Type 1 diabetes (T1D) afflicts more than half a million children aged 0-14 years worldwide [1]. Many individuals are diagnosed with T1D at a young age, which requires parents to take responsibility for managing their child’s chronic health condition [2]. Proper management includes monitoring multiple daily blood glucose measurements, physical activity, carbohydrate intake, and adjusting insulin doses with the use of multiple daily insulin injections or an insulin pump for basal and bolus insulin delivery [3,4]. Glycemic control is monitored quarterly at medical visits with the glycosylated hemoglobin (HbA1c) test. As children become teenagers, they steadily transition toward self-management, gaining more responsibility for their T1D care, and their parents gradually relinquishing control [5]. This transition period is often difficult because of the complexity of managing T1D in the context of suboptimal communication between the child and parent, often resulting in deleterious health outcomes [6]. Inadequate management during this transition increases the risk of severe hypoglycemia, which can lead to seizures or coma, and ketoacidosis, which can cause neurologic sequelae and even death [7,8]. The long-term consequences of poor diabetes management include cardiovascular disease, stroke, kidney disease, blindness, amputation, and premature death [7-9]. Successful transition is critical to the health and well-being of the teen. Identifying interventions to aid in the transition of care can ultimately increase positive health outcomes and the quality of life for adolescents with T1D [10].

As a result of the complexity of managing T1D, there is often a lack of adherence to proper T1D self-care, resulting in an unsuccessful transition [11]. Approximately 75% of teens are not achieving American Diabetes Association (ADA) HbA1c targets during this transition period [12]. Therefore, identifying and finding solutions to help adolescents attain the knowledge and skills needed to succeed in diabetes management is imperative to increasing positive health outcomes [13]. To ease the transition, interventions that aim to improve productive communication between parents and teens, help build trust and autonomy, and develop problem-solving skills which have demonstrated improved diabetes outcomes [14]. Unfortunately, parent-teen communication is difficult during this transition time, particularly when interactions regarding their diabetes management are often negative [15]. For instance, teens feel their parents are nagging them about their diabetes care [16]. Infusing positive communication into the parent-child relationship is fundamental to the success of this difficult transition [17]. Building trust and autonomy is also important to allow the parent to take a step back from care and give responsibility of disease management to the teen. Finally, problem-solving skills for both the parent and the teen can further improve communication and trust, ultimately improving diabetes outcomes.

Mobile health (mHealth), the use of mobile technology in health, has been used in many health contexts, including asthma, pain management, depression, pregnancy, and others for more than a decade and has demonstrated positive outcomes [18-26]. Despite the proliferation of these health apps (over 165,000 available in Google Marketplace and Apple Store), very few have been developed using a patient-centered design [27]. The use of mHealth in diabetes care has been reported as useful in improving trends, but interventions have not been tested over a period of time sufficient to determine long-term engagement [28,29]. A recent study has called for patients to be more engaged in the design and development process [30]. Other studies have explored the wants and needs of an app for people with diabetes. Researchers have found that ease of use, timely and accessible information, and good visual appearance of the technology are all important [31,32]. Building an mHealth intervention using a patient-centered process to increase communication, trust, and autonomy should facilitate the goals of a successful transition to independence and achieving positive health outcomes and quality of life.

Examining the literature regarding participatory design within T1D and apps, we were unable to find any research that reported including both the parents and the child in the design. Previous research describes working with clinicians and developers [33,34], adults with T1D [35,36], or only the adolescents with T1D (no discussion with the parents) [19]. However, parent-child communication is critical during the transition to self-care and should be of central concern when designing an app for this context. Thus, our study focuses on communication that occurs between the parent and their child surrounding the management of diabetes, as it relates to the development of our app. Our expectation is that an app focused on easing the difficulties of diabetes-specific communication will result in positive health outcomes and better quality of life for both adolescents and their parents [37]. Therefore, we sought to better understand the perspective of both groups and gather feedback regarding our prototype.

The objective of this study was to use patient-centered research methods to inform and improve the design and functionality of our type 1 diabetes app, MyT1DHero. Additionally, we hope to provide insight for others who are designing a health app for adolescents and parents. To inform development of the app, we conducted qualitative focus groups to further explore how teens with T1D and their parents communicate about diabetes, what motivates them to use different apps and games, and evaluate their perceptions of our proposed app.

MyT1DHero is a self-management mobile app that aims to encourage adolescents aged 10-15 years to self-manage their diabetes by recording their blood glucose (BG) values while allowing the parents seamless access to these same values. See Figure 1 for example screenshots of the app. The parent users have a separate login that allows them to access the app and review their child’s BG value history from their own phone. The parent and adolescent may also exchange messages via the app that are geared to facilitate positive communication. In addition, the app will have customizable test reminders, BG ranges, and an educational component that provides diabetes information and social support.

