the association with a health-related outcome. If you want to know whether a patient’s intake and excretion of a specific nutrient is in balance, the absolute individual intake of the patient is needed.

3. What is the reference period?

The reference period of interest depends on the aim of the study. Sometimes it is more suitable to determine the actual intake, while in other cases the habitual intake is needed. For example, to check compliance to an intervention you want to know the actual intake. But if you want to study the association between intake and a health related outcome, you may need the habitual intake over a longer time period.

4. What are the characteristics of the population?

The accuracy and applicability of the tools and methods can differ between age groups, between men and women, but also within groups. For example, for healthy elderly all methods are suitable, while for frail elderly with cognitive disorders, methods that rely on memory cannot be used.

Note: Our Compl-eat, FFQ-tool and Eetscore tool are only suitable for a Dutch population with a Dutch dietary pattern.

5. Do you want to compare the data with other or previous studies?

Data derived by different methods cannot be compared. Also, the outcomes of different FFQ’s are often not comparable. Therefore, the need for comparable results can be a reason to choose for a specific method.

6. How much resources are available?

It is most important to choose a method that is able to answer your research question. However, when more than one method is suitable the easiest and cheapest method can be chosen. For example, a web-based recall is less burdensome and cheaper for the researcher than a face-to-face recall by a trained interviewer.

These questions are also discussed and explained in a recent paper in the Dutch
journal for Dieticians (NTVD).

• Download the paper Voedingsnavraagmethoden (only Dutch)
• Download the overview paper of currently available methods