Output details

Analyte testing method and system

Coventry et al recognised the need for blood glucose monitors to log and analyse lifestyle data. Lifestyle data is any quantifiable information, which might affect or represent an individual’s physical condition. Examples of lifestyle data are eating, exercise, medication and illness. Coventry’s role was to design an interaction and ensure it was intuitive, thereby encouraging an individual to record information related to their lifestyle. Coventry designed this for use on the simple machines to flag lifestyle information next to a reading to help users understand where any problems of control arose from. This enables control interventions to be aimed at specific problems.

Understanding how lifestyle influences glucose control is crucial when designing interventions to improve control. Improving day-to-day control is vital to reduce the long-term effects of diabetes. Many people do not look at events surrounding a particular reading. Coventry researched this issue over a number of years, carrying out observations and reviews of blood glucose logs from patients, error analysis and evaluations to create an understanding of how to improve the information on the meter to help identify problem areas. Coventry designed a predictive flagging interface, which allowed separate presentation, of readings as fasting, before and after meals. This interface meant that there were no extra steps involved in creating the “flag” which could then be used to look at separate groups of readings to understand where any blood glucose control issues may arise from. This was the first predictive flagging system to be approved by the FDA. It is difficult to carry out evaluations of medical products with users as they must be deemed “safe” before they can be used in real settings. Coventry created a process involving multiple methods to get to the root cause of problems. The research and usability studies are not published as they are company confidential. They involved a user centred process, including observation, prototype testing with potential users, interviews and a human factors failure mode and error analysis.