Background: Nowadays, diabetes is one of the most common health problems worldwide that may cause severe side effects on the circulatory system, nervous system, kidneys, eyes, and feet. Health care education makes patients involved in caring and it is one of the main measures to attenuate the load of the disease on health system. The present study is an attempt to survey the effect of health belief-based educational intervention on improvement of metabolic indices in patients with diabetes type II in rural areas of Kermanshah-Iran. Methods: The study was carried out as a clinical trial in 2018 on 48 individuals grouped into intervention and control groups (each with 24 members) randomely. The participants were patients with diabetes type II visiting rural comprehensive services centers of Pave-Kermanshah-Iran. They were selected randomly so that visitors to Shamshir Village clinic of were selected as the intervention group and the patients in Serias and Darebaian villages’ centers were selected as the control group. Data gathering was done using demographics questionnaire and a metabolic indices checklist before and three months after the educational intervention. The participants in the intervention group were grouped into two groups of 12-15 members and the intervention was conducted based on health belief model in six sessions each for 60 minutes. Data analyses were done using SPSS (v.24) and descriptive statistics and Squared Chi test, Wilcoxon test, Mann Whitney test, independent t-test, and paired t-test. Findings: Before the intervention, there was no significant difference between the two groups in terms of demographics and diabetes metabolic indices (P>0.05). However, after the intervention, there was a significant difference between the two groups in terms of glycated hemoglobin (p=0.038) and fasting blood glucose level (P=0.006). The difference between the two groups was not significant in terms of BMI (p=0.301), cholesterol level (p=0.797), triglyceride (p=0.439), lipoprotein with low density (p=0.157), and lipoprotein with high density (p=0.664). Conclusion: The results showed that health belief-based educational intervention was effective in decreasing blood glucose level in diabetic patients. It is recommended using this approach as a part of therapeutic intervention and disease control in diabetic patients.