Development and Usability of REACH: A Tailored Theory-Based Text Messaging Intervention for Disadvantaged Adults With Type 2 Diabetes

Lyndsay A Nelson,Lindsay S Mayberry,Kenneth Wallston,Sunil Kripalani,Erin M Bergner,Chandra Y Osborn
DOI: https://doi.org/10.2196/humanfactors.6029
2016-09-08
JMIR Human Factors
Abstract:BACKGROUND: Among adults with type 2 diabetes mellitus (T2DM), adherence to recommended self-care activities is suboptimal, especially among racial and ethnic minorities with low income. Self-care nonadherence is associated with having worse glycemic control and diabetes complications. Text messaging interventions are improving the self-care of adults with T2DM, but few have been tested with disadvantaged populations.OBJECTIVE: To develop Rapid Education/Encouragement And Communications for Health (REACH), a tailored, text messaging intervention to support the self-care adherence of disadvantaged patients with T2DM, based on the Information-Motivation-Behavioral skills model. We then tested REACH's usability to make improvements before evaluating its effects.METHODS: We developed REACH's content and functionality using an empirical and theory-based approach, findings from a previously pilot-tested intervention, and the expertise of our interdisciplinary research team. We recruited 36 adults with T2DM from Federally Qualified Health Centers to participate in 1 of 3 rounds of usability testing. For 2 weeks, participants received daily text messages assessing and promoting self-care, including tailored messages addressing users' unique barriers to adherence, and weekly text messages with adherence feedback. We analyzed quantitative and qualitative user feedback and system-collected data to improve REACH.RESULTS: Participants were, on average, 52.4 (SD 9.5) years old, 56% (20/36) female, 63% (22/35) were a racial or ethnic minority, and 67% (22/33) had an income less than US $35,000. About half were taking insulin, and average hemoglobin A1c level was 8.2% (SD 2.2%). We identified issues (eg, user concerns with message phrasing, technical restrictions with responding to assessment messages) and made improvements between testing rounds. Overall, participants favorably rated the ease of understanding (mean 9.6, SD 0.7) and helpfulness (mean 9.3, SD 1.4) of self-care promoting text messages on a scale of 1-10, responded to 96% of assessment text messages, and rated the helpfulness of feedback text messages 8.5 (SD 2.7) on a scale of 1-10. User feedback led to refining our study enrollment process so that users understood the flexibility in message timing and that computers, not people, send the messages. Furthermore, research assistants' feedback on the enrollment process helped improve participants' engagement with study procedures.CONCLUSIONS: Testing technology-delivered interventions with disadvantaged adults revealed preferences and concerns unique to this population. Through iterative testing and multiple data sources, we identified and responded to users' intervention preferences, technical issues, and shortcomings in our research procedures.
What problem does this paper attempt to address?