Petit Cactus Study

Overview

The goal of this trial is to test a newly developed carbohydrate counting mobile application (named Petit Cactus) with photo recognition, for people with type 1 diabetes. To evaluate Petit Cactus, our objectives are: 1. To measure the association between app engagement and changes in blood glucose management over the same period. 2. To assess the usability of the app, focusing on engagement (i.e., frequency of meal logs) over 3 months and user satisfaction. 3. To validate the nutritional data obtained from the app's photo recognition algorithm by comparing it with a validated nutrient intake method (in this case, a web-based 24h recall). Participants will submit their Ambulatory Glucose Profile (i.e., CGM data) at the beginning and at the end of 3 months, download the app to use for carbohydrate counting for the 3-month duration, complete a usability questionnaire at the end of 3 months, and complete an end-of-study interview to report on their satisfaction with the app.

Design and description of methodology: A decentralized mixed-method proof of concept using a pre-post design. 1. Enrolment in the study: 1.1. Convenience sampling of 73 participants (aged 14 years and older) to use the Petit Cactus app for 3 months. 1.2. The link to the study will be distributed via Limesurvey to participants of the Canadian T1D BETTER registry as well as through social media and diabetes organization. 1.3. Upon reviewing the study objectives, procedures, and details of their participation on the link, potential participants will confirm eligibility, before being able to proceed with the electronic informed consent. 1.4. An invitation to sign their name, along with a clear explanation, stating that signing their name electronically is equal to a written signature, and is required to express consent. 1.5. Participants will be asked for their preferred method of communication, phone number or email, to be contacted by research personnel to schedule a call to confirm eligibility, review study methodology, and confirm enrolment in the study. 2. Demographic and diabetes management intake questionnaire: After completing the e-consent process and providing informed consent, participants will be invited to complete a demographic and diabetes management questionnaire on LimeSurvey. 2.1.1. Demographic variables: Age, self-identified gender, ethnicity, country of birth, highest level of education, work status, annual household income. 2.1.2. Diabetes management: Duration since diagnosis, method of insulin administration (multiple daily injections, insulin pumps, use of Automated Insulin Delivery; AID), type of CGM, A1c in the last 3 months (reported; continuous variable), reported number of serious (less than 3 mmol/L) and severe hypoglycemic (i.e., requiring external help) events in the last month. 2.1.3. Carbohydrate counting methods: a. Frequency of carb counting (at every meal and snack, at least once a day, at least once per week, less than once a week) b. Approximate duration of carbohydrate counting (in years; continuous variable) c. Use of apps to assist with carb counting (Yes/No) d. Reported perceived ability to perform CC: "I feel confident in my ability to determine the insulin dose given at mealtime based on the quantity of carbohydrates in my meals and/or snacks using an insulin-to-carb ratio" (Likert scale: Strongly disagree = 1; Strongly agree = 5). 2.1.4. Blood glucose management from CGM data (T=0): Estimated A1c, time-in-range, and hypoglycemic events at baseline. Participants will be invited to send us, through a personalized link to a One drive folder, their most recent Ambulatory Glucose Profile (AGP) report, which is an analysis generated using the last 14 days of glucose changes, accessible on their CGM device portal, to complete enrolment in the study. 3. Access to the mobile application: Petit Cactus for 3 months: participants will be enrolled in the study by receiving, through email: a. Link to download the app on their phone; from the App store if using an iPhone or Play store if using an Android phone device. b. Petit Cactus user guide PDF developed by our research team for research purposes. c. During the account creation process on the app, users will be informed that all identifiable data from the submitted demographic and diabetes management intake forms are kept confidential by the research team. Hence, participants have the option to refrain from disclosing information on age, date of birth, and gender in their profile settings at their own discretion. Pseudonyms can be used instead of real names if preferred. 4. Data collection on engagement throughout the trial duration The usability of the app will be assessed over 3 months through engagement with Petit Cactus, i.e., the frequency of meal logs in the app, using mobile app metrics (Google Analytics). 5. Completion of five (5) 24h recall using the validated R24W during the trial duration: To validate the app-based photo recognition algorithm, nutritional data (namely carbohydrates count) will be collected at 5 random days between T=0 and three-month T=3, to account for day-to-day variability in dietary intake. The days will encompass different contexts (weekday versus weekend). The data will be compared between a validated tool for nutritional intake and assessment, namely the R24W, and Petit Cactus. While participants complete the 24h recall on the R24W platform, their access to the app will be blocked until submission of the questionnaire. If participants do not complete the 24-hour recall within the allocated time, their access will be restored to the app, and the link to the R24W will expire. Participants will be contacted by phone or email to inquire on the reason for not completing the 24h recall and inform them that another invitation to the R24W platform will be sent out (the day will not be announced). A maximum of 6 invitations, where applicable, will be sent out to aim for 5 days of data collection. This process will be explained to the participants beforehand. To complete the R24W, participants will have to: 4.5.1.1. Click on the personalized link received by email at midnight to complete a 24-hour recall. Participants will have until midnight the following day to complete the questionnaire on their food and drink intake. No logins are required; answers are saved in real-time, and the questionnaire can be completed in multiple sessions, provided it's done before the end of the day. 4.5.1.2. Watch the mandatory 5-min tutorial on how to complete the 24h recall. 4.5.1.3. Answer the following: Date of birth, Sex at birth, Are you pregnant? Are you breastfeeding? Such information is important to contextualize the food intake. 4.5.1.4. Confirm the day for recalling their food and drink intake, e.g.: "You will need to indicate all the food and drinks you consumed on Thursday, April 25th, between midnight and midnight." 4.5.1.5. Complete the questionnaire by entering all meals consumed during the day: Type of meal, time of meal, context of meal i.e., location, alone/with company, screen time, manual food selection from database, portion size selection using visual cues, verification of meal entry, and supplementary intake: any food that could not be entered in the questionnaire, water intake, representativeness of everyday eating habits, other supplements taken in addition to food intake. 4.5.1.6. Confirm the reported intake and submit questionnaire. 4.5.1.7. Repeat for 4 other days. 6. Completion of an online satisfaction questionnaire with Petit Cactus At 3 months (end of the study), all participants will receive an online System Usability Scale (SUS) online questionnaire via email to gather quantitative feedback on their satisfaction with the app. In addition to the questionnaire, there will be an additional question on the reported perceived ability to perform CC: "I feel confident in my ability to determine the insulin dose given at mealtime based on the quantity of carbohydrates in my meals and/or snacks using an insulin-to-carb ratio" rated on a Likert scale from Strongly disagree (1) to Strongly agree (5). 7. Invitation to semi-structured individual interviews: After completing the online satisfaction questionnaire, a subset, randomly selected participants (20-28), will be invited to participate in individual interviews. Questions were derived from the SUS questionnaire to explore responses more deeply and facilitate triangulation of the results. 8. Blood glucose management from CGM data (T=3) To compare pre- and post-CGM data (i.e., estimated A1c, time-in-range, and hypoglycemic events) participants will be, once again, invited to send us their most recent Ambulatory Glucose Profile (AGP) report by email or download it directly into a personalized link to a One drive folder at the end of the trial (T=3).