Low ascorbate concentrations in diabetes may be secondary to inadequate dietary vitamin C intake or may relate to the varied metabolic roles of the vitamin. To determine whether inadequate dietary intake is a factor we calculated daily vitamin C intakes using both a vitamin C questionnaire and a 4-day food diary in a group of 30 patients with Type 2 diabetes (mean age 68.8 +/- 6.9 yr, 17M/13F) and in 30 community controls (mean age 68.0 +/- 5.5 yr, 12M/18F)). Measures of plasma glucose, serum fructosamine, and plasma ascorbic and dehydroascorbic acid were obtained from 20 subjects in each group. There was no significant difference in daily vitamin C intake between the two groups using both methods: food diary, 61.4 +/- 28.3 (patients) vs 69.5 +/- 33.4 (controls) mg; questionnaire, 54.0 +/- 28.9 (patients) vs 65.0 +/- 30.9 (controls) mg. Vitamin C intake derived from both methods was significantly correlated (p < 0.001). Plasma ascorbate (30.4 +/- 19.1 mumol l-1) and dehydroascorbate (27.6 +/- 6.4 mumol l-1) levels were significantly lower in patients vs in controls (68.8 +/- 36.0 and 31.8 +/- 4.8 mumol l-1, respectively), p < 0.0001 and p < 0.01. Plasma ascorbate levels were significantly correlated with vitamin C intake derived from the food diary (p < 0.01) and questionnaire (p < 0.01) methods in the diabetic group only. Low ascorbate levels in diabetes appears to be a consequence of the disease itself and not due to inadequate dietary intake of vitamin C. A short vitamin C questionnaire is a convenient and reliable estimate of vitamin C intake.(ABSTRACT TRUNCATED AT 250 WORDS)