Objective: To examine whether a carbohydrate reduction program combined with telemonitoring improves glycosylated hemoglobin (HbA1c) in patients with poorly controlled type 2 diabetes. Materials and Methods: A randomized controlled trial with two-group pretest-posttest design and follow-up was conducted in the Primary Care Unit at Siriraj Hospital, Bangkok, Thailand. Fifty-three patients with poorly controlled type 2 diabetes aged from 35 to 59 years were randomly assigned to two groups, an experimental group (n = 26) that received both routine care and a dietary carbohydrate program applying telephone monitoring, and a comparison group (n = 27) that received only routine care. HbA1c was assessed at baseline and at 12 weeks. Dietary self-efficacy and consumption behavior were measured at baseline, 8 weeks, and 12 weeks. Results: According to the findings, the HbA1c levels in the experimental group at 12 weeks showed significant improvement compared to baseline and the comparison group (p<0.05). The mean difference in dietary self-efficacy and consumption behavior at 12-week were significantly higher than baseline and greater than those in the comparison group (p<0.05). Conclusion: The study program was effective in decreasing HbA1c levels because the program enhanced patients’ confidence in dietary control, which improved consumption behavior.