Diabetes and Depression Text Messaging Intervention (DIAMANTE)

The main aim of the "Diabetes and Mental Health Adaptive Notification Tracking and Evaluation" trial (DIAMANTE) is to test a smartphone intervention that generates adaptive messaging, learning from daily patient data to personalize the timing and type of text-messages. We will compare the adaptive content to 1. a static messaging intervention with health management and educational messages and 2. a control condition that receives a weekly mood message. The primary outcomes for this aim will be improvements in physical activity at 6-month follow-up defined by daily step counts.

We utilized user-centered design (UCD) methods to iteratively develop the DIAMANTE content and text messaging system through three iterative phases of UCD with ten patients each (total n=30). The first phase consisted of 1.5-hour individual semi-structured interviews. Findings from phase 1 were used to inform content and information delivery decisions of the final intervention, including selecting the thematic message categories and the design. In the second phase, patients tested out an early prototype of the mobile application through usability testing. Patients tested the final DIAMANTE intervention including thematic message content and the application in the third, final UCD phase, in order to address any user-related issues prior to launching the randomized control trial. In the DIAMANTE Randomized Controlled Trial, we aim to examine the effect of a smartphone app that uses reinforcement learning to predict the most effective messages for increasing physical activity. We will recruit 276 low-income minority patients with depression and diabetes within he San Francisco Health Network. We will compare this intervention to static messages with health management content, and a control group that only receives a weekly mood message.

This study is a randomized, controlled, one-center superiority trial with three-arm groups and a primary endpoint of increase in daily steps during a 6 month intervention by a smartphone app. Randomization will be performed as block randomization with a 1:1:1 allocation. Patients will be automatically randomized into groups trough our secure server during on-boarding of the app. Patients need to be informed of the nature and frequency of the messages they will be receiving and discuss this with investigators during the course of the study. Therefore, investigators and patients will not be blinded. We used the SPIRIT checklist when writing our report.

The investigators and those completing the statistical analyses will not be aware of the true study assignments

We will send patients a total of seven messages per week (one per day) at 10.00 am. For the physical health management messages we use messages from established topics in the Diabetes Prevention Program(23) content with the emphasis on physical activity and stress management. The final message, on the seventh day will ask patients to rate their mood on a scale from 1 to 9. Physical activity (step-count/day) will be passively monitored via the app on their smartphone.

Patients in the adaptive messaging arm will receive the daily messages of the static arm, and additionally receive daily messages within different categories of feedback and motivational messages that are chosen using a reinforcement learning (RL) algorithm. Physical activity (step-count/day) will be actively monitored via the app on their smartphone.

Control patients will only install the app on their phone and will not receive any feedback messages. They will receive one message a week, on a fixed day, asking them to assess their mood in the previous week on a scale of 1 to 9. The message will be sent daily at 10:00 am. Non-responders will receive reminders to submit their mood self-assessments in two hour intervals.

In a three arm randomized controlled trial we will examine the effect of a text-messaging smartphone application to encourage physical activity in low-income ethnic minority patients with comorbid diabetes and depression. The adaptive intervention group receives messages chosen by a reinforcement learning algorithm.

The static intervention group receives health information text-messages, typical of existing text-messaging interventions for diabetes and depression.

Our primary outcome, change in daily step counts, will be passively collected by a mobile phone application during the time that patients remain in the intervention.

We will derive HbA1c, the average plasma glucose over the previous eight to 12 weeks, recommended as a means to diagnose diabetes (20), from patients' electronic health records (EHR). We will use the most recent, available measurement from a maximum of 12 months before participating in the study. After 6 months, we will again assess the most recent HbA1c (pulling from patients EHR), ensuring that at least 3 months elapsed between baseline and follow-up HbA1c levels.

We will compare the self-reported PHQ-8 from medical records at baseline, intervention completion, and at the 6 month follow-up.

