Assessing the Impact of Control-IQ Technology

Background Information \& Rationale

The management of type 1 diabetes (T1D) is rapidly evolving to offer technology that aims to improve metabolic control and decrease the burden of diabetes management. Hybrid-closed loop systems are at the forefront of this technology. They are comprised of an insulin pump and continuous glucose monitoring (CGM) system that interface in order to automatically adjust the delivery of basal insulin through the pump, based on CGM data. Insulin boluses for meals and snacks must still be entered manually by the user.

Studies of hybrid closed loop systems have primarily focused on clinical trials that show improved glycemic outcomes.1-3 Few studies have assessed real-world glycemic outcomes or patient-reported outcomes, and none have focused on a pediatric population.

A recent study published by Pinsker et al.4 showed improvement in psychosocial outcomes and persistent achievement of glycemic targets in a real-world setting in adults with T1D using the Tandem t:slim X2 pump with Control-IQ technology.

The Tandem t:slim X2 pump with Control-IQ is a hybrid closed loop system (using Dexcom G6 CGM) that has recently been approved for use by Health Canada, but has been used in the U.S. for the last year. This new technology will launch in Canada in March 2021, at which time existing Tandem t:slim X2 insulin pump users (with concurrent CGM use) will be able to upgrade their insulin pump software to access this technology (following review of an online module).

The Children's Hospital of Eastern Ontario (CHEO), is located in Ottawa, ON Canada. CHEO follows approximately 800 children with diabetes, approximately 85 of whom are managed on the Tandem t:slim X2 pump with concurrent Dexcom G6 CGM use.

The upcoming launch of Control-IQ presents a unique opportunity to gain a better understanding of the real-world impact of this technology on our Canadian pediatric population. The benefits of conducting this study in our population are two-fold:

1. The Canadian context offers the study a diverse patient population given our universal healthcare system and provincially-funded insulin pump program. Because pumps and supplies are funded by the government, they are available to all of our patients, regardless of socio-economic status and whether or not they have private health insurance.
2. Studying a pediatric population offers the unique perspective of children and youth (and their parents) whose management practices and goals differ from those of adults. While adult patients and parents of children with T1D are motivated by attaining glycemic targets for long term health outcomes, children and youth live in moment with emphasis on how they feel now. As such, patient reported outcomes, while important to all ages, are often the most important motivator for children and youth when choosing a treatment regimen.

Study Design This is a single-centre, mixed-methods, prospective study in pediatric patients with T1D initiating Control-IQ technology on the Tandem t-slim X2 insulin pump.

Participants will enroll in the study prior to initiating use of Control-IQ on their Tandem insulin pump.

Recruitment will occur over an eight-week period starting mid-March 2021. Participation in this study will last 16 weeks.

Study Objectives

Primary Objective: To determine pediatric T1D patients' and their parents' perceptions of the impact of Control-IQ on their psychosocial functioning and quality of life (as measured by the INSPIRE Questionnaire, post assessment).

Secondary Objectives:

1. To assess the effect of Control-IQ technology on other patient and parent reported outcomes (PROs) including:

• Diabetes Impact and Device Satisfaction (DIDS) scale,
• Hypoglycemia Fear Survey (child and parent version),
• WHO-5 (emotional well-being)

1. To determine the impact of Control-IQ technology on glycemic control in pediatric T1D patients who are experienced sensor-augmented pump users.

Exploratory Objectives:

1. To explore the effect of Control-IQ technology on patient and parents' sleep, school, activities/ sports, mood and family dynamics.
2. To explore whether or not youth and parents' expectations of Control-IQ align with their experience of using this this automated insulin delivery system.

Participant Characteristics:

The following data will be collected from each chart at baseline:

• Date of birth
• Gender
• Date of T1D diagnosis
• Date of initiation of insulin pump therapy
• Date of first insulin pump start
• Date of Tandem t:slim X2 pump start
• Current use of CGM (yes/ no)
• Last measured HbA1c ( if available within the last 3 months
• Number of hospitalizations due to diabetes complications within the past year

Sociodemographic measures will be collected by questionnaire at baseline, including:

• Family structure
• Parental education - highest level achieved
• Household income
• Private health insurance
• Ethnicity (self-reported)

Statistical Plan The patient population will be described using descriptive statistics. The study team will compare patient reported outcomes, measures of glycemic control, and measures of insulin pump management pre and 16 weeks post-initiation of Control-IQ technology, using paired t-test and Wilcoxon signed-rank test, depending on the distributional characteristics of the outcomes. Analysis of open-ended questions will be done by a qualitative research specialist.

Sample Size Determination CHEO currently has 85 patients on the Tandem t:slim X2 pump with CGM, with an estimated 10-15 patients more estimated to start before March 2021. Historically, the patient population at CHEO has been highly engaged with research with an anticipated participation rate of 90-95%. The Investigator expects that recruitment rates would be similar for this study, allowing for a sample size of 70 participants.

This sample size is sufficient to provide > 80% power for detecting a "medium size" effect (Cohen's d = 0.5), which corresponds to a mean score of 70 from the post assessment INSPIRE Questionnaire, 10 units larger than a mean score of 60 (the null hypothesis). Cohen's d is expressed by the ratio of the mean difference over the standard deviation. Here, we use sd = 20 as an estimate of the standard deviation of the score, which is conservative considering previously presented estimates (e.g. sd = 16.44 described in Weissberg-Benchell et al [5]).