Determinants of exercise behavior of veterans with diabetes: Does the theory of planned behavior apply

Exercise for treatment of diabetes can contribute to achieving glycemic control, thereby reducing complications and costs of the disease. Although persons with diabetes generally receive advice from their healthcare providers to exercise, most do not exercise. This study explores health behavior theory to learn why some individuals with diabetes exercise and others do not. The aim of this study was to propose and test a causal model that explains the relationships among predictors of exercise behavior in veterans with diabetes. A telephone survey assessed exercise behavior and its proposed determinants in one hundred-one veterans who attend the VA Greater Los Angeles Healthcare System for routine diabetes care and then re-assessed their exercise level three weeks later. Structural equation modeling was used to evaluate the fit of the data to a model based on the Theory of Planned Behavior. The data provided a significant fit to the overall model. However, the fit was improved by adding a pathway to future exercise behavior from patients' perceived physician belief that they should increase their exercise. Hierarchical models were used to test the importance of the individual determinants of exercise behavior. The only path not supported in this analysis was the indirect effect of perceived behavioral control on exercise behavior through behavioral intention. This supports the notion that perceived behavioral control is a direct determinant of a behavior when skill and self-efficacy are important motivators. Patients who reported receiving physician advice to exercise during the clinic visit reported more exercise at follow-up, regardless of whether they received encouragement alone or encouragement with specific exercise instruction.