The design and development of MyT1DHero: A mobile app for adolescents with type 1 diabetes and their parents

Abstract

Introduction

Type 1 diabetes impacts approximately 1.25 m Americans, many of them young children. As a child grows, there is a transition towards independence and they must learn to manage their diabetes independently. The objective of this study was to design, develop and conduct a prototype test to assess the satisfaction and feasibility of a mobile app for adolescents with type 1 diabetes and their parents to aid in this transition.

Methods

Ten parent/adolescent groups used MyT1DHero for four weeks. They were given a pre-test/post-test survey regarding family conflict with three tasks of diabetes management and the general tone of family communication. At post-test they were asked questions regarding their satisfaction and perception of ease of use of the app. They also participated in short interviews regarding the app. Additionally, we used server data to examine actual app usage.

Results

The parents’ perceptions of conflict around the results of the blood sugar tests increased (t(9) = 2.71, p = .02) as did perceptions of conflict around logging the blood sugar results (t(9) = 3.67, p = .005). The adolescents perceived increased conflict surrounding the results around logging the blood sugars results (t(9) = 3.09, p = .01).There was no change in the tone of general family communication. During the prototype testing, we discovered that the app crashed repeatedly and several of the functions were not working properly. In the interview data, three main themes emerged, app-crashing issues, problems with notifications and positive feedback.

Discussion

Through this process, all of the reported issues were corrected and new features were planned for subsequent versions. A smartphone application has the potential to be a novel intervention for engaging adolescents and their parents in positive communication to support type 1 diabetes management.

Keywords

Type 1 diabetes mhealth telemedicine

Introduction

Type 1 diabetes (T1D) is a serious and chronic health condition that impacts approximately 1.25 m Americans and it is estimated that 40,000 new cases are diagnosed each year.¹ Currently, there is no cure for T1D. In many cases, the diagnosis of T1D is made in young children and parents are responsible for managing the illness.² Management of T1D is complex and includes monitoring multiple daily blood glucose measurements, carbohydrate intake, physical activity and adjusting insulin doses for injections or, if the child has a pump, for basal and bolus insulin delivery. However, as a child grows, there is a gradual transition towards independence and the child is given more responsibility for the management of the diabetes. This transition often starts at the time of early adolescence.³ Due to the complexity of proper T1D management, during the adolescent transition, approximately 75% of teens do not achieve adequate control of their blood sugars, as measured by the American Diabetes Association (ADA) haemoglobin A1c (HbA1c) targets.⁴ Past research has demonstrated that improved and continuous family involvement has myriad positive benefits as the adolescent becomes self-managed.^5–8

mHealth is the use of mobile technology to impact health outcomes. It has been shown to improve results in a plethora of health issues, including diabetes.^9,10 mHealth offers a way to help improve the management of the disease for this age group as digital technologies are so frequently part of the daily routine.¹¹ Almost three-quarters of teens have access to a smartphone and 94% report going online daily.¹² Additionally, 76% of parents report having a smartphone.¹¹ mHealth may offer a possible method of educating and encouraging communication about diabetes tasks to ease the transition to full adolescent self-management. Therefore, we developed an app to help in the transition of management. The objective of this app is to improve positive communication, self-efficacy, trust and autonomy to aid in the adolescents’ transition to T1D self-management.

Theoretical underpinnings

The development of this app has been guided by theoretical components of social cognitive theory (SCT) and self-determination theory (SDT). The use of SCT to guide interventions has demonstrated success in changing behaviours around disease management.^5,13–15 SCT posits a reciprocal relationship between an individual’s environment, personal factors and behaviours.^16–18 The overall goals of this app and the app intervention were designed using SCT. For example, we hope to change the individual’s personal factors by providing social support through the parent/child messaging feature,^19–22 including prompts to set small and achievable daily goals, reminders and educational tips about living with T1D to improve the adolescents’ self-efficacy.^23–28 SDT is a theory of motivation that has been used in health domains in the past. To encourage the use and engagement of the app,²⁹ we used SDT to develop many of the gamification features (i.e. points for usage).³⁰

Study objectives

The aim of the present study was to discuss the design and development of a mobile phone app for adolescents with T1D, and to conduct prototype testing of the app. We followed a user-centred design approach in which we actively engaged all key stakeholders throughout the process. The main objectives of the prototype testing was to assess the feasibility, satisfaction and ease of use of the app. Additionally, we were interested in issues around communication of managing the adolescents’ diabetes to prepare for our future intervention study.

