BACKGROUND: Achieving and maintaining controlled glycemic levels are challenging in people with insulin-treated diabetes mellitus, suboptimal insulin injections being the main obstacle to treatment success. METHODS: This research article presents the main functionalities and performance tests on Insulclock®, an electronic device plugged into insulin pens and connected with a smartphone app to improve insulin management. Insulclock tracks the date, time of day, dose, type of insulin, temperature, and duration of insulin injections. Information is stored in the device memory and is available for analysis by patients and health care providers. Insulclock device also has a reminder system with visual and acoustic alerts to reduce insulin omissions and mistiming. RESULTS: The main performance tests reveal that Insulclock can detect seven types of insulin pens with a 97% correct classification rate. Among 556 injections, most of the doses were accurately detected (deviation = 0), with relative errors ranging from 3% to 7% across different dosages. The duration of injections recorded by this device strongly correlated with that detected by an external chronometer (R2 = 0.99, root-mean-square deviation [RMSD] = 0.39). Moreover, the precision of the temperature sensor was evidenced by high correlation of temperatures detected by Insulclock and by an external thermometer (Pearson's R2 = 0.90, RMSD = 0.45). CONCLUSIONS: Insulclock is a novel device capable of tracking dosing, timing, and missing insulin injections. The promising capabilities it offers for diabetes mellitus self-management may help health care providers, researchers, and insulin users avoid frequent errors in insulin administration.
BACKGROUND: Achieving and maintaining controlled glycemic levels are challenging in people with insulin-treated diabetes mellitus, suboptimal insulin injections being the main obstacle to treatment success. METHODS: This research article presents the main functionalities and performance tests on Insulclock®, an electronic device plugged into insulin pens and connected with a smartphone app to improve insulin management. Insulclock tracks the date, time of day, dose, type of insulin, temperature, and duration of insulin injections. Information is stored in the device memory and is available for analysis by patients and health care providers. Insulclock device also has a reminder system with visual and acoustic alerts to reduce insulin omissions and mistiming. RESULTS: The main performance tests reveal that Insulclock can detect seven types of insulin pens with a 97% correct classification rate. Among 556 injections, most of the doses were accurately detected (deviation = 0), with relative errors ranging from 3% to 7% across different dosages. The duration of injections recorded by this device strongly correlated with that detected by an external chronometer (R2 = 0.99, root-mean-square deviation [RMSD] = 0.39). Moreover, the precision of the temperature sensor was evidenced by high correlation of temperatures detected by Insulclock and by an external thermometer (Pearson's R2 = 0.90, RMSD = 0.45). CONCLUSIONS: Insulclock is a novel device capable of tracking dosing, timing, and missing insulin injections. The promising capabilities it offers for diabetes mellitus self-management may help health care providers, researchers, and insulin users avoid frequent errors in insulin administration.
Authors: Lutz Heinemann; Oliver Schnell; Bernhard Gehr; Nanette C Schloot; Sven W Görgens; Christoph Görgen Journal: J Diabetes Sci Technol Date: 2021-01-11