A household food inventory is not a good measure of fruit and vegetable intake among ethnically diverse rural women

Environmental measures of food availability are surrogates of consumption. Such measures may be useful among populations for whom standard dietary assessment is difficult. The objective of this cross-sectional study was to test whether a measure of the household dietary environment would perform as well as or better than a standard fruit and vegetable assessment among ethnically diverse rural women. Participants were 154 non-Hispanic white, 157 Hispanic, and 102 Native American adult women residing in rural Washington state. Participants completed an interviewer-administered household inventory of fruits and vegetables and a standard measure of fruit and vegetable intake used in the 5 A Day for Better Health Program. Pearson correlation coefficients assessed the validity of the measures against biomarkers of fruit and vegetable consumption (serum carotenoids). Pearson correlations were poor to between the household inventory and serum carotenoids (r = 0.06 to 0.22) and between the 5 A Day responses and serum carotenoids (r = -0.08 to 0.17). There were no differences by ethnic group; both short tools performed poorly compared with the biomarkers across Hispanic, non-Hispanic white, and Native-American participants. In conclusion, both the household inventory and the popular 5 A Day measure were poor indicators of fruit and vegetable intake in this sample of ethnically diverse rural women.