App development

The development process consisted of four key stages: 1. inductive focus groups and interviews with adolescents and teens with T1D, parents of children with T1D and certified diabetes educators; 2. development of an application wireframe; 3. app development; and 4. app prototype testing. Stages 1 and 2 have been discussed elsewhere;^31,32 therefore a brief summary of the findings is presented here. Stage 3 includes information regarding the app development process and Stage 4 describes the prototype testing study.

Stage 1: Inductive focus groups and interviews

The first stage of this process included conducting focus groups and interviews with key stakeholders. This was done in order to gain a broader context of T1D management as an adolescent and parent. Through this process, we learned that communication between the adolescent and parent was important in the management of the disease. However, this communication was perceived by the adolescents as frequently negative.³¹ Therefore, we decided to focus on promoting positive family communication as a way to improve adherence to management.

Stage 2: Wireframe development

In this stage, a multi-disciplinary team of paediatric endocrinologists, nurse practitioners who specialise in diabetes, health communication experts and technology developers used the information from Stage 1, along with a review of current literature to develop a wireframe of the app. This included use cases for the flow of the messaging, logging blood sugars and development of pre-programmed messaging for the parents and the adolescents.

In order to understand the perspective of the parents and the adolescents, focus groups were conducted to demonstrate our vision of the app. We received feedback that aided in finalising our design.³² Through this exercise, it was decided to include gamification (i.e. the application of game-design elements in other environments) aspects in our app for the adolescent interface.

Stage 3: App development

Once feedback from all of the user groups was compiled, the app development group began developing the app. This process included finalising the back-end server process for registration (i.e. linking the parent app to the child’s app), completing design components, and adding game features and educational components. Full design and development of the app took 10 months.

App description

MyT1DHero is an app that links parents and their children by creating two separate app interfaces, one for the parent and one for the child, that work together to help them communicate about diabetes management. The app has customisable blood glucose reminders and ranges, videos of support from other adolescents with T1D, snack lists, and other educational information on T1D.

On the parent app interface, parents have the ability to send and receive messages from their child and view their child’s blood glucose trends. Parents receive notification of a blood glucose test that has been entered into the app and are able to communicate with their child about their plan for care via the app. Parents also receive social support through positive messages sent through the app. Additionally, there are pre-populated supportive parent messages that can be sent to their child. See Figures 1 and 2 for parent screenshot examples.

Figure 1.

Parent screenshot.

Figure 2.

Parent screenshot.

On the child app, the adolescents have a place to log their blood glucose, and send and view their blood glucose trends. The app also uses a super-hero avatar that can be customised. By using the app, the adolescents can earn points, which can be used to purchase ‘accessories’. Furthermore, the app has pre-programmed messages that the child can send to the parent asking for specific help in managing their diabetes. See Figure 3 for an example of an adolescent screenshot.

Figure 3.

Adolescent screenshot.

Methods

Stage 4: Prototype testing

Participants and recruitment

The prototype testing literature states that five users would be sufficient to test the app.³³ In order to get a wider variety of opinion and feedback we hoped to enrol 10 family pairs.

To be eligible for the study, the adolescents had to be 10–15 years old, have had a T1D diagnosis for at least six months, be fluent in English, have a parent/guardian willing to participate and be allowed to use a mobile phone for the study. The parent had to have a child with T1D, be fluent in English and have daily access to the Internet. Participants were recruited through the Southeast Michigan JDRF (formerly Juvenile Diabetes Research Foundation) social media groups, word-of-mouth and snowball sampling. Ethics approval to conduct the study was obtained from the Michigan State University Human Research Protection Programme.

Procedures

Once the participants were screened, enrolment and training sessions were held at local coffee shops to enrol the families and demonstrate the app. If participants did not have an Android phone, they were provided with a phone and a service plan for the duration of the study.

