This study will include a type 1 hybrid effectiveness implementation study to evaluate the effectiveness of the PROMOTE (peer support plus remote patient monitoring) program among Black adults with uncontrolled type 2 diabetes recruited from local primary care practices in Jefferson County, Alabama. Additionally, a mixed methods evaluation to characterize the contextual factors relevant to implementation of PROMOTE using Practical Robust Implementation Science Model (PRISM) will be conducted.

In the proposed study, the study team will conduct a type 1 hybrid effectiveness-implementation randomized controlled trial (RCT) of peer support plus remote patient monitoring as well as a mixed methods evaluation of contextual factors relevant to program delivery and implementation. Black adults with uncontrolled type 2 diabetes (A1C >8%) from primary care practices in Jefferson County, Alabama will be recruited to participate in this RCT.

Specific Aim 1: To evaluate the effectiveness of the PROMOTE (peer support plus remote patient monitoring) program among Black adults with uncontrolled type 2 diabetes in Jefferson County, Alabama. Hypothesis: The PROMOTE program will be effective, leading to clinically meaningful improvements in glycemic control.

A type 1 hybrid effectiveness-implementation RCT among Black adults with uncontrolled type 2 diabetes (A1C >8%) will test the PROMOTE program. The study team will be working with primary care practices in Jefferson County, Alabama through a consortium of community-based primary care providers who aim to improve healthcare delivery and outcomes by integrating research into their daily practice of medicine. Participants will be randomized 1:1 to intervention (PROMOTE) or control condition. Both study arms will receive diabetes self-management education and support (DSMES) delivered virtually. Participants randomized to the PROMOTE program will receive core components of peer support, remote patient monitoring of blood glucose, and integration with clinical care. Peer support will be delivered by trained ConnectionHealth CHWs with an initial in-person visit, followed by phone visits weekly for 3 months, then monthly for 3 months. ConnectionHealth is a 501c3 nonprofit organization that was founded in December 2014 in Birmingham, Alabama dedicated to training and deploying community health workers (CHWs) to address the needs of Alabama's most vulnerable citizens. Currently, ConnectionHealth serves citizens in Jefferson County, Shelby County, and the Black Belt Region of rural Alabama, providing community-based support with the goal of assisting clients to achieve healthier lives. The study team will partner with UAB eMedicine to deliver remote patient monitoring program. UAB eMedicine has an active remote patient monitoring program that is currently monitoring over 900 patients. All glucose values are transmitted via cellular enabled devices; the patients do not need access to their own internet plan or to their own smart device, which ensures that there are no disparities related to the technology. Furthermore, test strips are provided to patients on a regular basis by the program based on the number of readings performed, which eliminates a barrier of cost. Participants will be asked to check blood glucose with remote patient monitoring daily for 6 months, frequency tailored to their medication regimen. Monthly reports of peer support and remote patient monitoring activities will be provided to primary care providers. Clinical and patient-reported outcomes will include change from baseline to 6 months for the following: A1C (primary), as well as secondary outcomes of blood pressure, low density lipoprotein levels, body mass index, diabetes self-management, and diabetes distress. Clinical outcome will be extracted from the electronic health record and patient reported outcomes will be obtained via patient survey using validated measures. The main analysis will used linear mixed effects models to model change in each outcome (e.g. A1C) by study group (intervention vs. control arms), adjusted for the baseline value, accounting for any meaningful clustering using random intercepts. The hypotheses of an intervention effect will be tested by the significance of the coefficient for study group at the p<0.05 level.

Specific Aim 2: To characterize the contextual factors relevant to implementation of PROMOTE using Practical Robust Implementation Science Model (PRISM) through mixed methods evaluation.

PRISM is a context-oriented framework that aligns well with this study because it is intended to be used across program phases from pre-implementation, implementation, evaluation, and sustainment phases with a focus on the contextual domains closely tied to the outcomes of the RE-AIM framework (Reach Effectiveness Adoption Implementation and Maintenance). The study team will use purposeful recruitment to identify provider participants with a goal of 2-4 key informant provider participants per site with the following roles represented including physicians and advanced practice providers, nurses, care coordinators, and administrative staff. The study team will use parallel convergent mixed methods utilizing PRISM to assess perspectives and characteristics at the organizational and patient levels, as well as implementation and sustainability infrastructure and external environment, to determine key contextual factors that serve as implementation determinants.23 These results will be applied to determine future implementation strategies.

Inclusion Criteria:

• Age 19 years or older
• Black race (self-reported)
• Diagnosis of type 2 diabetes
• Uncontrolled A1C >8%
• Receipt of care at one of the study sites
• Speaks and reads in English

Exclusion Criteria:

• Limited English proficiency
• Currently pregnancy
• Non-community dwelling.
• Speaks and reads in English

De-identified Individual Participant Data for the following measures will be restricted with a data use agreement: 1) Biometric and lab data including hemoglobin A1c, height, weight, body mass index, blood pressure, remote monitoring of blood glucose; 2) Demographics and health status per NIH required common data elements; 3) Survey assessment: Diabetes Distress Scale, Patient Health Questionnaire, Diabetes Self-Management Questionnaire, Diabetes Self-Efficacy, and Social Support