The adolescents will also earn points based on app usage. The accumulated points will allow for the adolescent to purchase “accessories” to customize their avatar, or a virtual character on the app.

Participants were recruited via the Juvenile Diabetes Research Foundation (JDRF) southeast Michigan’s family network, including Facebook, Twitter, and their email listserv. Inclusion criteria included: the adolescents must: (1) have a T1D diagnosis according to the ADA practice guidelines, (2) be 10-15 years old, (3) have had a diagnosis of T1D for at least six months, and (4) be fluent in English. The parent or guardian must: (1) have an adolescent with T1D who is 10-15 years old and who has had a diagnosis for at least 6 months and (2) be fluent in English. The study was approved by the Michigan State University Institutional Review Board. There were 5 adolescent participants in the focus group aged 10-13 years. The parent focus group had 7 participants, each with a child having a diagnosis of diabetes for 4-6 years. Three of the parents did not have an adolescent participating in the focus group.

Figure 2 summarizes the content of the focus group protocol. Parent and adolescent focus groups were conducted separately. The adolescent focus group comprised 5 individuals, which included 4 males with a mean age of 11.6 years old (range 10-13 years), and the adolescents had a diagnosis duration of 4-6 years. The parent focus group comprised 7 individuals and included 6 females. They ranged in age from 35 to 60 years, all white, two had some college, two had a bachelor’s degree, and two had master’s degrees. The average income based on the participants’ zip code was US $80,902.20 (range: US $46,649 to US $112,774). They all resided in southeast Michigan.

Each group consisted of two sections; in the first, participants were asked about their daily diabetes management routine, how their child’s diabetes is managed (ie, amount of child responsibility), and overall communication with each other, including how often they communicate, the tone or feel of the communication (eg, positive, happy, nagging), and the channel in which they communicate (eg, phone, text, face-to-face). Gaining the perspectives of both parent and adolescent is key to designing a patient-centered app that fits the context of disease management from the point of view of those directly affected.

Additionally, we asked about current apps that they used for both diabetes management and for general entertainment. For the second portion of the focus group, the participants were given a demonstration of our app prototype. We asked them general questions about the app, what they liked about it, what they would change, and what would motivate them to use it. The parent focus group lasted for 107 min and the adolescent focus group lasted for 72 min.

All of the data collected in the focus groups were audio-recorded, transcribed verbatim, and coded using NVivo version 11 (QSR International Pty Ltd) software. Using inductive thematic analysis, part of the team iteratively developed a set of themes that captured the focus group dialog (BH, DH, AH, KM, and MK) [38]. The codes were presented to the whole group for clarification and feedback. Then, two members of the team (KM and MK), coded a random selection of the transcripts to ensure reliability. Once reliability was obtained, each transcript was coded independently by the coders. They then came together to compare and evaluate the codes. Disagreements were resolved through discussion.

Four key themes were identified. These included (1) diabetes is unpredictable, (2) negative and frustrated communication, (3) motivations to use an app, and (4) feedback specific to our app. The following section describes each theme, along with descriptive quotes.

This part of the focus group asked participants about how they manage their diabetes and general communication around diabetes. We believe that this is key in understanding how our app could help them in their daily lives.

In this section of the focus group, we asked them about their current use of apps to help manage diabetes and what keeps them using technologies. We then conducted a demonstration of our app prototype and gathered feedback.

mHealth used for diabetes management is growing as a method to improve adherence and health outcomes. Using a patient-centered approach, this study sought to understand parent and adolescent perspectives of on-going development of a T1D app. To fully understand how an app could be engaging to both the parent and the child, it was important to understand the environment in which an app would be used. To achieve this goal, we asked parents and adolescents to tell us about how they currently manage the child’s diabetes and how they communicate about the disease. We then showed them a demonstration of our prototype app to obtain their perspectives on the app and additional components they thought the app should have.

These focus groups acknowledge the complexity of disease management and the frustrations in communication for families with an adolescent with T1D. Parents feel the need to remain in control because the day-to-day management of diabetes is very unpredictable. Consistent with other research, the unpredictability fuels some of the frustration that the parent and the adolescent feel and ultimately leads to inadequate communication strategies [23]. Our proposed app is designed to help improve the communication between the parent and their child. The app helps monitor the adolescent’s blood glucose and provides positive message prompts to the parent for better communication.

In order for an app to be successful and provide benefit to users, people must actively use it. We asked about different strategies that would keep the users engaged with the app. Customization, interactivity, and tangible rewards were discussed as motivating factors of using the app [37]. This is similar to other studies that have concluded that customization is important to appeal to the needs of various users and that it is imperative to discuss what options should be customizable [39-41]. Having the app be interactive was also stated as a motivation for engaging and using the app. Currently, there is limited interactivity in the downloads of their meter or their pump, which could be one cause of their low usage. Past research has demonstrated that having interactive functions have improved usability and resulted in higher engagement [42-45]. Through these focus groups, we have decided to integrate more interactivity in the app through some of the customization features, as well as through how the parents and child communicate through the app.