Study variables

Short pre-test/post-test surveys were administered using three items from the Diabetes Family Conflict Scale (DFCS), a validated scale to measure perceived frequency of family conflict³⁴ surrounding remembering to test blood sugars, results of the blood sugar tests and the logging of the results. Four items assessed the tone of general family communication; for this scale we used adjectives that were used in our focus groups and the open-ended survey responses. These items were measured on a five-point Likert scale (1 = always to 5 = never). The main objective of this study was to investigate the perceptions of usability and satisfaction with the app. Therefore, we did not use the full versions of the DFCS, but wanted to understand if there was any change in a few of these key measures. For the post-test, validated scales were used to measure the usability and satisfaction of the app,³⁵ along with a short interview regarding feedback about using the app. The interviews ranged from 12–30 min long, were audio-recorded and transcribed verbatim. Finally, server data was extracted to understand actual use of the app.

Data analysis

The survey data was analysed using SPSS version.³⁶ Descriptive statistics were used to analyse the demographic questions as well as the ease of use and satisfaction questions. For the satisfaction questions, there were no significant differences between the parent and adolescent responses so they were analysed together. For the pre-test/post-test questions, we conducted a paired independent sample t-test for the parent results and the adolescent results. The interview data was analysed by hand, using thematic analysis.³⁷ Two members of the research team independently read through the transcripts and developed key themes, which were then compared, and codes assigned. Disagreements were resolved through discussion.

Results

These results focus solely on the prototype testing (Stage 4).

Participants

Thirteen families came to the enrolment sessions, but three families dropped out of the study before completion. One family was unable to get phone service on the provided phones in their home. Two families never completed the post-test survey and did not respond to follow-up, see Figure 4.

Figure 4.

MyT1DHero prototype testing participant diagram.

All of the parent participants were female (n = 10), the biological mother of the adolescent and white. Most were between the ages of 45 and 54 (n = 7) and were married (n = 6). Four of the parents had some level of college, but no degree. The average income of the families by county was $72,209 (range: $61,097–$84,783). The adolescent participants ranged in age from 11–15 years, half were 12 years old (n = 5) and six were male. Eight of the adolescents managed their diabetes by using an insulin pump. Four of the 10 used a continuous glucose monitor (CGM). Half of our adolescents had been diagnosed for between 13 months and five years (n = 5); two had been diagnosed for a year or less, and three greater than five years. See Table 1 for more detailed information.

Table 1.

Participant demographics.

Age of Adolescent Participants		Age of Parent Participants
Age (years)	Count	Age range (years)	Count
11	1	25–34	1
12	5	35–44	2
13	2	45–54	7
14	1
15	1	Education	Count
		Some college, no degree	4
Length of Diagnosis	Count	Associate’s degree	1
6 mths–12 mths	2	Bachelor’s degree	1
13 mths–5 yrs	5	Master’s degree	1
6 yrs–10 yrs	3	Professional degree (Ex. JD, MD)	1
		Did not answer	2
Management of Diagnosis	Count
Pump & CGM	3	Marital Status	Count
Injections & CGM	1	Single, never married	1
Pump	5	Married or domestic partnership	6
Injections	1	Divorced	1
		Did not answer	2

Quantitative results

The pre-test/post-survey asked both the parents and the adolescents to report on conflict around diabetes management tasks during the past month. The parents reported no change around perceived conflict surrounding remembering to check blood sugars (t(9) = .36, p = .72). However, subjects reported arguing more with their child about the results of the blood sugar tests (t(9) = 2.71, p = .02) and logging the blood sugar results (t(9) = 3.67, p = .005). The adolescents reported no change around perceived conflict surrounding remembering to check blood sugars (t(9) = .61. p = .56) or the results of the blood sugar tests (t(9) = −.39, p = .7). They did perceive an increase in conflict around the logging of the blood sugar results (t(9) = 3.09, p = .01). See Table 2 for more information.

Table 2.

Conflict around diabetes management.

