This clinical study was conducted to investigate the safety and effect of D-psicose on postprandial blood glucose levels in adult men and women, including borderline diabetes patients. A randomized double-blind placebo-controlled crossover experiment of single ingestion was conducted on 26 subjects who consumed zero or 5 g of D-psicose in tea with a standard meal. The blood glucose levels at fasting and 30, 60, 90, and 120 min after the meal were compared. The blood glucose level was significantly lower 30 and 60 min after the meal with D-psicose (p<0.01, p<0.05), and a significant decrease was also shown in the area under the curve (p<0.01). The results suggest that D-psicose had an effect to suppress the postprandial blood glucose elevation mainly in borderline diabetes cases. A randomized double-blind placebo-controlled parallel-group experiment of long-term ingestion was conducted on 17 normal subjects who took 5 g of D-psicose or D-glucose with meals three times a day for 12 continuous weeks. Neither any abnormal effects nor clinical problems caused by the continuous ingestion of D-psicose were found.