The interface between human behaviour and technology in youth with Type 1 diabetes mellitus

Utilisation of technology in the management of youth with type 1 diabetes has increased substantially over the past decade, however glycaemic results remain suboptimal for a significant proportion of users. Continuous subcutaneous insulin infusion (CSII) and continuous glucose monitoring systems (CGMS) represent cost-effective modalities, contingent upon achievement of target glycaemic control. Controversy exists surrounding who to prioritise for these therapies, given current resource-limited healthcare systems. As metabolic control has been shown to correlate with amount of interaction at the user-technology interface, the premise of this work was the exploration of this human factor and the possibility of using personal characteristics to predict successful engagement with technology. The first study in this work examined the impact on self-care parameters of CSII-automated blood glucose level (BGL) delivery versus the standard manual-entry method. This randomised crossover study displayed a mean increase of one BGL per day during the automated phase, but this did not translate to improvement in metabolic control, or other self-care behaviours. Thereafter, I conducted a systematic review of the literature in order to identify reproducible demographic, inter- and intrapersonal characteristics robustly associated with, or predictive of, self-care or metabolic control. This examination of seventy empiric studies resulted in thirteen factors which were then coalesced into a questionnaire-based tool. Finally, I conducted a clinical trial of this questionnaire in 97 youth with type 1 diabetes, 50 of whom subsequently commenced using 'real-time' CGMS and 47 of whom subsequently commenced CSII utilisation. The questionnaire resulted in a 92% accuracy in prediction of participant usage of CGMS and exhibited 95% accuracy in the CSII cohort. This work advances our knowledge regarding the human element of diabetes-related technology and culminated in a pilot study of the first successful tool shown to be predictive of technology usage in children and adolescents with type 1 diabetes