Inclusion Criteria: HbA1c (Hemoglobin A1c) > 7 PHQ-8 (Patient Health Questionnaire-8) > 5 Exclusion Criteria: high levels of physical activity (>30 minutes of moderate to vigorous activity per day) pregnancy inability to exercise due to physical disability serious mental illness unable to read and write in English or Spanish plans to leave the country for extended periods of time during the trial

We will submit study-results for publication in peer reviewed journals and presentation at (inter)national meetings, taking into account relevant reporting guidelines (e.g. CONSORT(32)). We will attempt to publish all findings in open-access journals when possible, or in other journals with a concurrent uploading of the manuscript content into PubMed Central for public access. Curated technical appendices, statistical code, and anonymized data will become freely available from the corresponding authors upon request.

Curated technical appendices, statistical code, and anonymized data will become freely available from the corresponding authors upon request.

Aguilera A, Schueller SM, Leykin Y. Daily mood ratings via text message as a proxy for clinic based depression assessment. J Affect Disord. 2015 Apr 1;175:471-4. doi: 10.1016/j.jad.2015.01.033. Epub 2015 Jan 29.

Lyles CR, Ratanawongsa N, Bolen SD, Samal L. mHealth and Health Information Technology Tools for Diverse Patients with Diabetes. J Diabetes Res. 2017;2017:1704917. doi: 10.1155/2017/1704917. Epub 2017 Feb 23. No abstract available.

Nouri SS, Avila-Garcia P, Cemballi AG, Sarkar U, Aguilera A, Lyles CR. Assessing Mobile Phone Digital Literacy and Engagement in User-Centered Design in a Diverse, Safety-Net Population: Mixed Methods Study. JMIR Mhealth Uhealth. 2019 Aug 29;7(8):e14250. doi: 10.2196/14250.

Avila-Garcia P, Hernandez-Ramos R, Nouri SS, Cemballi A, Sarkar U, Lyles CR, Aguilera A. Engaging users in the design of an mHealth, text message-based intervention to increase physical activity at a safety-net health care system. JAMIA Open. 2019 Oct 11;2(4):489-497. doi: 10.1093/jamiaopen/ooz049. eCollection 2019 Dec.

Hernandez-Ramos R, Aguilera A, Garcia F, Miramontes-Gomez J, Pathak LE, Figueroa CA, Lyles CR. Conducting Internet-Based Visits for Onboarding Populations With Limited Digital Literacy to an mHealth Intervention: Development of a Patient-Centered Approach. JMIR Form Res. 2021 Apr 29;5(4):e25299. doi: 10.2196/25299.

Figueroa CA, Aguilera A, Chakraborty B, Modiri A, Aggarwal J, Deliu N, Sarkar U, Jay Williams J, Lyles CR. Adaptive learning algorithms to optimize mobile applications for behavioral health: guidelines for design decisions. J Am Med Inform Assoc. 2021 Jun 12;28(6):1225-1234. doi: 10.1093/jamia/ocab001.

Aguilera A, Figueroa CA, Hernandez-Ramos R, Sarkar U, Cemballi A, Gomez-Pathak L, Miramontes J, Yom-Tov E, Chakraborty B, Yan X, Xu J, Modiri A, Aggarwal J, Jay Williams J, Lyles CR. mHealth app using machine learning to increase physical activity in diabetes and depression: clinical trial protocol for the DIAMANTE Study. BMJ Open. 2020 Aug 20;10(8):e034723. doi: 10.1136/bmjopen-2019-034723. Erratum In: BMJ Open. 2023 Jul 11;13(7):e034723corr1.

Figueroa CA, Deliu N, Chakraborty B, Modiri A, Xu J, Aggarwal J, Jay Williams J, Lyles C, Aguilera A. Daily Motivational Text Messages to Promote Physical Activity in University Students: Results From a Microrandomized Trial. Ann Behav Med. 2022 Feb 11;56(2):212-218. doi: 10.1093/abm/kaab028.