Assessing Dietary Intakes: Current Projects & Research
The Risk Factor Monitoring and Methods Branch (RFMMB) conducts extensive research on
the development, use, and evaluation of food frequency questionnaires (FFQs) and
statistical methods to assess usual dietary intake.
The design and accuracy of FFQs is an important issue because these instruments are
commonly used in epidemiological research to evaluate the relationship of diet to outcomes
such as cancer, heart disease, or other chronic diseases. FFQs are also used in dietary
interventions or in public health settings to monitor or evaluate absolute intakes. FFQs
consist of a list of foods about which individuals are asked to record usual frequency of
intake over a specified period of time (such as the past year). FFQs vary in length, but
often include more than 100 food items. Questions about usual portion size are often, but
not always, included. In addition to questions about food intake, FFQs often include
questions about vitamin, mineral, or other dietary supplement use.
Since the mid-1990s, we have worked to improve and validate FFQs for epidemiological
and public health research. We have developed an FFQ, called the Diet History Questionnaire (DHQ),
for use by researchers and others. Along with its associated software, Diet*Calc,
the DHQ can be used to estimate nutrient and food group intakes of
individuals. We have conducted several studies to evaluate the DHQ and other FFQs:
- The Eating at America's Table Study was a 1997-1998
validation study of the DHQ that compared the DHQ to two other FFQs in wide use at that time. EATS
used 24-hour dietary recalls as the reference instrument. We are currently analyzing EATS data in two
analyses: 1) a validation study of Food Guide Pyramid food group intake using four 24-hour dietary recalls
as the reference instrument, and 2) a study of the relationship of blood biomarkers to reported FFQ
intakes of carotenoids and tocopherols in a subgroup of participants.
- The Observing Protein and Energy Nutrition Study is a biomarker-based
study using the DHQ, 24-hour dietary recalls, doubly labeled water to measure energy
expenditure, and 24-hour urines to measure nitrogen, sodium, and potassium intakes. This
study was designed to assess the the structure of measurement error associated with FFQs
and 24-hour dietary recalls. Although several papers from OPEN are already published we
are continuing to analyze these data to assess 1) the impact of energy adjustment on
measurement error, 2) food intake reporting differences between low- energy reporters
compared to accurate reporters, and 3) reporting issues related to body mass index.
- The Re-OPEN Study is a study in which OPEN Study participants were
recontacted to further assess measurement error issues related to food records and a
combined food frequency questionnaire and a 7-day checklist. These data are currently
being collected and will be available for analysis in Fall 2004.
- The Women's Health Initiative Ancillary Study of the Relative Risk
Differences between FFQs and Food Records will assess whether relative risk findings
differ between two dietary assessment instruments, the FFQ and food records, conducted at
baseline among approximately 600 breast cancer cases and 1,200 controls.
We are also developing an improved method of assessing long-term average, or "usual"
dietary intake that builds on the strengths of both FFQs and 24-hour dietary recalls. As
part of this effort, we have developed the NHANES Food Frequency
Questionnaire (formerly called Food Propensity Questionnaire), which is similar to the
DHQ but without portion size questions. This FFQ was pilot tested for the National Health and Nutrition Examination Survey (NHANES) and
was introduced in the NHANES in 2003.