Food Frequency Questionnaire (FFQ)
Food Frequency Questionnaire · Also known as: FFQ, food-frequency-assessment
The Food Frequency Questionnaire is a self-administered dietary assessment tool designed to measure habitual food and nutrient intake over an extended period (typically 6–12 months). Developed by epidemiologists, particularly Walter Willett at Harvard, the FFQ has become a cornerstone of nutritional epidemiology research, enabling large-scale studies to assess dietary patterns and examine diet-disease relationships. FFQs vary in length (50–200+ items) and focus, but all share the purpose of estimating average dietary intake in a time-efficient manner suitable for population studies.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
FFQs are essential for large cohort studies, national surveys, and epidemiological research examining diet-disease associations. Clinical applications include: detailed dietary assessment in patients with chronic diseases (diabetes, cardiovascular disease, renal disease, cancer) to identify patterns and guide medical nutrition therapy; assessment of dietary adherence to specific eating plans (Mediterranean, DASH, low-sodium, renal diet); and research examining relationships between nutrient intake and biomarkers or health outcomes. FFQs are less practical for rapid screening in busy clinical settings (use short dietary screeners instead) but are gold standard for assessing usual intake over extended periods in research contexts.
Strengths & limitations
- Practical for large-scale studies—captures 6–12 months of typical intake efficiently; low cost and burden enable inclusion of thousands of participants in cohort studies.
- Validated extensively—numerous FFQ versions have been compared against food diaries, biomarkers (e.g., 24-hour urine sodium for sodium intake), and doubly-labeled water for energy; validity correlations typically 0.4–0.7 (reasonable for epidemiology).
- Reflects habitual intake—asks about typical consumption over an extended period, reducing within-person variation captured by single 24-hour recalls.
- Enables nutrient assessment—detailed food list and portion information allow calculation of total nutrient intakes, enabling examination of relationships between specific nutrients and disease.
- Population-based ranking—excellent for dividing populations into dietary groups (e.g., high vs. low vegetable consumers) and comparing health outcomes across groups.
- Flexible and customizable—FFQ design can be tailored to specific populations (e.g., inclusion of ethnic-specific foods, removal of irrelevant items), research questions, and resource constraints.
- Recall bias—individuals may not accurately remember consumption over the past year; food preferences and current dieting may bias recall of typical intake.
- Portion size uncertainty—FFQ portion estimates are often crude; respondents may not accurately estimate 'medium' or 'large,' leading to misclassification.
- Food list limitations—FFQ includes finite food list; newly popular foods may be missing; aggregate categories (e.g., 'other vegetables') may mask important variations.
- Assumes static patterns—FFQ asks about past-year habits, but actual intake varies seasonally and over time; assumes recent diet is typical.
- Measurement error and attenuation—FFQ estimates contain substantial error; correlations with biomarkers are 0.4–0.7, meaning roughly 50% of variance is noise.
- Expensive detailed validation—validating a new FFQ against biomarkers or food records requires additional studies and cost.
- Limited portion-size detail—most FFQs include simple portion categories; detailed portion assessment (using scales, photographic atlases) is more burdensome.
- Socioeconomic and literacy barriers—completing FFQ requires literacy and cognitive ability; may be challenging for some populations (elderly, low education).
Frequently asked
How accurate is the Food Frequency Questionnaire compared to other dietary assessment methods?
FFQ accuracy relative to reference methods (food records, 24-hour recalls, biomarkers) varies. Validity correlations typically range 0.4–0.7 for energy and major nutrients; correlations are lower for micronutrients and food groups. This means FFQ rankings (high vs. low consumers) are reasonably valid for epidemiology, but absolute intake estimates may be off by 20–50%. For individual nutrition counseling, FFQ alone is insufficient; dietitian assessment and additional recalls are recommended.
What is the best FFQ to use for research?
Choice depends on research question, population, and resources. Validated FFQs for general populations include the Harvard FFQ (used in major cohorts) and NCI FFQ (US-focused). For specific populations (older adults, children, ethnic groups), population-specific FFQs with established validity are preferable. Consultation with a nutritional epidemiologist or dietitian is recommended to select the most appropriate instrument.
Can FFQ be used to assess dietary adherence in a patient with type 2 diabetes?
Yes, a diabetes-specific FFQ can assess carbohydrate quality, fiber, sodium, and other nutrients relevant to diabetes management. FFQ provides a comprehensive snapshot of typical intake, identifying patterns (e.g., high refined carbohydrate, low fiber) that guide counseling. However, FFQ does not measure meal timing, portion sizes, or glycemic control; integration with self-monitoring blood glucose and dietitian counseling is essential.
How often should FFQ be repeated in longitudinal studies?
FFQ is typically administered at baseline and then repeated every 2–4 years to capture changes in dietary patterns. More frequent assessment (annually) increases burden but improves temporal resolution. Less frequent assessment (every 5+ years) may miss important dietary transitions. Optimal interval depends on the expected rate of dietary change and study resources.
Does FFQ capture supplement use?
Most FFQs do not include detailed supplement questions. Some FFQs have optional supplement modules; others rely on separate supplement questionnaires. Because supplements can significantly contribute to nutrient intake (e.g., vitamin D, calcium), researchers should ask about supplement use separately and add estimated nutrient amounts from supplements to FFQ-derived nutrient totals.
Can FFQ be used in populations with low literacy or language barriers?
Standard FFQs require reading and writing ability; populations with low literacy or non-English speakers may struggle with self-administered FFQs. Trained interviewers can administer FFQs verbally in the respondent's language, though this increases cost. Pictures or photographs of foods can aid comprehension. Some FFQs have been translated and validated in specific languages; use of population-appropriate FFQ versions is preferable to administering untranslated tools.
Sources
- Willett, W. C. (1998). Nutritional Epidemiology (2nd ed.). Oxford University Press. DOI: 10.1093/acprof:oso/9780195122978.001.0001 ↗
- Subar, A. F., Freedman, M. R., Tooze, J. A., et al. (2015). Addressing current criticism regarding the value of food frequency questionnaires. The Journal of Nutrition, 145(12), 2639-2645. link ↗
How to cite this page
ScholarGate. (2026, June 3). Food Frequency Questionnaire. ScholarGate. https://scholargate.app/en/nutritional-science/food-frequency-questionnaire
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- DASESNutritional Science↔ compare
- DQI-INutritional Science↔ compare
- FNSNutritional Science↔ compare
- MEDASNutritional Science↔ compare
- MNANutritional Science↔ compare