The effect of blood glucose selfmonitoring in patients with non-insulin treated type 2 diabetes

Significant debate exists about the utility of blood glucose self-monitoring (BGSM) in patients with non-insulin treated type 2 diabetes. Previous studies lacked any theoretically based intervention and produced conflicting results. The novel randomized controlled trial reported in this thesis used a theoretically based intervention to evaluate the effectiveness of BGSM as a strategy of glycaemic feedback to promote self management. Psychological theory was used to develop an intervention to support use of glycaemic feedback to promote self-management. Measures based on similar psychological theory were also developed to assess the effectiveness of the intervention in targeting the hypothesized mediators of behaviour. Participants were randomized to one of three intervention arms: regular HbA1c, BGSM with facilitator interpretation and BGSM with patient interpretation and application. There was no significant difference in HbA1c between the three groups after one year (p=0.653), although significant reductions were noted in all groups in patients with HbA1c over 8%. There were no significant differences in secondary clinical outcomes of change in weight (p=0.649) or systolic or diastolic blood pressure (p=0.890 and p=0.860 respectively). Significant changes in beliefs and behaviour were limited to change in illness coherence (p=0.041) and in eating high fat diets (p=0.009) and five portions of fruit and vegetables daily (p=0.017). Semi-structured qualitative interviews were used to explore patients' experiences and perceptions of BGSM. Participants who had difficulty in adhering to self-management recommendations found BGSM helpful, whereas those who had previously followed recommendations were discouraged by BGSM results if they did not reflect their efforts. BGSM with patient interpretation and application did not significantly improve HbA1c or other clinical parameters when compared to BGSM with facilitator feedback or regular HbA1c measurement. It also had little effect on illness beliefs and self-reported behaviour. It did, however, appear to be beneficial in patients who had difficulty in following self-management recommendations.