	Parent			Adolescent
During the past month, I have argued with my parent/my child about	Pre-test Mean SD	Post-test Mean SD	Cohen’s d	Pre-test Mean SD	Post-test Mean SD	Cohen’s d
Remembering to check blood sugars	3.50 (.97)	3.40 (.70)	0.12	3.70 (1.16)	3.30 (1.34)	0.31
Results of blood sugar monitoring	4.30 (.68)	3.70 (.82)	0.79	4.10 (.99)	4.30 (.95)	0.21
Logging blood sugar results	4.60 (.84)	3.40 (.84)	1.42	4.70 (.68)	3.50 (1.08)	1.33

1= always, 2 = very often, 3 = sometimes, 4 = rarely, 5 = never

When examining general perceptions of communication, there were no significant differences pre- and post-test. We asked about positive communication (parent results, t(9) = -.56, p = .59; adolescent results, t(9) = .68, p = .21), strained communication (parent results, t(9) = −.61, p = .55; adolescent results, t(9) = 1.5, p = .33), stressful communication(parent results, t(9) = −1.46, p = .18; adolescent results, t(9) = 1.37, p = .43) and calm communication (parent results, t(9) = −1, p = .35; adolescent results, t(9) = .71, p = .25). See Table 3.

Table 3.

General communication characteristics.

	Parent			Adolescent
During the past month, my communication with my parent/my child has been…	Pre-test Mean (Std. Dev.)	Post-test Mean (Std. Dev.)	Cohen’s d	Pre-test Mean (Std. Dev.)	Post-test Mean (Std. Dev.)	Cohen’s d
Positive	1.80 (.42)	1.90 (.57)	0.2	1.70 (.68)	1.60 (.70)	0.14
Strained	3.80 (.63)	4 (.67)	0.3	4 (1.5)	4.4 (.84)	0.33
Stressful	3.60 (.70)	4.10 (.57)	0.78	3.90 (1.37)	4.40 (.70)	0.46
Calm	2 (.54)	2.13 (.35)	0.29	2.13 (.64)	1.75 (.71)	0.52

1= always, 2 = very often, 3 = sometimes, 4 = rarely, 5 = never

Regarding the user’s ease of use and satisfaction of the app, the participants found it understandable, easy to use and were satisfied with the app. Additionally, they disagreed with the statement that they thought that the app was confusing. Participants were neutral about whether they liked using the app and that the app was fun to use. See Table 4.

Table 4.

Ease of use and satisfaction.

	Mean (Std. Dev.)	Range
The MyT1DHero app was understandable	4.10 (.97)	1–5
It was easy to become skilled at using the app	4.00 (1.08)	1–5
I found the MyT1DHero app easy to use	4.16 (.69)	3–5
The MyT1DHero was confusing	2.15 (.67)	1–4
Learning to operate the MyT1DHero app was easy to use	4.40 (.59)	3–5
I liked using the MyT1DHero app	3.63 (1.07)	1–5
I thought using the MyT1DHero app was fun	3.40 (.68)	3–5

1= strongly disagree, 2 = disagree, 3 = neutral, 4 = agree, 5 = strongly agree

App server data revealed that the app was used every day for the full month. We defined this as the number of times that each individual logged into the app. The daily mean ranged from 2–86 times daily. There was an early peak of app use at the start of the prototype testing and then a drop-off around day 21. See Figure 5. Additionally, the users explored many features each time they logged in. The dashboard was the most used feature, followed by the blood glucose logging, and the avatar character section (child participants only).

Figure 5.

Mean app usage by day.

Qualitative results

There were three main themes that emerged from the data: app-crashing issues, problems with notifications and positive feedback.

App-crashing issues

The most frequently reported issue during prototype testing was that the app often crashed. At the beginning of the study, the user registration process was not clear to the participants and the app often crashed during this time. The registration process included having the parent set up the child’s information, including blood sugar ranges and blood sugar testing times. Developers began working to fix the app-crashing issues immediately after being notified, and although most of the crashing issues were fixed before the end of the study, participants reported some crashing for the entire study duration. One parent said,

Mornings are the worst for the app to work. I’ve tried logging out but then the login button will not respond. Sometimes if I reboot the phone the login will either automatically log him back in but usually it will ask to login again and the button will not respond, then minutes or hours later he will be logged in again.

One participant noted, ‘I would go to something else and when I would go back to the app itself it would always crash with an error message so I would have to open it back up again.’ Toggling between two different apps caused the app to crash, even after the major crashing issues had been fixed.

