Estimation of fruit and vegetable intake using a two-item dietary questionnaire: a potential tool for primary health care workers.

BACKGROUND AND AIM: High fruit and vegetable intake is associated with health benefits for cancer and cardiovascular disease. An increase is therefore integral in recommendations for the prevention of chronic disease. However, measuring intake requires either extensive dietary assessment or the measurement of specific bio-markers which is neither cheap nor feasible for the routine assessment of an individual's diet in a community or primary care setting. Within the context of a study evaluating a dietary counselling programme to increase fruit and vegetable intake our aim was to assess the use of a simple tool to estimate fruit and vegetable intake. METHODS AND
RESULTS: We studied associations between bio-markers [plasma ascorbic acid, beta-carotene and alpha-tocopherol 24-hour urinary potassium excretion] and a two-item fruit and vegetable consumption questionnaire in 271 subjects (105 men and 166 women), aged 18 to 70 years. After controlling for age, sex, vitamin supplement use, smoking and body mass, those reporting a daily intake of > or = 5 portions of fruit and vegetables had higher potassium excretion (difference 15.6 [95% confidence interval: 6.2 to 25.0] mmol/24 h), urinary potassium/creatinine ratio (1.2 [0.5 to 2.0]) and plasma vitamin C (10.0 [-0.9 to 20.8] mumol/L) than those reporting < or = 2.5 portions per day. beta-carotene (p = 0.04), vitamin C (p = 0.01) and potassium excretion (p < 0.001) were associated with fruit rather than vegetable intake. The two-item questionnaire had high specificity; over 3/4 of participants who reported low intake also had bio-markers below the upper third of the distribution.
CONCLUSION: Self report of fruit and vegetable intake through a simple questionnaire is confirmed by bio-markers for those eating less than five portions of fruit and vegetables a day. Although the tool is amenable to improvements for the detection of vegetable portions, it may prove useful for monitoring dietary preventive approaches in primary care without the use of invasive and costly biochemical measurements.