This study aimed to estimate the effects of a health mentoring program on fasting blood sugar, total cholesterol, triglyceride, physical activity, self care behavior and social support changes among community-dwelling vulnerable elderly individuals with diabetes.

A non-equivalent control group pre-post-test design was used. Participants were 70 community-dwelling vulnerable elderly individuals with diabetes. They were assigned to the experimental (n=30) or comparative (n=30) or control group (n=28). The experimental group participated in the health mentoring program, while the comparative group participated in health education program, the control group did not participate in any program. Data analyses involved a chi-square test, Fisher's exact test, a generalized linear model, and the Bonferroni correction, using SPSS 23.0.

Compared to the control group, the experimental and comparative groups showed a significant decrease in fasting blood sugar, total cholesterol, and triglyceride. Compared to the comparative and control groups, the experimental group showed significant improvement in self care behavior. However, there were no statistical differences in physical activity or social support among the three groups.

These findings indicate that the health mentoring program is an effective intervention for community-dwelling vulnerable elderly individuals with diabetes. This program can be used as an efficient strategy for diabetes self-management within this population.