An Application of Health Behaviour Models to Diabetic Treatment Adherence: A Comparison of Protection Motivation Theory and The Theory of Planned Behaviour

Abstract

Diabetes is a medical condition in which affected individuals have relatively low adherence to the treatment regimen consisting of medications, diet, exercise, and blood glucose monitoring. Models of health behaviour such as Protection Motivation Theory (PMT; Rogers, 1983) and the Theory of Planned Behaviour (TPB; Ajzen, 1991) have been developed in order to understand the many factors proposed to influence health behaviours but have not been sufficiently applied to diabetes. Both PMT and TPB predict intention to perform health behaviours but focus on different cognitions relating to intention. In addition to predicting intention and behaviour, the models may also be valuable in predicting emotional responses as the cognitions are similar to those specified in models of anxiety. The purpose of the present study was to compare PMT and TPB models in capacity to predict intentions to adhere and diabetic treatment adherence (i.e.,diet, exercise, medication, blood glucose monitoring). A second objective was to examine the influence of anxiety within the models. Volunteers with Type 1 or 2 diabetes from across Canada (N = 418) completed online or paper questionnaires. Follow-up questionnaires assessing treatment adherence were also completed one and two weeks after the initial questionnaire. Participants were highly adherent to medication and blood glucose monitoring activities (for individuals with Type 1 diabetes); therefore, the models could not be tested for these activities due to a lack of variance. A large percentage of variance was accounted for by PMT and TPB in each of the treatment behaviours. For both TPB and PMT, the strongest paths were found between intention and treatment behaviour. Perceived behavioural control/self-efficacy was the strongest predictor of intention in the majority of the models. Subjective norms in TPB and perceived severity in PMT were not found to be statistically significant in the models. Anxiety constructs were found to mediate relationships between certain cognitions and intention and between cognitions and treatment adherence. Anxiety symptoms were found to be negatively associated with treatment adherence. The results serve to improve our understanding of treatment adherence in individuals with diabetes, and also inform interventions for individuals who have difficulties with treatment adherence.