Knight & McIntyre et al., from the Queensland Diabetes and Endocrine Center in Brisbane, Australia have published a study “Qualitative Assessment of User Experiences of a Novel Smart Phone Application Designed to Support Flexible Intensive Insulin Therapy in Type 1 Diabetes” in the journal of BMC Medical Informatics and Decision Making. The article outlined specific features that users felt were useful for a mobile app used to manage insulin therapy for type 1 diabetes.
Treatment for type 1 and 2 diabetes consists of a wide spectrum of medications to manage blood glucose. Insulin injections are one of the more complicated medications used to manage diabetes. Patients need to remember when to administer their insulin, as well as differentiate how long each type of insulin stays active. Most people require both a long and short-acting insulin to achieve optimal control. Some regimens require patients to tailor their carbohydrate intake with their insulin dosing, which requires the onerous task of carbohydrate counting and dose calculations. Insulin underdosing leads to inadequate control of blood glucose, while overdosing can lead to a low glucose state. Both of these can easily lead towards life threatening conditions. There are many potential areas of innovation being explored by the mobile health community. More software and mobile applications can now wirelessly sync glucose data with from meters, insulins pumps, and continuous glucose monitors (CGMs). Each iteration brings more features for patients and providers. This study gives insight into what features are important in a smart phone-based app for diabetes 1 management.
This paper specifically studied multiple dose insulin (MDI) injection therapy. This regimen allows for greater flexibility but also requires that patients perform calculations to determine the appropriate dose of their insulin based on carbohydrate intake. 7 subjects were chosen to give their opinions during an educator led focus group after they used RapidCalc to record and manage their blood glucose for one month. Participants in the study were already experienced in using a flexible MDI regimen. The majority frequently overrode dosing recommendations due to their own personal sense that RapidCalc dose was off, especially when trying to account for exercise. All participants valued features of efficient dose calculations and the ability to quickly record and present data in an organized way. The majority also initially mistrusted recommended doses, which quickly dissipated as they grew more familiar with the application. Eventually participants found that RapidCalc was accurate and helpful for the majority of routine insulin dosing. Overall the participants felt RapidCalc was useful but wanted more features to streamline data collection, expand ways to catalogue food for more efficient carb counting, and the ability to pull data from glucometers.
It would be interesting to see more studies that focus on what determines the level of trust that patients have in health technology. The subjects from this paper found that having RapidCalcs show how it calculated its recommended dose allowed for a higher degree of trust. As RapidClac and software play a larger role in using collected data to guide medical management, knowing what characteristics bolster the trust of users will be important. Further exploring this concept can help designers achieve more buy-in for new features that further automate tasks.
This paper also brings up the important concept of user engagement in medical treatment. The management of diabetes relies heavily on data. Engagement of patients and MDs are required on both ends to maximize outcomes. It is still standard for patients to keep paper based journals of glucose levels and caloric intake. Maintaining this journal keeps patients engaged, but also requires a fair bit of monotonous work. Innovations have now achieved the ability to wirelessly collect glucose data from devices and present them in a way that allows for remote monitoring by physicians. These innovations have reduced the burdens of glucose monitoring, but also carry with it the potential for less engagement by patients. Overall, these innovations have been shown to improve management of diabetes, but use of these products requires a high level of health literacy. As health technology streamlines medical management, it is important that more studies guide how innovators implement features.