The role of knowledge, attitude, and self -efficacy in predicting glycemic control among persons with diabetes

This study employed a cross-sectional design (n = 147) to further investigate the roles that diabetes-related knowledge; patient satisfaction; and diabetes self-efficacy play in glycemic control among adult persons with diabetes. The independent variables included: diabetes-related knowledge, satisfaction and self-efficacy. These variables were measured using a paper and pencil instrument developed for the study during the patient\u27s initial visit to a diabetes self-management training program. Three self-efficacy sub-scales were derived from factor analysis: diet, barriers, and blood glucose monitoring/medications. The dependent variable used in the study to measure glycemic control was HbA1c which was measured during the initial visit. Knowledge about diabetes did not explain significant variance in HbA1c beyond that explained by socio-demographic variables in a regression model. Satisfaction with care did show a significant correlation (r = −0.25, p = 0.009) with glycemic control in persons using medications, suggesting that glycemic control improves as patient satisfaction increases. Satisfaction with care explained approximately 4% of variance in HbA1c beyond the socio-demographic variables in a regression model. Satisfaction with glycemic control was also significantly correlated with Hba1c in the total sample (r = −0.20, p = 0.02) and the sub-sample using medications (r = −0.29, p = 0.002). In regression models for the total sample and the medication sub-sample, satisfaction with glycemic control significantly added to the variance explained in glycemic control beyond the socio-demographic variables. Self-efficacy scores for the total scale and the sub-scales were not predictive of glycemic control. The diet self-efficacy sub-scale and the total self-efficacy scale both demonstrated that the proportion of subjects in good control with relatively high self-efficacy scores was significantly greater than those in poor control with high self-efficacy scores. In the total sample, a high overall self-efficacy score was a significant predictor of glycemic control (p = 0.0496) and explained 3% of the variance in glycemic control beyond the variance accounted for by socio-demographic variables. The results of logistic regression showed that the probability of having poor control was approximately 4.0 times greater for subjects with low self-efficacy scores. The results of this project suggest that diabetes-related self-efficacy and satisfaction are factors that are positively related to glycemic control