OBJECTIVES:Food frequency questionnaires (FFQ) are frequently used in epidemiologic studies of nutrition and food intake. However, the use of FFQs in patients receiving maintenance dialysis has not been extensively studied. We hypothesize that FFQ is a useful tool to assess the food intake differences between patients receiving dialysis and patients not receiving dialysis. DESIGN:Matched exposed-unexposed study with case-controlled design. SETTING:Outpatient dialysis unit affiliated with a tertiary-care community medical center. PATIENTS:From a pool of 102 maintenance hemodialysis (MHD) outpatients in a community dialysis unit, 30 adult MHD outpatients (15 men, 15 women, aged 55.8 +/- 14.6 years) were selected randomly as case subjects. They included 16 African Americans, 8 whites, 4 Hispanics, and 2 Asians. Eleven MHD patients took the multivitamin, Nephrovite (R&D Laboratories, Marina del Rey, CA), regularly. From an archive of 1,610 nondialytic individuals with known FFQ data, 30 control subjects were selected randomly to match the age, race, and sex of the case subjects. INTERVENTION:We used Block's FFQ (version 98), an 8-page self-administered questionnaire that has been widely used in epidemiologic studies. A group of trained research assistants supervised the FFQ administration and interviewed those patients who were not able to answer all of the questions without assistance. Student t test was used to compare group means in form of daily dietary intake, and conditional logistic regression was used to calculate odds ratios for predetermined dichotomizing cutoff levels. MAIN OUTCOME MEASURES:Food intake characteristics of MHD patients as compared with control patients not receiving dialysis. RESULTS:Statistically significant differences between MHD case subjects and nondialytic control subjects were observed between the amounts of daily intake for vitamin C (84 +/- 63 mg/d v 127 +/- 70 mg/d, P = .01), dietary fiber (12 +/- 6 g/d v 18 +/- 11 g/d, P = .02), potassium (2,024 +/- 1,088 mg/d v 2,701 +/- 1,429 mg/d, P = .04), cryptoxanthin (56 +/- 88 microg/d v 140 +/- 118 microg/d, P = .003), and lycopene (2,052 +/- 2,234 microg/d v 4,524 +/- 3,979 microg/d, P = .004). These data indicate that MHD patients had a significantly lower intake of vitamin C, dietary fibers, potassium, and 2 of the carotenoid compounds when compared with individuals not receiving dialysis. Moreover, the daily intake of vitamin B(6) was significantly higher in MHD patients probably because of the high pyridoxine content in Nephrovite. By using the conditional logistic regression analysis, the odds ratios for lower than predetermined cutoff levels in patients receiving dialysis were significant for vitamin C, potassium, and the 2 previously mentioned carotenoids (odds ratio between 3.50 and 7.50, P < .05). CONCLUSIONS:Patients receiving dialysis may consume significantly lower amounts of potassium, vitamin C, and dietary fibers as well as lower amounts of some carotenoids. The FFQ seems to be a useful tool to compare dietary intake of MHD patients with other groups, although it may underestimate the amount of daily protein and energy intake and, hence, may not be an accurate tool for individual assessment of food intake. More studies are required to evaluate the validity of the FFQ in dialysis patients. The lower vitamin C, fiber, and carotenoid intake of MHD patients may be atherogenic. Hence, the hypothesis is proposed that prescribed restrictions in potassium in MHD patients may lead to reduced fruit and vegetable intake, leaving meat and fats as the main source of calories. This may contribute to atherosclerosis and increased cardiovascular morbidity and mortality in these patients. This hypothesis needs to be evaluated in future studies.