Dietary Assessment Methods in GDD

Dietary Assessment Methods


About Diet Assessment Methods used in GDD

The Global Dietary Database is comprised of approximately 1,216 surveys from across the world. There are a variety of ways in which nutrition researchers collect dietary data. The image above indicates the different assessment methods used by scientists to gather dietary data and the breakdown of each type within the Global Dietary Database. Each survey collection method comes with a variety of strengths and weaknesses, outlined below. The pros and cons of these methods contribute to the unique nature of nutrition research but are often used to critique the field.

This makes a resource like GDD even more essential. Since GDD is comprised of all of these dietary assessment methods, the nutrition community can further validate these commonly-used tools against one another and build a foundation for a credible and reliable set of dietary data that spans across regions and populations.


Food Frequency Questionnaires (FFQ)

FFQ’s are a common tool for scientists to measure dietary intake on the individual level. This method generally requires the individual to recall what foods they have consumed over the past year and asks specific questions to understand the frequency and amount of consumption of certain foods. These questionnaires can have upwards of 150 food items listed.


  • Estimates usual intake over a period of time, generally the past year
  • It is a commonly used methodology that is relatively easy and quick to administer


  • Requires an individual to remember what they ate over the past year so it is subject to recall bias
  • Often underestimates what an individual truly consumes
  • The questionnaire is limited to certain foods listed, thus; infrequently consumed foods are not accounted for


Single/Multiple Recalls

Dietary recalls are another way scientists measure intake at the individual level. Recalls are generally conducted in-person or over the phone and require the participant to remember what they ate over the last 24 hours. A multiple recall follows the same protocol on multiple days to capture an individual's average diet rather than one particular day.


  • Participants only must recall recent memory rather than intake over the past year
  • It can capture unusual foods as well as specific foods at restaurants
  • The recall is administered by an interviewer, and therefore literacy is not a limitation 


  • Requires multiple assessments in order to gauge an average or usual diet
  • Requires administration from a trained professional and therefore can be resource-intensive
  • The open-ended nature of this assessment method can cause issues when categorizing foods for analysis


Demographic Health Surveys

We highlighted DHS surveys in the last edition of our newsletter (click here). DHS surveys are often referred to as diet diversity questionnaires because they collect data on general food consumption and intake rather than specific quantities.


  • Often collected in low-resource countries where nutrition data is limited
  • Conducted over multiple years so the repeated measurements allow examination of trends
  • These surveys target children and women, two groups that are of particular focus to GDD
  • Tend to encompass large groups and therefore have substantial sample sizes


  • Do not measure specific intake
  • Questions are often framed at the household level rather than the individual level


Biomarker Assays

Biomarker data uses biological specimens to indicate intake of certain foods. Examples of biomarkers include but are not limited to blood/serum, urine, stool, hair and nails, and adipose tissue. The biomarker data primarily included in the GDD come from urine samples measuring sodium consumption.


  • Do not require self-report and may, therefore, be more objectively accurate
  • If stored and labeled properly, the samples can be used for more than one analysis
  • Often does not require burden or discomfort when obtained from the participant


  • Unless careful, biomarkers can be influenced by time of day that the sample was collected and the fasted or non-fasted state of the individual
  • In order to actually get an estimate for nutrient intake for an individual, lab procedures using the specimen must be performed and the specimen must be properly stored, all of which incur additional costs
  • Cannot isolate food sources contributing to the particular biomarker being measured


Household Availability/Budget Surveys

These surveys are generally conducted to categorize aspects of household socioeconomic conditions. However, more recently, they are being used for food security and nutrition-related research.


  • These surveys tend to be nationally representative
  • They are often collected every few years and are therefore useful for looking at trends in consumption
  • They are useful for looking at questions surrounding food security and nutrition
  • They include other variables such as socioeconomic status and education


  • Food consumed away from the house is often not included
  • These data measure apparent consumption but not actual consumption
  • Does not allow for individual-level measurements, data is on the house-hold level