Approximately 20% of patients with type two diabetes mellitus (T2DM) use insulin in Australia (1).  Many international studies identify gaps in core principles of patient knowledge and skills in insulin self- management (2-12).

Purpose:  To develop and validate a tool which can be used by nurses and general practitioners to assess T2DM patients' knowledge and skills in their self-management of their insulin and be completed within twenty minutes.

Method:  The 26 items in the tool were developed from the literature in five core domains and included a patient demonstration component. Face validation was undertaken by an expert review panel of endocrinologists and credentialed diabetes educators (CDEs). Item scoring was developed with expert psychometrician input. Readability and interpretability was assessed using a peer read-a-loud process and the items revised appropriately. The tool was piloted in regional New South Wales, Australia with 40 patients with T2DM on a variety of insulin regimens. The interviews were recorded and transcribed verbatim for qualitative analysis and completion of blinded scoring by a second independent CDE to assess inter-rater reliability.

Findings:   The burden of time for completion of the tool was within 20mins.  The majority of items demonstrated good correlation between raters and internal reliability with a Cronbach’s alpha  value of  0.808.   A Wilcoxon signed rank test was used to analyze the statistical significance between raters’ scoring. Four questions of the 25 approached significant difference and were deemed not to demonstrate reliability between raters. The remaining 21 questions were found to be non-significant, suggesting that statistical difference between rater 1 and rater 2 were not significant, (p > .200) two-tailed.  Spearman’s rho analysis was also conducted with scores ranging between .522-.871, suggesting that responses between raters were strongly correlated. Qualitative analysis in underway.

Conclusion:  The tool demonstrated promising psychometric properties and appeared capable of identifying patients requiring further education. The pilot has produced a tool which has the potential for high utility in general practice for use by nurses and general practitioners.  Further refinement of the tool and research trials in clinical practice are required. Final results will be prepared for dissemination in 2015.