Adolescence constitutes a critical developmental period during which youth establish their autonomy while grappling with the social, emotional, and physical challenges of becoming young adults. This period represents a time of tremendous vulnerability, during which psychiatric and mood disorders often manifest [1]. Adolescence is even more challenging for youth with type 1 diabetes (T1D), as they contend with the daily challenges of disease management in addition to the developmental challenges inherent to this phase of development.

Diabetes distress (DD) is an emotional construct that refers to the negative emotions that arise from living with and managing diabetes. The majority of youth with T1D experience some level of DD [2,3], with adolescent females experiencing greater DD than males [4-6]. More than one-third of youth with T1D suffer from substantial DD [7]. DD is apparent in youth for many reasons, including fear of stigma from peers [8], feeling overwhelmed by their diabetes regimen [9], and worry of hypoglycemia and long-term complications of a potentially fatal condition [9].

Self-compassion, a positive psychological construct, involves being open to one’s own suffering and treating oneself with the same care one would show to loved ones [10]. Youth can have varying degrees of innate self-compassion [11]. Higher levels of self-compassion are associated with lower levels of distress in youth with other chronic medical conditions [12], and it is a psychological characteristic that correlates with favorable outcomes, including psychological well-being, self-care, and hemoglobin A_1c (HbA_1c) among individuals with diabetes [13].

Among adults with diabetes, an intervention to increase self-compassion resulted in reductions in DD [14]. However, this relationship has yet to be shown in youth with T1D. Self-compassion interventions lead to better psychological well-being and decreased psychopathology in healthy adolescents and those with some chronic conditions [15,16]. Since self-compassion is a modifiable behavior on which one can be trained [17,18], we have a largely untapped opportunity for intervention among youth with T1D.

Mindfulness-based therapy has led to short-term improvements in self-reported stress among adults and emerging adults with T1D [19]. However, a sole focus on mindfulness is not sufficient, as it does not provide enough training around active self-comforting in suffering. Mindfulness also does not provide the aptitude to address the feelings of failure and self-blame that are common among youth with T1D [20]. Therefore, our goal is to study the effects of a psychological intervention combining mindfulness and self-compassion (mindful self-compassion for teens [MSC-T]) on DD among youth with T1D.

Our primary objective is to compare DD at 3 months among youth with T1D who participated in the MSC-T intervention to those in a waitlist control group. Our secondary objectives are to (1) compare anxiety, depression, diabetes-related disordered eating, and suicidal ideation at 3 months among youth with T1D who participated in the MSC-T intervention with those in the waitlist control group; and (2) compare DD, anxiety, depression, diabetes-related disordered eating, and suicidal ideation at 12 months among youth with T1D who participated in the MSC-T intervention with those in the waitlist control group. Finally, our exploratory objectives are to (1) explore if the effect of the MSC-T intervention changes over time throughout the 12-month follow-up period, (2) compare glycemic control (HbA_1c and time in range) among adolescents with T1D who participated in the MSC-T intervention to those in the waitlist control group, and (3) explore if the effect of the web-based MSC-T intervention on DD differs based on self-reported gender.

We hypothesize that adolescents who participate in the MSC-T intervention will report lower DD, anxiety, depression, diabetes-related disordered eating, and suicidal ideation at 3 months compared with the waitlist control group.

This study was reviewed and approved by the local Research Ethics Board at the Children’s Hospital of Eastern Ontario (CHEO; protocol 22/09E, approved July 17, 2022). Following an introduction from a member of the participant’s circle of care, in line with requirements outlined in the Personal Health Information Protection Act, a member of the research team will obtain written informed consent, either in person or remotely, using our REDCap (Research Electronic Data Capture; Vanderbilt University) system [21,22]. Participants will consent on their own behalf and will be compensated with gift cards for completing the study questionnaires (US $15 for baseline, US $20 for week 8, US $30 for month 3, US $30 for month 6, and US $35 for month 12) and up to 15 volunteer hours.

We will conduct a single-center, parallel-group randomized controlled trial assessing the effect of a web-based MSC-T intervention on DD in adolescents with T1D followed for 12 months, compared with a waitlist control group (standard of care).

We will recruit 140 adolescents with T1D, aged between 12 and 17 years, from the Diabetes Clinic at CHEO in Ottawa, Ontario, Canada.

The inclusion criteria and exclusion criteria for this study are presented in Textbox 1.

As shown in Figure 1, participants from the diabetes clinic at CHEO will be randomized to the MSC-T intervention group or the waitlist control group (standard of care).

We will use the randomization module in REDCap [21,22] to allocate participants to their group. The randomization table will be created by a study team member who is not directly involved in patient recruitment or data collection using R (R Core Team) [24]. The trial coordinator will generate the allocation for each participant following obtaining informed consent and completion of the baseline assessments (ie, there will be hidden allocation).

