Background: The effectiveness of health education based on the Health Belief Model in diabetes mellitus patients can improve diabetes mellitus prevention behavior. This study aimed to determine the effect of the application of the Health Belief Model on health education in patients with diabetes mellitus. Subjects and Method: A systematic review and meta-analysis was performed using the PRISMA guidelines and the PICO model including Population= Type 2 diabetes patients; Intervention = Education based on the Health Belief Model; Comparison= Not using an educational method based on the Health Belief Model; Outcome= Perceived vulnerability, Perceived compliance, Perceived benefits, Perceived obstacles. Articles are collected from PubMed, Science Direct, and Google Schoolar. The keywords used “Health Belief Model” AND “DM” OR “Diabetes Mellitus” AND “Type-2”. A total of 8 articles met the inclusion criteria, namely primary full text paper, randomized controlled trial study design, based on the Health Belief Model, and outcomes in the form of perceptions of vulnerability, perceived compliance, perceived benefits, perceived obstacles for the meta-analysis were then assessed using RevMan 5.3. Results: Meta-analyses from Iran, India and Turkey showed that diabetes patients who were given education had an average perceived susceptibility score 1.37 units higher than those without education (SMD= 1.37; 95% CI= 0.74 to 2.01; p= 0.050). Diabetes patients who received education or education had an average severity perception score of 0.86 units higher than those without education (SMD= 0.86; 95% CI= 0.23 to 1.48; p= 0.007). Diabetes patients who received education on average had a perceived benefit score of 1.02 units higher than those who did not receive education (SMD= 1.02; 95% CI= 0.54 to 1.50; p < 0.001). Conclusion: Education based on the health belief model is effective for increasing perceptions of vulnerability, perceptions of adherence, and perceptions of benefits in diabetes patients with type 2 diabetes mellitus.