Aims This meta-analysis aims to update former meta-analyses from randomized controlled trials (RCT) focused on the efficacy of CBT for diabetes.

Methods Five databases were searched for RCTs. Primary outcomes were glycated hemoglobin (HbA1c), fasting blood glucose (FBS), systolic blood pressure (SBP), diastolic blood pressure (DBP), and body mass index (BMI). Secondary outcomes were depression, anxiety and distress symptoms, quality of life, sleep quality.

Results 32 RCTs were included. Results revealed that CBT could reduce HbA1c: -0.14% (95% CI: -0.25 to -0.02%, P = 0.020); FBS: -15.48 mg/dl (95% CI: -30.16 to -0.81 mg/dl, P = 0.040); DBP: -2.88 mmHg (95% CI: -4.08 to -1.69 mmHg, P < 0.001); depression symptoms: -0.90 (95% CI: -1.22 to -0.57, P < 0.001); anxiety symptoms: -0.28 (95% CI: -0.50 to -0.07, P = 0.009); improve sleep quality: -0.92 (95% CI: -1.77 to -0.07, P = 0.030). Subgroup analysis indicated that CBT has siginificantly reduced HbA1c when delivered as a group-based and face-to-face method, and psycho-education, behavioral, cognitive, goal-setting, homework assignment strategies were applied as central strategies.

Conclusion CBT was an effective treatment for diabetes patients, significantly reduced their HbA1c, FBS, DBP, depression and anxiety symptoms, and improved sleep quality.