Participants will be randomly allocated to either the MSC-T intervention group or the waitlist control group using a 1:1 ratio block randomization. We will stratify each block by age group (12-14 years [early adolescence] or 15-17 years [middle adolescence]). The cohort will be divided by these age groups to deliver the intervention with the goal of fostering a developmentally appropriate and open environment. Stratification by age group is therefore necessary to ensure an equal age distribution between the treatment groups, since the developmental age of a participant could impact their response to the intervention [25]. We will also randomly vary the block size, with blocks of either 6 or 8, to prevent prediction of group allocation as participants are accrued [26].

We will collect the following demographic information at baseline: age, postal code (to determine socioeconomic status using neighborhood-level material and social deprivation indices [27,28]), optional self-reporting of gender identity (participants will be presented with a range of options [29,30]; Table 1), race or ethnicity, and indigeneity [31] (Table 2), and participant’s social context (living arrangements, etc). Medical history will be collected at baseline, and at intervals, it will be reviewed at each follow-up visit to determine if participants have been newly diagnosed with a serious mental illness (exclusion or withdrawal criteria).

Our primary outcome, DD, has been shown to be the psychological construct most related to metabolic outcomes in adolescents with T1D [3]. DD will be measured using the Problem Areas in Diabetes-Teens (PAID-T) version [32], a validated measure to evaluate diabetes-specific emotional distress in adolescents with high internal consistency (Cronbach α=.96) [32]. It consists of 26 items, each scored on a 6-point ordinal scale, allowing for the calculation of an overall distress score of 26-156, with scores of <70, 70-90, and >90 representing none-to-mild, moderate, and severe DD, respectively [33]. The PAID-T has excellent psychometric properties [34], is sensitive to change from behavioral intervention [35], and is a commonly used measure to assess DD in adolescents [7,36].

Our secondary outcomes are anxiety, depression, diabetes-related disordered eating, and suicidal ideation. Symptoms of anxiety will be measured using the Generalized Anxiety Disorder 7-item (GAD-7) scale, a self-report tool to detect clinically significant anxiety symptoms. The GAD-7 is scored from 0-21, with 0-4, 5-9, 10-14, and 15-21 indicating minimal, mild, moderate, and severe anxiety symptoms, respectively [37]. It has been validated in adolescents [38] and has previously been used to assess adolescents with T1D [39,40].

Symptoms of depression will be measured using the Patient Health Questionnaire-9 (PHQ-9) by self-report [41]. The PHQ-9 is a valid and reliable screening tool for identifying adolescents meeting the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) criteria for major depression [41]. Scores range from 0-27, and in adolescents, scores of 0-4, 5-10, 11-14, 15-19, and ≥20 indicate absence of, mild, moderate, moderately severe, and severe depression, respectively [41].

Symptoms of diabetes-related disordered eating will be measured using the Diabetes Eating Problem Survey-Revised (DEPS-R) version [42], a diabetes-specific measure of disordered eating symptoms that is valid for use in adolescents with T1D [42]. The DEPS-R is scored from 0-80, with higher scores indicating more disordered eating behaviors, and a score of ≥20 is suggestive of disordered eating behaviors [42]. The use of a diabetes-specific tool to measure disordered eating is important, as generic measures may overestimate the prevalence of disordered eating symptoms given that diabetes management involves an emphasis on food intake and carbohydrate counting and that insulin restriction is not captured in generic measures [43].

Suicidal ideation will be assessed using 1 question on the PHQ-9, which asks the respondent if they have had thoughts that they would be better off dead or of hurting themselves in the previous 2 weeks. This question is scored on a scale from 0 (not at all) to 3 (nearly every day). In line with previous use of the PHQ-9 [44], we will consider any score of >0 to be indicative that suicidal ideation is present.

Our exploratory outcomes are the effects of time, HbA_1c, time in range, and effect of gender. We will assess whether the effect of the MSC-T intervention on DD changes over time by exploring the interaction of time (ie, study time point) with DD. We will assess both HbA_1c and time in range, as not all our clinic patients wear a continuous glucose monitor (CGM), which is necessary to calculate time in range (>70% of our clinic patients wear a CGM).

HbA_1c is a measure of the average blood glucose levels over the previous 3 months and is a routinely followed biochemical marker used to gauge metabolic control in individuals with T1D. We will explore changes in HbA_1c over the duration of our trial, as mental health morbidity has been associated with higher HbA_1c levels [45,46]. Time in range is a metric that denotes the proportion of time that a person’s glucose level is within a desired target range, calculated using a CGM. A total of 14 days of data from CGM provides a good approximation of glucose data [47]; time in range will allow us to evaluate the glucose control at the various time points. We will collect CGM data at all study visits. Given the web-based nature of the trial combined with the high proportion of our clinic population that uses CGM, which is a more reliable measure of metabolic control [48], we will minimize additional blood collection by recording results from clinically collected HbA_1c at baseline and 1 year.