Problems with notifications

Notifications sent by the system included reminders to the child that they needed to test, notification to the parents that their child had tested, and communication between the parent and the child. Because of problems with this feature of the app, the users were not receiving clear notifications to go into the app. In fact, the prototype version of the app did not have reliable notifications for any of the tasks. According to many of the parents, this resulted in very low adherence in terms of the child entering their blood glucose values. As for feedback from the participants, the majority of individuals, both parent and child, stated that notifications would be helpful for them in remembering to check and enter blood glucose values, in seeing their child’s blood glucose values and in exchanging messages within the app.

Positive feedback

Despite the many changes needed to fine-tune the app based on participant feedback, their overall view of the app was very positive. One parent said,

I know things have to be tweaked, they always do in the beginning. We love to be able to help get this to be an awesome app. We both love the idea of the app and when it works it’s a great tool and a way the child can become independent with a little supervision.

Many of the kids also really enjoyed the app, ‘It was very fun and I couldn’t stop using it and I kept wanting to check my sugar to put it in and get new stuff to upgrade the person [hero].’

Discussion

This paper describes the design, development and prototype testing of the MyT1DHero mobile phone app. This app seeks to improve the communication around diabetes management between an adolescent and parent. While both participant user types generally liked the app, technical issues were found and impacted many of the overall results of prototype testing.

Our prototype testing included 10 family dyads (n = 20), which is adequate for prototype testing,³³ over a four-week period. In regards to the family conflict results, we believe that we may have triggered a Hawthorne effect,³⁸ which was exacerbated by the lack of functioning reminders to log glucose values throughout this test period. The app was intended to help the adolescents remember to test the blood glucose and to log those results, which would then be sent to the parent to view. However, with this function not working properly, it is possible that the app caused more conflict. Alternately, as the participants were aware conflict was being examined as part of the study, they may have thought about and identified this behaviour in a more salient manner.³⁹ Past literature has suggested that family conflict is normative, especially between parents and adolescents⁴⁰ and conflict provides an opportunity for a more open conversation.⁴¹ However, it has been demonstrated that conflict specifically surrounding diabetes can lead to poorer adherence and blood glucose control⁴² and thus should be monitored. Again, we suspect that this was likely to have been around the parents asking if their child had entered their blood sugar number because the app was not sending notifications, as well as conflict around the app crashing, both of which have been fixed.

While there was no significant effect on general communication, those results trended in a positive direction. We hypothesise that with a properly working app used over a longer period of time, family communication would improve in the future. Additionally, while our effect sizes are somewhat larger than expected, they should be taken with extreme caution, as we believe that they are due to the small sample size and the homogeneity of the sample.⁴³

Due to technical difficulties, the app user data may have underestimated potential future benefit. All of the technical issues reported by the participants have been resolved, and user experience is likely to improve. Additionally, a spike in use was noted at the beginning of the trial, which is expected in the use of new apps. There was a slight drop in use at day 21; this also seems to be common.⁴⁴ Therefore, the frequent addition of new content or updates to the app may be necessary to keep people engaged in using it.⁴⁵ Furthermore, as this was a test to see what the users would do with the app, we did not require as specific a ‘dose’ of the app during the prototype testing. However, we did suggest at least four times a day as the minimum recommended blood glucose testing frequency. Based on user feedback, planned app enhancements include a free-text option for messages, additional badges for uses, daily schedule changes and other usability fixes (e.g. making the ranges easier to program) will be included in the next iteration of the app. Overall, the participants generally thought that the app design was easy to use, learn and understand.

Limitations to this study include a small and homogeneous sample size, which limits generalisation of results. Additionally, the app was only used by mothers; father participants may have provided a different perspective.

Although the number of apps targeting diabetes and specifically T1D is increasing, there is no other current app (of which the authors are aware) that specifically focuses on adolescent–parent communication. Overall, the feedback was positive, despite the technical issues. The qualitative results provided a number of suggestions to improve the app. Through this testing and feedback process, issues were corrected and new features are planned for subsequent versions. A larger intervention is currently being conducted to test the efficacy of the revised version of the app and will include assessment of blood glucose control by measuring the child’s HbA1c. Smartphone applications have the potential to be a novel intervention for engaging adolescents and their parents in positive communication to support T1D management.

