Development and Pilot Testing of a Tailored Text Messaging Intervention to Improve Self-efficacy and Health Beliefs About Self-care in People with Heart Failure

Abstract

Introduction Self-efficacy and health beliefs are important modifiable factors influencing adherence to self-care behaviors in patients with heart failure (HF). Text messaging (TM) is a low-cost and scalable solution for delivering behavior change intervention and improving patient outcomes, yet it is largely unexplored in patients with HF. Hypotheses We developed a tailored TM intervention and conducted a pilot randomized control trial (RCT) to assess feasibility and preliminary efficacy in urban-dwelling patients with HF over 60 days. We hypothesized that patients receiving TM would report higher HF self-care efficacy, as well as fewer barriers and more benefits to medication, low-sodium, and self-monitoring compliance compared with patients receiving usual care. Methods We translated a validated, theory-based program of tailored Heart Messages into a Short Message Service TM intervention, and further enhanced its content using stakeholder input. We conducted a pilot RCT with two arms (TM intervention vs usual care). Eligibility criteria were medical diagnosis of HF (Stage C), access to a cell phone, and ability to read and speak English. Self-efficacy was assessed with the Self-Care Heart Failure Index (self-confidence subscale). Health Beliefs (perceived benefits and barriers) were assessed with the Beliefs about Medication, Diet and Self-monitoring Compliance Scales. Participants in the TM group received 3 individually tailored TM per week for 60 days based on their responses to the Health Belief scales at baseline and 30 days. Results Twenty-five patients (mean age 56±8.3, female=11, African American=23) were enrolled and completed the study (TM=11, control=14). Baseline characteristics were similar between groups. Although not reaching statistical significance, participants in the TM group reported better self-efficacy (mean difference (MD) 10.0, 95% Confidence Interval (CI): -2.21 to 22.21; Cohen's d=0.66, p=0.166), and more perceived benefits with medication (MD 1.88, 95% CI: -0.53 to 4.29; Cohen's d=0.64, p=0.15) and low-sodium diet compliance (MD 2.35, 95% CI: -1.20 to 5.90; Cohen's d=0.54, p=0.105). TM significantly improved perceived benefits of HF self-monitoring (MD 3.48, 95% CI: 0.40 to 6.56; Cohen's d=0.86, p=0.047) and trended to reduce perceived barriers (MD -1.32, 95% CI: -6.03 to 3.39; Cohen's d=-0.24, p=0.473). Compared with usual care, TM did not reduce barriers of medication (MD 0.64 95% CI: -2.46 to 3.75; Cohen's d=0.18, p=0.977) or low-sodium diet compliance (MD 1.37 95% CI: -0.56 to 3.30; Cohen's d=0.58, p=0.166). Satisfaction with the TM intervention was high (2.94±0.07; Low=1, High=3). Conclusions The TM program was feasible to deliver and was associated with beneficial trends. It is now ready for testing in intervention studies with larger sample sizes.