All the mental health outcomes being assessed are self-reported by participants directly into the study database. The principal investigator or delegate and study coordinator will be immediately alerted through an automated REDCap algorithm when (1) an individual self-reports as having suicidal ideation >0 on question 9 of the PHQ-9 or (2) an individual intentionally does not fill out the suicidality question on the PHQ-9. The principal investigator, delegate, or study coordinator will contact the participant’s endocrinologist or the endocrinologist on call, who will conduct a same-day risk assessment using the Ask Suicide-Screening Questionnaire (ASQ) [49]. If the participant is judged to have acute suicidality or suicidal ideation as per the ASQ, the assessor will either suggest a referral to the emergency department or contact the psychiatrist on call, who will suggest clinical follow-up as appropriate. The assessor will report the results of the ASQ screening back to the study coordinator within 1 day. Participants with acute suicidality as per the ASQ will then be excluded or withdrawn from the study.

To foster an environment where the participants feel comfortable, the web-based MSC-T intervention will be conducted with groups of 12-18 participants of the same age group [18]. To ensure that there is no exclusion of youth based on socioeconomic status or access to technology, participants who do not have a computer, laptop, tablet, or webcam to attend the web-based sessions will be loaned a study tablet, and mobile internet devices will be provided to those who do not have high-speed internet at home. All participants will receive a subsidy to offset the cost of increased home internet use for the duration of the study. To maximize participant engagement, the trial coordinator will provide weekly reminders for the duration of the program through SMS text messages or email, depending on the participant’s preference.

At each follow-up visit (Table 3), the study coordinator will review the interim medical history using a study case report form and have the participants complete the study questionnaires.

To ensure that this study is feasible, interesting, and relevant to youth with T1D, we have engaged and plan to continue to engage with people with lived experience throughout all stages of the study. We have 2 people with lived experience as collaborators (SH and OI), both of whom are former patients of the CHEO diabetes clinic, who will help spearhead engagement strategies. Multimedia Appendix 1 provides more information about our people with lived experience engagement activities.

We have considered gender and diversity in all aspects of this study, from recruitment through to knowledge translation. Multimedia Appendix 2 [4-6,50-52] provides more information about our equity, diversity, and inclusion considerations.

In this study, we will evaluate MSC-T as a way to improve DD and mental health outcomes among adolescents with T1D. We will also explore the impact of the MSC-T intervention on glycemic control. In adolescents with T1D, psychological interventions have largely targeted improvement in diet and insulin regimen adherence and strategies to cope with specific stressors [61]. While some approaches, including cognitive behavioral therapy (CBT), have shown promise for improving diabetes management, there is limited evidence that these traditional approaches translate to improved coping with distress [61]. A recent review [62] on the use of CBT for stress and coping in youth with T1D showed conflicting results, with a number of studies showing no improvement in stress [63] or DD [64]. CBT may also highlight potential causes of distress related to diabetes that the teenager has not yet considered without providing adequate skills and practices to manage these emotions [64].

We hypothesize that the combination of mindfulness and self-compassion practices in the MSC-T program will result in improved DD and mental health outcomes for adolescents enrolled in the program. We also anticipate that these improvements may translate to improved clinical outcomes through glycemic control.

To our knowledge, this is the first randomized controlled trial using the MSC-T intervention to address DD in adolescents with T1D. One of our instructors (KB) is a person with lived experience of T1D; having an instructor who is also a person with lived experience of the condition has been identified as a key component in previous studies using MSC interventions [56]. The longer-term follow-up of 12 months is also a clear strength. Throughout adolescence, youth are typically required to become more independent with their diabetes self-management while receiving diminished support from their caregivers. Mental health morbidity, including DD, is associated with poor glycemic control, diabetic ketoacidosis, and hypoglycemia in youth with T1D [7,65,66]. The MSC-T program will help adolescents with T1D develop skills to better cope with their emotions around managing their diabetes.

There are, however, some potential limitations to this study. The trial is single-blind; participants and the trial coordinator are not blinded to their assignment group. However, the statistician conducting the primary analysis will be blinded to participant allocation to the MSC-T intervention or waitlist control group. Given that the intervention is web-based, it does require the use of high-speed internet and a device that has webcam functionality, which may make it difficult for some youth to attend should these items be lacking. However, in Canada and the United States, over 96% [67] and 95% [68] of youth and emerging adults own smartphones, respectively, suggesting that if the program were to be scaled nationally or internationally, attending such a program would be feasible for the majority of North Americans.

Our proposed research has the potential to lower DD among youth with T1D by increasing their mindfulness-based self-compassion. We expect that our work will enhance the psychological well-being of youth with T1D, providing them with self-compassion skills to face adversity as they develop into young adults, ultimately leading them to live happier and healthier lives. If effective, further research should be done toward the implementation of MSC-T into standard clinical care for youth with T1D through partnerships with diabetes organizations.