A Mobile App for the Self-Management of Type 1 Diabetes Among Adolescents: A Randomized Controlled Trial

Abstract

Background: While optimal blood glucose control is known to reduce the long-term complications associated with type 1 diabetes mellitus, adolescents often struggle to achieve their blood glucose targets. However, their strong propensity toward technology presents a unique opportunity for the delivery of novel self-management interventions. To support type 1 diabetes self-management in this population, we developed the diabetes self-management app bant, which included wireless blood glucose reading transfer, out-of-range blood glucose trend alerts, coaching around out-of-range trend causes and fixes, and a point-based incentive system.

Objective: The primary objective was to evaluate bant 's effect on hemoglobin A1c (HbA1c) through a randomized controlled trial (RCT). Secondary measures (eg, self-monitoring of blood glucose [SMBG]) were also collected to assess bant 's impact on the self-management behaviors of adolescents with type 1 diabetes.

Methods: We enrolled 92 adolescents into a 12-month RCT, with 46 receiving usual care and 46 receiving usual care plus bant. Clinical outcome data were collected at quarterly research visits via validated tools, electronic chart review, glucometer downloads, and semistructured interviews. App satisfaction was assessed at 6 and 12 months, and at trial end, users ranked bant components based on perceived usefulness. Mobile analytics captured frequency of blood glucose uploads, which were used to categorize participants into high, moderate, low, or very low engagement levels.

Results: Linear mixed models showed no changes in primary and secondary clinical outcomes. However, exploratory regression analysis demonstrated a statistically significant association between increased SMBG and improved HbA1c in the intervention group. For a subgroup of bant users taking SMBG ≥5 daily, there was a significant improvement in HbA1c of 0.58% (P=.02), while the parallel subgroup in the control arm experienced no significant change in HbA1c (decrease of 0.06%, P=.84). Although app usage did diminish over the trial, on average, 35% (16/46 participants) were classified as moderately or highly engaged (uploaded SMBG ≥3 days a week) over the 12 months.

Conclusion: Although primary analysis of clinical outcomes did not demonstrate differences between the bant and control groups, exploratory analysis suggested that bant may positively impact the use of SMBG data and glycemic control among youth. The next generation of bant will aim to remove barriers to use, such as deploying the app directly to personal devices instead of secondary research phones, and to explore the utility of integrating bant into routine clinical care to facilitate more frequent feedback. Future evaluations of mHealth apps should consider more robust research tools (eg, ResearchKit) and alternative RCT study designs to enable more rapid and iterative evaluations, better suited to the nature of rapidly evolving consumer technology.

Trial registration: ClinicalTrials.gov NCT01899274; https://ichgcp.net/clinical-trials-registry/NCT01899274 (Archived by WebCite at http://www.webcitation.org/6qWrqF1yw).

Keywords: adolescent; behavior change; blood glucose; cell phones; diabetes mellitus; gamification; mobile applications; mobile phone; self-care; self-management.

Conflict of interest statement

Conflicts of Interest: The Hospital for Sick Children and University Health Network jointly own intellectual property rights to the bant app. Under the respective agreements with their organizations, Joseph A Cafazzo, Mark R Palmert, Debra K Katzman, and Shivani Goyal are entitled to personally benefit from any commercial use of the intellectual property.

©Shivani Goyal, Caitlin A Nunn, Michael Rotondi, Amy B Couperthwaite, Sally Reiser, Angelo Simone, Debra K Katzman, Joseph A Cafazzo, Mark R Palmert. Originally published in JMIR Mhealth and Uhealth (http://mhealth.jmir.org), 19.06.2017.

Figures

Figure 1

The intervention includes an iPhone 4S loaded with bant, as well as a Bluetooth adapter attached to the OneTouch UltraMini blood glucose meter. Circles represent individual readings at different times of the day, with the bedtime reading having been selected to display further information; the blue region represents a particular participant’s target blood glucose range. The different colors of the circles represent the different reading contexts (eg, breakfast readings are blue).

Figure 2

Participant enrollment.

Figure 3

Mean hemoglobin A1c values for the intervention and control groups from baseline to 12 months.

Figure 4

Regression analysis for self-monitoring of blood glucose (SMBG) and hemoglobin A1c.

Figure 5

Longitudinal mean hemoglobin A1c for intervention and control participants with 12-month self-monitoring of blood glucose of 5 or more per day.

Figure 6

Overall satisfaction with bant at the 6- and 12-month time points.

Figure 7

Number of times (measured as days per month) users uploaded blood glucose data to bant across the study duration.