An adaptive personalized text message intervention for cardiac prevention

点击翻译按钮获取中文摘要

The majority of patients at risk for heart disease are unable to adhere to key health behaviors, including regular physical activity and healthy eating. This is a major public health problem, as heart disease is the leading cause of death in the U.S., yet could be prevented in many cases through greater participation in these behaviors. Existing programs to promote health behaviors and prevent heart disease can be effective but have several limitations. First, they are costly, intensive, and attended by relatively few patients (with especially low attendance among minorities). In addition, these programs have not included positive psychology (PP) activities, which promote constructs (e.g., positive affect) that are independently linked to healthy behaviors and superior cardiac health. Finally, they are not designed to meet patients’ individual preferences and needs. Text message interventions (TMIs) may address many of the shortcomings of existing programs. TMIs are simple, low-burden, customizable, and highly accessible, and they have thus far shown promise in promoting health behaviors. Simple PP activities are well-suited to TMIs, and a TMI delivering both PP and health behavior messages could be a powerful approach to improving health and reducing cardiac risk, as combined psychological-behavioral interventions appear to have greater effects than either approach alone. A TMI that utilizes an adaptive algorithm to choose messages could be highly effective. As opposed to a single fixed set of messages, a flexible program that is deeply personalized for each individual could greatly improve engagement and impact. We have developed a novel adaptive TMI that delivers PP and health behavior text messages in English and Spanish. Based on a participant’s adherence gaps and feedback about the utility of prior messages, the program uses a dynamic, patient-specific algorithm to select new messages via multiple message-specific attributes (e.g., PP vs. health behavior-focused). Over time, the adaptive TMI increasingly delivers message content that matches the individual’s specific preferences and needs. Consistent with PA-18-389 focused on mHealth interventions for self-management, we propose a randomized pilot trial of this adaptive 12-week TMI (with daily text messages from the adaptive algorithm, twice-weekly goal-focused messages, and two brief phone check-ins to provide support) plus provision of a step counter. The TMI will be compared to enhanced usual care (step counter alone) in 60 patients with 2+ cardiac risk conditions (hypertension, type 2 diabetes, hyperlipidemia) and suboptimal physical activity and diet. The primary outcome of this R21 trial is feasibility and acceptability, measured via rates of text message feedback and participants’ ratings of message utility. As a key secondary aim, we will compare between-group differences in improvements in psychological, behavioral, and health-related outcomes at 12 and 24 weeks. If the intervention is feasible, well-accepted, and has promising effects on clinically relevant outcomes, we will conduct a larger next-step trial to assess the TMI’s impact on adherence and heart disease development.