Wearable Technology and a Virtual Lifestyle Program for Type 2 Diabetics

This study will assess the feasibility of integrating tailored aerobic exercise prescriptions and wearable activity trackers into a virtually delivered lifestyle medicine program in an outpatient clinic for adult patients with T2D.

This will be a 6-week single-group cohort study assessing the feasibility of delivering a lifestyle medicine program in a virtual setting with the incorporation of wearable activity trackers and personalized aerobic exercise prescriptions. STAND is a lifestyle medicine program that is currently run by the St. Joseph's Primary Care Diabetes Support Program (PCDSP). Participants will be new clinic, adult (aged ≤18 years old) patients with T2DM. Routinely, all new patients at the PCDSP are referred by their practitioner to a general 1-hour STAND informational class after their initial intake. Starting in June 2020, these classes will be offered virtually on the WebEx video platform. As such, participating patients must have access to technology that enables WebEx participation. At the end of this class, patients will be invited to participate in the study, STAND-VAT (Virtual Aerobic Therapy). If they agree, they will be contacted by a study investigator. Patients will be emailed the letter of information to review, along with copies of four brief surveys they will later be answering, to review. Once confirmed, the investigator will review the study details and answer any questions. Participants will authorize consent to participate through e-signature on REDCap. After consent, the study investigator will verbally go through the four short surveys (a brief demographics review, the PAID-5, the Stanford Self-Efficacy Scale, and a technology use survey). If patients do not wish to participate in the study, they will still be offered the opportunity to receive standard, biweekly virtual STAND classes delivered by SJHC staff (certified diabetes nurses, nurse practitioner, physician and dietician). These classes use motivational coaching and focus on practicing low carbohydrate nutrition, teaching patients to recognize glucose patterns using wearable flash glucose monitors (Libre Freestyle (Abbott Ltd., Canada)), and general advice and suggestion to get active, and reduce and break daily sitting.

The intervention, STAND-VAT will take the same format as STAND, but instead of general advice for exercise, patients will be provided with individualized exercise prescriptions and exercise goal setting/counselling via an exercise specialist (UWO School of Kinesiology MSc Student). This study will comprise her Master's thesis. The intervention program will also offer the opportunity to loan a wearable activity tracker to track steps and physical activity levels.

To ensure participants are medical safe to participate in exercise, after informed consent is obtained, two PCDSP staff and study co-investigators will review the patients electronic medical record to ensure inclusion criteria are met. A modified PAR-Q will be completed by a PCDSP clinician to ensure the patient does not suffer from unstable heart disease, vascular disease, etc. If they are not eligible, the PCDSP staff will inform them and work on optimizing their health as per routine clinical care. A functional fitness test (the 2-minute Step Test; number of steps completed during a 2-minute time period) will be completed in clinic at baseline after the participant has been medically cleared to exercise, but before the first group session. The test will be conducted one-on-one with the participant and the Master's student, but monitored by either a physician or nurse practitioner.

The virtual intervention STAND-VAT classes will occur biweekly and be co-led by a Certified Diabetes Educator and Master's Kinesiology student. The classes will flow based on patient questions, issues, or concerns; this format will not be didactic. The classes will review everyone's exercise prescription as created by the exercise specialist and deliver exercise progression recommendations. It will also include an opportunity for participants to reflect on exercise barriers and successes, and the resultant glucose patterns from adjusting diet and exercise (using data from their glucose monitors). One day prior to each class, participants will always be medically screened (by completing a modified PAR-Q) by a clinic nurse practitioner or physician for contraindications to exercise/participation.

Study participants will be given FitBit Inspire©'s to wear daily, using them to track daily step counts, exercise minutes and hourly movement goals (250 or more steps in an hour is a completed hour move goal). The aerobic exercise prescription will be individually tailored, adjusted bi-weekly and be prescribed as a daily step count goal. The daily step count goal will be progressed up to a maximum of 3,000 steps above baseline. Bi-weekly prescription adjustments will be prescribed using the mean or median daily step count from the past two weeks (whichever number is lower) plus an additional 500 steps. Participants will be strongly advised to exercise for a minimum of 10 minute bouts at least 3 times a day (or i.e. 30 minutes a day) at a brisk walking pace (Haskell et al., 2007) (utilizing the "talk test strategy" which elicits a moderate intensity (Reed \& Pipe, 2014)). Participants will be given instructions on how to use their FitBits© prior to the first virtual group class by the Master's Student.

