Background:Self-care is a valuable strategy to improve health and reduce events of hospitalization and the duration of hospital stay in elderly diabetic patients. This study aimed to examine the model of self-care behaviors in elderly diabetic patients.Materials and Methods:A survey was conducted among 209 diabetic elderly patients who were admitted in three hospitals affiliated with the Isfahan University of Medical Sciences, Iran. Convenience sampling method was used to recruit the participants. Depression, anxiety, stress, and perceived social support were considered as predicting exogenous variables and elderly patients' self-care activities were treated as endogenous variables. The data were collected by a four-part questionnaire consisting of demographic and health-related characteristics; 21-item depression anxiety stress scale, multidimensional scale of perceived social support, and Diabetes Self-care Activities scale. Structural equation modelling by Statistical Package for the Social Sciences (SPSS) version 16 and Analysis of Moment Structures-7 (AMOS) software was applied for data analysis.Results:Mean (standard deviation) of depression, anxiety, stress, perceived social support, and self-care activities of participants were 14.29 (4.3), 13.62 (3.74), 16.83 (4.23), 57.33 (14.19), and 44.56 (13.77), respectively. The results showed that the overall model fitted the data (χ2/df = 3.8, goodness-of-fit index (GFI) = 0.52, incremental fit index (IFI) = 0.48, and root mean square error of approximation (RMSEA) = 0.14). Three out of four variables (i.e., perceived social support, anxiety, and depression) significantly predicted adherence to self-care behaviors among diabetic elderly patients (p < 0.05).Conclusions:The perceived social support, anxiety, and depression were identified as key constructs which need to be taken into account and well managed by health care professionals to enhance adherence to self-care activities in diabetic elderly patients.