Self-monitoring of blood glucose (SBGM) is widely recommended for both type 1 and type 2 diabetic patients despite the lack of evidence of benefit in glucose control or as an aid in weight loss in type 2 subjects. This study tested the hypothesis that combined use of SMBG and dietary carbohydrate (CHO) counting, using the blood monitoring results to shape dietary CHO quotas, is beneficial in managing type 2 diabetes. Twenty-three overweight (body mass index, BMI 27.5-44 kg/m2) patients aged 40-75 participated in a 28-week behavioral weight control program. Baseline hemoglobin HbA1c ranged between 9.5% and 13.5% (normal range 5.5%-7.7%). Subjects were matched for weight, sex, and HbA1c and assigned to small (4-8 participants) groups which met weekly for 12 weeks and then monthly for 16 weeks. After 8 weeks, the groups were randomized either to continue the behavioral program or to have SMBG and dietary CHO counting. Glucose monitoring was performed 6 times daily (pre- and 2 h postprandially) for the first month, focusing on the meal increment and correlating this to dietary CHO intake. Weight loss was identical in both groups during the year of follow-up. The HbA1c level showed a progressive decline in experimental subjects (P < 0.05), whereas there was no improvement in control subjects.