Footnotes

Acknowledgements

We would like to recognise, Bion Bilateral who helped us develop the app, and JDRF Metro Detroit/Southeast Michigan Chapter for their support.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the American Diabetes Association (ADA).

References

Juvenile Diabetes Research Foundation. Type 1 diabetes facts, http://jdrf.org/about-jdrf/fact-sheets/type-1-diabetes-facts/ (2017, accessed 31 July 2017).

Pierce

et al.

Health care transition for young adults with type 1 diabetes: stakeholder engagement for defining optimal outcomes. J Pediatr Psychol 2017; 42(9): 970–982.

Ersig

et al.

Stressors in teens with type 1 diabetes and their parents: immediate and long-term implications for transition to self-management. J Pediatr Nurs 2016; 31: 390–396.

American Diabetes Association. 11. Children and adolescents. Diabetes Care 2016; 39: S86–S93.

Anderson

et al.

Family conflict, adherence, and glycaemic control in youth with short duration type 1 diabetes. Diabetic Med 2002; 19: 635–642.

Anderson

et al.

An office-based intervention to maintain parent–adolescent teamwork in diabetes management. Impact on parent involvement, family conflict, and subsequent glycemic control. Diabetes Care 1999; 22: 713–721.

La Greca

et al.

I get by with a little help from my family and friends: adolescents’ support for diabetes care. J Pediatr Psychol 1995; 20: 449–476.

Laffel

et al.

Impact of ambulatory, family-focused teamwork intervention on glycemic control in youth with type 1 diabetes. J Pediatr 2003; 142: 409–416.

Hamine

et al.

Impact of mHealth chronic disease management on treatment adherence and patient outcomes: a systematic review. J Medical Internet Res 2015; 17: 2–2.

10.

Holtz

Lauckner

. Diabetes management via mobile phones: a systematic review. Telemed J E Health 2012; 18: 175–184.

11.

Anderson M, Singh A and Page D. Parents, teens and digital monitoring. Report, Pew Research Center, US, January 2016.

12.

Lenhart A, et al. Teens, social media & technology overview 2015: Internet & American Life Project. Report, Pew Research Centre, US, April, 2015.

13.

Huang

et al.

Preparing adolescents with chronic disease for transition to adult care: a technology program. Pediatrics 2014; 133: e1639–e1646.

14.

Lubans

et al.

Preventing obesity among adolescent girls: one-year outcomes of the Nutrition and Enjoyable Activity for Teen Girls (NEAT Girls) cluster randomized controlled trial. Arch Pediatr Adolesc Med 2012; 166: 821–827.

15.

Martin

et al.

Using social cognitive theory to predict physical activity and fitness in underserved middle school children. Res Q Exerc Sport 2011; 82: 247–255.

16.

Pendley

et al.

Peer and family support in children and adolescents with type 1 diabetes. J Pediatr Psychol 2002; 27: 429–438.

17.

Greco

et al.

A peer group intervention for adolescents with type 1 diabetes and their best friends. J Pediatr Psychol 2001; 26: 485–490.

18.

Ravert

Hancock

Ingersoll

. Online forum messages posted by adolescents with type 1 diabetes. Diabetes Educ 2004; 30: 827–834.

19.

Skinner

John

Hampson

. Social support and personal models of diabetes as predictors of self-care and well-being: a longitudinal study of adolescents with diabetes. J Pediatr Psychol 2000; 25: 257–267.

20.

Hanna

Guthrie

. Adolescents’ behavioral autonomy related to diabetes management and adolescent activities/rules. Diabetes Educ 2003; 29: 283–291.

21.

Hanna

et al.

Emerging adults with type 1 diabetes during the first year post-high school perceptions of parental behaviors. Emerging Adulthood 2013; 2(2): 128–137.

22.

Skinner

Hampson

. Social support and personal models of diabetes in relation to self-care and well-being in adolescents with type I diabetes mellitus. J Adolesc 1998; 21: 703–715.

23.

Stupiansky

et al.

