Do diabetes cognitions and poor emotional health predict adherence to diabetes self-care? A longitudinal test of the Common Sense Self-Regulation Model using Structural Equation Modelling

  • Joanna Hudson

    Student thesis: Unknown


    Depression and anxiety are prevalent in adults with diabetes and are associated with poorer adherence to diabetes self-care. The Common Sense Self-Regulation Model (CS-SRM) hypothesises that both illness cognitions and emotions determine adherence to health behaviours but little is known about the causal pathways through which poor emotional health impairs adherence. Using the CS-SRM as a theoretical framework this thesis aimed to identify the pathways that exist between diabetes cognitions and poor emotional health and explored their combined and independent effects on diabetes self-care.A systematic review with meta-analysis identified that all previous observational studies used cross-sectional designs to explore the relationships between diabetes cognitions and poor emotional health, excluding one. Randomised controlled trials showed that changes in diabetes cognitions were associated with changes in poor emotional health but these mechanisms of action were not confirmed by mediation analyses. Few observational studies have rigorously tested how diabetes cognitions and poor emotional health operate together to determine diabetes self-management behaviours. This thesis used a longitudinal design to test whether: i) diabetes cognitions and poor emotional health have a longitudinal bi-directional relationship and ii) diabetes cognitions and poor emotional health have a combined (mediated) and independent (direct) longitudinal effect on diabetes self-care. Outpatients with Type 2 diabetes (N=261) were recruited at baseline and completed self-report measures of poor emotional health (Well-being Questionnaire), cognitions (Revised Illness Perception Questionnaire; Beliefs about Medicines Questionnaire) and diabetes self-care (Summary of Diabetes Self-Care Activities Scale) at baseline and six months follow-up. Hypothesised pathways were tested simultaneously using structural equation modelling.Participants who were more anxious at baseline perceived diabetes to be unpredictable and were apprehensive about their medications at follow-up. These effects were not observed for depression. Baseline diabetes cognitions did not predict change in poor emotional health at follow-up. Equally, neither baseline depression or anxiety predicted change in diabetes self-care behaviours at six months (directly or indirectly via diabetes cognitions). Baseline personal control beliefs remained independent of poor emotional health: greater personal control beliefs were associated with reduced adherence to diabetes self-care.A bi-directional relationship between diabetes cognitions and poor emotional health was absent in this sample. The main direction of effect was from anxiety only to diabetes cognitions. Depression and anxiety had no relationship with diabetes self-care. Whilst personal control remained independent of poor emotional health, mastery beliefs appear to be insufficient on their own to sustain adherence behaviours over time suggesting that interventions should also provide patients with action plans whilst managing outcome expectations. The unexpected findings for the relationship between poor emotional health, cognitions, and diabetes self-care may be because the sample did not include individuals with more severe depression or anxiety. This study needs to be replicated among people experiencing clinically significant levels of depression and anxiety in diabetes.
    Date of Award31 Dec 2013
    Original languageEnglish
    Awarding Institution
    • The University of Manchester
    SupervisorDavid Reeves (Supervisor), Edna Bundy (Supervisor), Alexander Wood (Supervisor) & Peter Coventry (Supervisor)


    • Self-care behaviours
    • Depression
    • Anxiety

    Cite this