Relative validity of the Geisinger Rural Aging Study food frequency questionnaire.

Abstract

To determine the relative validity of a population specific food frequency questionnaire (FFQ) and evaluate the effectiveness of the instrument for assessing nutritional risk in older adults. A cross-over design with participants completing two different dietary assessment instruments in random order. The Geisinger Rural Aging Study (GRAS), a longitudinal study of over 20,000 adults living in the central, northern and eastern counties of Pennsylvania. A subset of GRAS consisting of 245 older adults (60% women) ranging in age from 70 to 95 years. Energy and nutrient intakes were assessed from two instruments: a population specific food frequency questionnaire (FFQ) and four 24-hour dietary recalls conducted over a two week period. Pearson correlation coefficients between the FFQ and dietary recalls for most nutrients were 0.5 or higher which suggests that the FFQ provided relatively valid estimates of macro and micronutrient intakes examined. Bland-Altman plots were generated to examine the agreement between instruments. Data are shown for energy, folate and zinc with close agreement at lower intakes indicative of risk for folate and zinc. Sensitivity results also showed that the FFQ was able to correctly classify individuals adequately at risk for most nutrients examined. This population specific FFQ appears to be a valid instrument for use in in evaluating risk for many nutrients that are of particular concern in older adults residing throughout many predominately rural counties in Pennsylvania.