Financial Incentives and Text Messaging to Improve African American Womens' Glycemic Control

Overview

Diabetes is a major public health problem that is associated with substantial health disparities for African American women compared to African American men or white men and women. This project will incorporate a) new technology whose use is high among African Americans (cell phone text message reminders), b) support from social networks, and c) strategies from the emerging field of behavioral economics to encourage and support self care behaviors that lead to improvements in diabetes control. In Stage I of the study, the investigators conducted focus groups with women with diabetes and their family members and friends. The investigators used information from these groups to work with a team of community partners to develop recruitment protocols, prepare study documents (including flyers, Frequently Asked Questions sheets, and informed consent forms), and finalize participant surveys. In the next phase of the study the investigators will evaluate ways that friends and family members can use direct support and text messages to improve control of blood sugar for African American women with type 2 diabetes. A team of three people--the Person With Diabetes (PWD) and 2 friends or family members, also known as the Support Team Members (STMs)--will be assigned at random to one of two groups to compare approaches to improving control of blood sugar. Each team will receive information and resources on improving diabetes control and will be linked by a text messaging program. The intervention team will receive additional text messaging support that may help them improve the PWD's diabetes control.

Control Condition. Control group participants will consist of the person with diabetes (PWD) and their two support team members (STMs). The PWD will receive regular medical care from their usual providers. They will also receive an introductory educational session and written materials on diabetes, the importance of its control, and diabetes self-care strategies, and a blood glucose and physical activity monitoring log. They will be invited for study visits, distinct from their routine clinic visits, at 3, 6, and 12 months to measure cholesterol, A1c (for those with diabetes), blood pressure, height, and weight measurement. Participants in the control condition will receive generic health promotion information during in-person visits or by mail. They will receive follow-up messages by text to remind them about study related events. Intervention Group. Participants in the intervention condition will consist of the PWD and their two STMs. The PWD will receive the same protocols that the "Usual Care" participants receive, including an introductory educational session, written materials on diabetes, a blood glucose log, and a physical activity monitoring log. The PWDs in the intervention group will receive training in action planning that is linked to their role self-identity. Role identity will be kept highly salient throughout the study by using identity primes that remind the individual that both processes (diabetes self-care behaviors) and goals (achieving A1c targets) are important responsibilities associated with that role. For example, participants may be asked to keep a picture or memento that reminds them of their role identity (e.g., a picture of a child or a photo of a place she would like to visit) on a blood sugar and physical activity log that the study will provide to the patient. The STMs in the intervention group will discuss these role identities and motivators with the PWD in their team intake session. They will also obtain similar training in action planning and in ways to support the PWD in order to better control her diabetes. Intervention team members will be reminded of the PWD's intrinsic motivations (IM), and how engaging in regular diabetes self-care can improve diabetes control, during interactions designed to boost the participants' confidence that they can accomplish and sustain the behavior change that contributes to interim and longer term goals. These interactions include text messages between the support team and the study team and between members of the support team. The text messages will include individually- and/or culturally-tailored messages, some of which are linked to the motivating factors identified through the questionnaire and probes. The text messages will take varying forms: (a) General diabetes messages derived from sources such as the ADA African American Programs and the National Diabetes Education Program; (b) Reminders about blood glucose monitoring, medication adherence, diet, and exercise, and how these behaviors contribute to maintaining role identity; (c) Prompts or queries about home glucose monitor readings and their frequencies that allow participants to communicate by text with study staff on non-urgent matters; and (d) Challenges and contests for which they receive supportive text messages. Face-to-Face Meetings will occur at the baseline and months 3, 6, and 12 to gather study data. For the intervention group, this will also include contact with a study team member who helps the intervention PWD and STMs identify and agree to desired behaviors and A1c targets. The study team member will review glucometer readings, medication refills, and A1c levels; encourage each PWD to engage in processes (e.g., self-monitoring or adhering to medications); encourage STMs in their