BACKGROUND: Although therapeutic diets are critical to diabetes management, their acceptability to patients is largely unstudied. OBJECTIVE: To quantify adherence and acceptability for two types of diets for diabetes. DESIGN: Controlled trial conducted between 2004 and 2006. SUBJECTS/SETTING: Individuals with type 2 diabetes (n=99) at a community-based research facility. Participants were randomly assigned to a diet following 2003 American Diabetes Association guidelines or a low-fat, vegan diet for 74 weeks. MAIN OUTCOME MEASURES: Attrition, adherence, dietary behavior, diet acceptability, and cravings. STATISTICAL ANALYSES: For nutrient intake and questionnaire scores, t tests determined between-group differences. For diet-acceptability measures, the related samples Wilcoxon sum rank test assessed within-group changes; the independent samples Mann-Whitney U test compared the diet groups. Changes in reported symptoms among the groups was compared using chi(2) for independent samples. RESULTS: All participants completed the initial 22 weeks; 90% (45/50) of American Diabetes Association guidelines diet group and 86% (42/49) of the vegan diet group participants completed 74 weeks. Fat and cholesterol intake fell more and carbohydrate and fiber intake increased more in the vegan group. At 22 weeks, group-specific diet adherence criteria were met by 44% (22/50) of members of the American Diabetes Association diet group and 67% (33/49) of vegan-group participants (P=0.019); the American Diabetes Association guidelines diet group reported a greater increase in dietary restraint; this difference was not significant at 74 weeks. Both groups reported reduced hunger and reduced disinhibition. Questionnaire responses rated both diets as satisfactory, with no significant differences between groups, except for ease of preparation, for which the 22-week ratings marginally favored the American Diabetes Association guideline group. Cravings for fatty foods diminished more in the vegan group at 22 weeks, with no significant difference at 74 weeks. CONCLUSIONS: Despite its greater influence on macronutrient intake, a low-fat, vegan diet has an acceptability similar to that of a more conventional diabetes diet. Acceptability appears to be no barrier to its use in medical nutrition therapy.