Participants also agreed that providing tangible rewards or financial incentives would motivate users to remain engaged with the app. Some research has shown that financial incentives are key motivators in weight loss and smoking cessation [46]. Ethical issues remain related to incentivizing individuals in this manner and what this practice might mean particularly for disadvantaged populations [47]. Additionally, there are issues of long-term sustainability both in terms of the health impacts and how the program would be able to continue to provide those incentives [48]. Because of these issues, we have decided to use some gamification techniques to reward use of the app. The utilization of gamification seeks to encourage users to complete specific tasks within a nongame environment [49-51]. There are several ways to do this, which include giving points every time a user opens the app, and having participants earn badges or points to “buy” items for avatars. Past research has suggested that having an environment that is enjoyable to use can increase health knowledge, efficacy, and adherence to their daily care can improve health outcomes [52-56].

A striking feature from the parent focus group was how parents reported communicating with their child about diabetes. In the current communication cycle, parents often report becoming frustrated and upset with their child, which does not help them learn how to handle the issue in a positive manner [57]. This may lead to the adolescent lying to their parents in hopes of preventing them from becoming frustrated or upset. This leads to further deterioration in the quality of the parent-child communication. In short, the parents talk at their adolescent, commanding them to pay attention to their care rather than working with them to successfully manage diabetes. Therefore, we believe adding problem-solving resources and features to the app would be beneficial. Problem solving, defined by Yeates and colleagues [58], is conceptualized as defining a problem, generating alternative strategies, evaluating outcomes, and selecting a strategy. Some literature has demonstrated positive associations between problem-solving behaviors and improved HbA1c; however more research is needed in this area [59].

An inherent feature of technology is its distance between users. Past research has suggested that “online interaction reduced embarrassment that might be experienced in face-to-face interactions and thus encourages self-disclosure” [60]. There are many studies demonstrating that computer mediated communication enhances self-disclosure [61-62]. This is especially important for adolescents who, developmentally, are often hyper self-aware and self-conscious [62]. As the proposed app will notify parents that their child has measured their blood glucose level, and in some cases report what the blood glucose level is, parents will be able to respond with affirmative preprogrammed messages through the app. We believe that the outcome of app use will change the dynamic of the communication. The adolescent will be more likely to provide true numbers and the parents will have preprogrammed messages with which they can respond, thereby making communication less frustrating and a more honest and constructive exchange.

One of the desirable app features that the adolescents mentioned was to join a team. A growing number of researchers are starting to focus on social interaction in mobile apps. Although fun is the primary reason many use apps, social interaction was reported as the second reason to play games [63]. Using technology in a group leads to greater enjoyment, lower dropout rate, and longer engagement compared with individual play [64]. It has been shown in health and sports studies that most people enjoy playing in a group more than by themselves [65]. Additionally, past research has demonstrated improvement in glycemic control for children with T1D can result from the responsibility of caring for others [66]. Researchers hypothesized that the structured care provided cues that translated to the performance of diabetes self-management behaviors. This suggests that by caring for another, possibly through a team setting, individuals learn to care better for themselves.

Implications for other groups developing an app for children with a chronic disease and their parent include understanding the current communication context that disease management takes place in. As a result of this insight, we were able to add a problem-solving framework into our app. We were also able to understand the importance of having the messages and reminders come from the app and not the parent. Team participation within the app appears to be a way to keep the adolescents engaged and responsible for their own care. Finally, we learned that some type of reward needs to be added. To achieve this goal, we have added the ability for the adolescent to purchase different accessories for their in-app avatar. Without understanding the communication that currently takes place between the parent and the child, we would have not been able to make these adjustments.

Whereas there are some limitations to this work, the goal was to gain a patient-centered perspective on the app design to date. These focus groups are not representative of all mHealth apps, as ours is specifically for adolescents with T1D and includes their parent. Additionally, our sample size was small. However, we believe that this work provides information that may help others in their app development process. Future studies of this type should include larger samples, including a wider age range of youth.

This app has the potential to provide parents the security they need to start transitioning diabetes management to their children and allowing them to develop the confidence needed for ongoing optimal control of their diabetes. This study used an original app idea and demonstrated a prototype of the app to help guide both the parents’ and the adolescents’ perspectives to design this patient-centered diabetes app. Overall, the app review was positive and the participants thought it would be useful to aid in the transition to adolescent self-management. Understanding the needs of patients and caregivers will help better inform the development and design of the app. This study also provides others with strategies that can be incorporated into the development of other health apps, including use of a patient-centered design, incorporating interactivity, use of customization, and building teams within the app.