supportive roles; congratulate those PWDs who near or reach A1c targets; and motivate those having difficulty reaching their targets. Baseline Protocols and Data Collection After randomization, all participants will be asked to provide verbal consent for the survey tailored to their study arm and their role as a PWD or STM, and will be administered a 35-45 minute baseline in-person survey in English by phone. There will be four different versions of the survey: Intervention PWD, Intervention STM, Control PWD, and Control STM. Following the administration of the baseline survey, all intervention participants will be allowed to select the message themes that they would like to receive from a list of options based on factors that they identified as intrinsic motivators and the intervals at which they would like to receive text messages (range 5 per week to 2 per day). The investigators plan to send an average of three messages per day, and elicit on average 1-2 responses per day. Additionally, the intervention PWDs will be told their target A1c goals and the incentives associated with specific drops in A1c. At this stage, the study staff will again discuss with the PWDs and STMs intervention groups the potential risks and harms of using inappropriate strategies to try to attain glycemic targets and, to obtain the monetary incentives. All enrolled patients will participate in a small group 30 minute session with study staff in the clinic to discuss diabetes self-care. All participants will also be invited to participate in ongoing diabetes classes and activities, and each person will be given a list of diabetes resources available in their clinic (e.g. group and individual diabetes education) and a calendar of local events and activities, such as the ADA Project Power diabetes classes in local churches, exercise classes, and healthy eating groups. Follow-up Protocols and Data Collection Once a month for the 6 month intervention period, participants in the intervention groups will receive a phone call from study staff to reinforce intrinsic motivation for improved glycemic control. At months 3, 6, and 12, participants (PWDs) in the control and intervention group will receive a reminder that the A1c test is due (via phone, mail, and text message). At months 3 and 6, the intervention team will also receive a message (via phone, mail, and text) about the deadline for an in-person incentive payment. In addition, intervention participants (PWDs) who attain target A1c levels will receive the appropriate monetary incentives based on their new A1c levels and through participation in the lottery. Compensation and incentives will be delivered at the time of testing. At the 6 month visit, all study subjects in both groups will be administered an abbreviated version of the baseline survey. At this point the financial incentives will be withdrawn and the IM, reminder, and generic text messages will stop for the intervention arm, who, like the control arm, will only receive reminder messages for clinic visits to get A1c tested (A1c testing will only be completed with PWDs in the control and intervention groups). At 12 months, all PWD and STM participants will complete a final survey. In the event the investigators are unable to reach the study participants, the investigators will obtain alternate contact information during study enrollment for all participants using a locator form. This form will be stored separate from the study participants' data. This will help to increase retention over the course of the study, and the investigators will update contact information regularly via the text messages. Once the 12-month study visit has been completed, the locator form will be destroyed. Assess the costs of the interventions In order to determine feasibility of importing the results of this study into public community clinics a costs analysis is planned at the end of the study. Costs fall into two main categories: program costs and costs of health services used by study participants. Program costs will be based on actual costs incurred by the study, including staff time and supplies, as well as the opportunity cost of time incurred by participants for attending quarterly study visits and participating in study phone calls. Costs related to personnel time will be measured as the product of personnel time spent and unit cost of labor and fringe benefits for the study staff. Personnel time spent for each study condition will be estimated from self-report. Costs related to use of resources (equipment and supplies, overhead) will be measured as the product of quantities used and unit cost derived from accounting data. The program developmental and training costs will be measured, but will be divided equally for the two intervention arms since identical protocols and provider trainings are required in each arm. Costs that are solely for research purpose, such as data management and analysis, will be excluded. The investigators will derive health services costs from survey data collected in the surveys from all study participants. These data include information on the number of outpatient medical visits, hospital stays, and Emergency Department visits. Using data in the published literature (e.g. using Medicare data), the investigators will map costs onto units of utilization, such as the cost for one physician office visit or one day in the hospital, thereby allowing the investigators to estimate total costs incurred for health services.