Adaptive Internet-based Stress Management Among Adults With a Cardiovascular Disease: A Pilot Sequential Multiple Assignment Randomized Trial (SMART) Design

BACKGROUND: Stress has a negative impact on the experience of individuals with a cardiovascular disease-both directly (e.g., physiologic effects of stress may worsen cardiovascular disease) and indirectly (e.g., leads to higher perceived symptom severity). Therefore, stress management has become an extensively researched form of psychological therapy for these individuals, and is increasingly offered through the internet. Advantages of internet-based interventions include being time efficient (e.g., cut travel time), reducing waiting-lists, and increasing access for some sub-groups of the population (e.g., those living in rural and remote areas). Internet-based interventions also have the potential to integrate effective features of face-to-face interventions (e.g., tailored feedback, monitoring), but offer the scalability needed for public health interventions. However, up to 60% of individuals do not respond to these interventions, and for these individuals offering different types of support or increasing the intensity of the support provided is necessary. An innovative trial design to develop time-varying, adaptive interventions to maximize clinical effectiveness is the Sequential multiple assignment randomized trials (SMARTs). However, despite increasing popularity, SMARTs remain relatively new to intervention researchers.

GOAL AND OBJECTIVES: The goal of this pilot Sequential Multiple Assignment Randomized Trial (SMART) is to inform the planning of a subsequent larger SMART to evaluate an adaptive internet-based stress management program that follows a stepped-care model for adults with a cardiovascular disease. The primary objectives are to examine the (a) feasibility of the SMART procedures and different combinations of the stress management program (recruitment, retention, and questionnaire completion rates, reach, and fidelity) and (b) acceptability of the combinations of the stress management program, including adapting the program for non-responders and assessing adherence, satisfaction, and stress management strategies learned. The secondary objective is to estimate the program's effect size to inform sample size calculations for a full-scale SMART.

METHODOLOGY: 56 patients with a physician confirmed diagnosis of a cardiovascular disease will be recruited through clinic referrals and community-based study advertisement. Inclusion criteria are: (a) moderate stress (stress subscale score of 15 or more on Depression, Anxiety, and Stress Scale), (b) at least 18 years of age, (c) not participated in a stress management intervention in the past year, (e) regular access to a computer with internet and e-mail capabilities, and (f) reports understanding English or French. Eligibility will be confirmed mainly by self-report at the time of a screening interview. After completing their consent form and baseline questionnaire (T0), participants will be randomized to either: (a) a 6-week self-directed, web-based stress management program or (b) the same program plus weekly lay coaching. At the end of 6 weeks, intervention response will be assessed (T1 questionnaire), and non-responders will include those who have a stress score that did not improve (i.e., less than 50% decrease) or have a stress score above the cut-off point for mild stress (primary tailoring variable). In both groups, non-responders will then be randomized a second time to either (a) continue with their first stage program or (b) professionally-led motivational interviewing for another 6 weeks. During this time, responders will continue their first stage programs. The internet-based stress management program is available through my Health CheckUp. All participants will then complete their 12-13 weeks follow-up questionnaire (T2) to examine changes in stress and quality of life (primary outcomes), anxiety, depression, illness appraisal, self-efficacy, physical activity, and coping (secondary outcomes). Acceptability will be assessed by the satisfaction measure included in the follow-up questionnaire and further explored during an exit interview with participants. Feasibility measures will focus on the characteristics of the participants reached, protocol fidelity, percent of missing data, and recruitment and retention rates.

SIGNIFICANCE: The potential reach of internet-based interventions promises considerable public health impact. However, to maximize impact, the most effective interventions need to be developed and adapted to patients' evolving needs. This study will use the innovative SMART design to evaluate an adaptive internet-based intervention for individuals with a physical chronic illness.