Immediately before class, participants will receive a summary email of their new step count prescription. Since temporal landmarks have been shown to influence health behaviours (e.g., January 1, Mondays, etc.), referred to as the fresh start effect (Dai et al., 2014). It will be emphasized to participants in class and in the email that they will get to "start over" (get a 'clean slate') every two weeks, to increase motivation. This is timed at the same time the FreeStyle Libre patch needs replacing, and with a new step count prescription.

The week in between classes, the Master's student will call participants to ensure exercise adherence and resolve any issues. If a medical question or safety issue arises, the student will contact a PCDSP nurse practitioner or physician to follow-up with the patient. One day prior to the bi-weekly classes, participants will be required to sync and share FitBit Inspire HR© data to the FitBit© app, then sync/connect the app to the internet. Once synced, participants will log on to online desktop (https://www.fitbit.com) and export their weekly FitBit© data to a Microsoft Excel file. The exported file will be uploaded via the SJHC secure file transfer website, https://filesafe.lhsc.on.ca; the Master's student will review the step count information and create initial exercise step prescription Simultaneously, participants will also re-complete the modified PAR-Q (bi-weekly) to ensure continued medical safety.

Baseline demographics and a Technology Use survey will be completed at time of consent. The Stanford Self-Efficacy Scale, and the PAID-5 will also be completed over the phone with a study investigator at time of consent, and over video call within one week after week six (the end of the intervention). Outcomes related to step count data will be collected daily throughout the study by the FitBit Inspire HR©. The 2-minute Step Test will be completed in clinic at baseline and again immediately after the intervention. Glycemic control (measured by the Freestyle Libre) will be assessed using bi-weekly average time in glycemic target and estimated A1c (glycated hemoglobin, a clinical measure of longer-term glucose levels). These averages will be collected throughout the six-week intervention from flash glucose monitor data. The change in exercise volume will be measured using repeated, average daily step counts each week collected by the FitBit Inspire HR© during baseline, week one to week six.

Data analysis will performed on SPSS version 25 (SPSS Inc., Chicago, Illinois, USA). Recruitment levels will be defined as the number of people signed up over number of eligible people approached, presented as a percentage of sign up rate. Retention will be assessed as number of participants completing the post-intervention follow up divided by number of initial sample size, presented as a percentage retention rate. Acceptability data gathered from the exit survey will be analyzed in a qualitative or descriptive fashion, using frequencies and means (i.e. "this many rated the lottery draw as a motivation to get them to attend class as "strongly agree": 93%"). Additionally, number of reported technology issues will be reported as a total in addition to average mean # of technology issues per person. Counting will be used based on type of issue using categories like a) lost/disrupted internet connection or unable to connect or hear or see during class; b) issues utilizing FitBit/FitBit app c) Issues with uploading FitBit data; d) other.

Adherence will be measured by the percent of biweekly classes attended and check-in phone calls answered, number of data submissions (all out of three) as well as percent of days with FitBit worn (>500 steps) and average percent Libre sensor is active. Change in weekly average step counts and sedentary minutes will be analyzed using a one-way repeated measures ANOVA. Exercise prescription adherence will be calculated as percent of step goals achieved compared to total days participated in the study over the course of the six weeks. Weekly averages (over the six weeks) of estimated A1c, percent coefficient of variation, and "time in target" from the FreeStyle Libre data will be analyzed using a one-way repeated measures ANOVA. Changes of functional fitness levels from baseline to post-intervention will be compared using a paired t-test. The change in scores from baseline to week six for the Stanford Self-Efficacy Scale and PAID-5 will be evaluated using paired t-tests.