Impulse control, diabetes-specific self-efficacy, and diabetes management among emerging adults with type 1 diabetes. J Pediatr Psychol 2012; 38: 247–254.

24.

Law

et al.

Adolescent and parent diabetes distress in type 1 diabetes: the role of self-efficacy, perceived consequences, family responsibility and adolescent–parent discrepancies. J Psychosom Res 2013; 74: 334–339.

25.

Iannotti

et al.

Self-efficacy, outcome expectations, and diabetes self-management in adolescents with type 1 diabetes. J Dev Behav Pediatr 2006; 27: 98–105.

26.

Anderson

et al.

Diabetes empowerment scale: a measure of pscyo-social self-efficacy. Diabetes Care 2000; 23: 739–743.

27.

Nouwen

et al.

Comparison of the role of self-efficacy and illness representations in relation to dietary self-care and diabetes distress in adolescents with type 1 diabetes. Psychol Health 2009; 24: 1071–1084.

28.

Viner

et al.

Motivational/solution-focused intervention improves HbA1c in adolescents with type 1 diabetes: a pilot study. Diabetic Med 2003; 20: 739–742.

29.

Cafazzo

et al.

Design of an mHealth app for the self-management of adolescent type 1 diabetes: a pilot study. J Med Internet Res 2012; 14: e70–e70.

30.

Deci

Ryan

. The general causality orientations scale: self-determination in personality. J Res Pers 1985; 19: 109–134.

31.

Holtz B, et al. Three perspectives on the transition to self-management. Presented at the 14th Annual Pediatric Research Conference, Detroit, MI, US.

32.

Holtz

et al.

Developing a patient-centered mhealth app: A for adolescents with type 1 diabetes. JMIR mHealth uHealth 2017; 5(4): e53–e53.

33.

Nielsen J and Landauer TK. A mathematical model of the finding of usability problems. In: Proceedings of INTERCHI ’93 conference on human factors in computer systems (eds. S Ashlund, A Henderson, E Hollnagel, et al.), Amsterdam, the Netherlands, April 24-29 1993, pp. 206–213, Amsterdam: IOS Press.

34.

Hood

et al.

Updated and revised diabetes family conflict scale. Diabetes Care 2007; 30: 1764–1769.

35.

Lewis

. IBM computer usability satisfaction questionnaires: psychometric evaluation and instructions for use. Int J Hum Comput Interact 1995; 7: 57–78.

36.

IBM Corp. IBM Statistics for Mac, Version 23. Released 2016, Armonk, NY: IBM Corp.

37.

Braun

Clarke

. Using thematic analysis in psychology. Qual Res Psychol 2006; 3: 77–101.

38.

Roethlisberger

Dickson

. Management and the Worker, Vol. 5: Psychology Press, 2003.

39.

Fagan

Design and code inspections to reduce errors in program development. In: Broy

Denert

(eds). Software pioneers, Berlin: Springer, 2002, pp. 575–607.

40.

Sillars

Canary

Tafoya

Communication, conflict, and the quality of family relationships. In: Vangelisti

(ed). LEA’s communication series. Handbook of family communication, Mahwah, NJ: Lawrence Erlbaum Associates, 2004, pp. 413–446.

41.

Fitzpatrick

Ritchie

. Communication schemata within the family. Hum Commun Res 1994; 20: 275–301.

42.

Anderson

. Family conflict and diabetes management in youth: clinical lessons from child development and diabetes research. Diabetes Spectr 2004; 17: 22–26.

43.

Ioannidis

. Why most discovered true associations are inflated. Epidemiology 2008; 19: 640–648.

44.

Nielsen Corp. So many apps, so much more time for entertainment, http://www.nielsen.com/us/en/insights/news/2015/so-many-apps-so-much-more-time-for-entertainment.html?afflt=ntrt15340001&afflt_uid=kHO3g3-FBdk.4_iT8AmqXblFk7h5gTxIGuaU53OB8WY5&afflt_uid_2=AFFLT_ID_2 (2015, accessed 7 July 2017).

45.

Dennison

et al.

Opportunities and challenges for smartphone applications in supporting health behaviour change: qualitative study. J Med Internet Res 2013; 15(4): ).