Mindfulness-Based College: Stage 1

Overview

Mindfulness interventions are increasingly offered to undergraduate students at universities world-wide, however the evidence base is very limited. The objective is to evaluate effects of a customized mindfulness intervention (called Mindfulness-Based College) on undergraduate student health. A superiority randomized controlled trial with parallel groups will be performed with 30 participants in each arm. Participants will be randomly assigned to Mindfulness-Based College or health education waitlist control. Investigators will be blinded to treatment allocation. Participants will be assessed at baseline, 10 weeks, and six months. The primary outcome is a college health summary score, including seven evidence-based determinants of health particularly relevant to college student well-being: body mass index, physical activity, diet, alcohol consumption, sleep quantity, perceived stress, and loneliness. Primary intention-to-treat analyses will evaluate whether MB-College vs. control is associated with the summary score, utilizing generalized linear models. Secondary analyses will evaluate which, if any, of the seven determinants of health are driving associations.

1. Background and Rationale Mindfulness interventions in school- and work-place settings are expanding exponentially, often with a limited evidence base. Increasingly, mindfulness interventions are customized to specific populations and health outcomes to increase intervention efficacy and efficiency. Emerging adulthood and the college undergraduate experience is a sensitive life course period, with strong peer effects, diminished parental influence, and compelling opportunities for exploring health behaviors having sustained effects, including social relationships, sleep patterns, alcohol and substance use, dietary patterns, and physical or sedentary activities.6 Few mindfulness interventions focus specifically on this population, but preliminary results are promising. Specifically, a randomized controlled trial (RCT) evaluating effects of the low time-burden Koru mindfulness intervention showed significant improvements in perceived stress, sleep, mindfulness and self-compassion immediately following the 4-week intervention, vs. waitlist control. An intensive mindfulness intervention RCT, customized to enhance physical activity, nutrition, sleep, mindfulness, compassion, relationships and well-being, demonstrated significant improvements in physical health, working memory, standardized test performance, mood, self-esteem, self-efficacy mindfulness and life satisfaction vs. control, at 6 weeks follow-up.8 We developed the moderate intensity Mindfulness-Based College (MB-College) intervention, which in a pre/post pilot study showed significant improvements in emotion regulation, diastolic blood pressure, stress, mindfulness and cognitive performance at 8 weeks follow-up (Table 1). Overall, early studies suggest potential benefit for college student health, however these studies require replication, longer term follow-up, and rigorous study designs. 2. Objective The primary objective is to perform an RCT to evaluate effects of MB-College vs. health education waitlist control, on a college undergraduate student health summary score at 6 months follow-up. 3. Methods Overall, a superiority randomized controlled trial with parallel groups will be performed with 40 participants per arm. Participants will be randomly assigned to MB-College or health education waitlist control. Investigators will be blinded to treatment allocation, including staff performing assessments, randomization, and data analyses. Participants will be assessed at baseline, 10 weeks, and 6 months follow-up. Standard CONSORT guidelines will be followed, including trial registration at ClinicalTrials.gov. The study population inclusion criteria are 18-28 years of age, current undergraduate student at any university, and English communication. Exclusion criteria follow standard recommendations, specifically current regular meditation practice, substance abuse, suicidal ideation, or history of bipolar or psychotic disorders or self-injurious behaviors. Participants will be recruited using print medium (e.g. recruitment cards and posters distributed throughout campuses), distributing advertisement graphics via social media, and advertisement emails sent to student listservs. The intervention is MB-College, which is an 8-week, 9-session curriculum providing systematic and intensive training in mindfulness meditation practices, applied to health behaviors relevant to college students. The curriculum (available upon request) is based on the manualized and standardized Mindfulness-Based Stress Reduction (MBSR) developed at the University of Massachusetts Medical School. The course builds a foundation of mindfulness self-regulation skills, including attention control, self-awareness and emotion regulation. It then directs those skills towards participants' relationships with health-related factors particularly salient in college undergraduates, including physical activity, diet, alcohol consumption, sleep, stress, social relationships, cognitive performance, and emotion regulation. Health behavior goal setting, and support for behavior change are integrated in the curriculum. The control condition is a health education wait list control, which participants will be offered a 30 minute, one-on-one meeting with the MB-College instructor. There, participants will learn about the MB-College curriculum, and have opportunities to share their relationship with common determinants of undergraduate student health and performance, described above. Together, the student and instructor will explore ways the course may assist in shifting these parts of their lives, as they see best. A deliberate relationship will be formed with the instructor. The primary outcome is a college health summary score, assessed using the following seven evidence-based determinants of health particularly relevant to college student well-being: body mass index (height and weight, directly assessed using standard epidemiologic methods); physical activity (MET minutes and step counts per week, using validated actigraphy and IPAQ - International Physical Activity Questionnaire); diet (mean daily fruit and vegetable consumption, utilizing validated Willet food frequency questionnaire); alcohol consumption (mean drinks per day, via standard questions from Behavioral Risk Factor Surveillance Survey); sleep quantity (mean sleep hours per night, using Pittsburgh Sleep Quality Index); perceived stress (validated Perceived Stress Scale score); and loneliness (validated R-UCLA \[University of California Los Angeles\] Loneliness Scale score). Specifically, analyses will assess changes in mean Z-scores of college health summary score, using a single mean health summary score across the seven aforementioned college health domains. Secondary analyses will evaluate which of the seven domains are most driving associations. Further primary analyses will determine if participant-identified health domains to focus on showed improvements in MB-College vs. control, utilizing the z-score approach described above, but restricting only to health domains identified by participants as having high readiness to change. Secondary outcomes, utilizing the National Institutes of Health (NIH) Science of Behavior Change Framework and NIH Stage Model for Intervention Development, include measures of self-regulation hypothesized to be proximal targets to the mindfulness intervention such as self-awareness (e.g. via validated Multidimensional Assessment of Interoceptive Awareness) and attention control (e.g. via validated Sustained Attention to Response Task and Mindful Attention Awareness Scale). Potential effect modifiers assessed include age, race, ethnicity, and socioeconomic status, as well as adverse childhood experiences utilizing the Adverse Childhood Experiences Questionnaire. All in-person assessments will take place at the Brown School of Public Health by trained research staff experienced in assessing all aforementioned measures. For the estimated timeline, please see Figure 1. Stratified randomization will be used for intervention allocation, performed using Research Randomizer (www.randomizer.org). Variables used to create strata include gender and race/ethnicity. Blinded, simple random sampling will occur within each strata. The analytic approach will evaluate whether MB-College vs. control is associated with college health summary score, described above. Analyses will incorporate generalized linear models (GLM) with properly chosen link functions, performed using generalized estimating equations (GEE) with robust standard error estimators. Following "intention-to-treat" principles, analyses will be conducted on all participants, regardless of intervention completion. Focus groups of participants will take place following MB-College, facilitated by Prof. Abigail Harrison, who is an experienced focus group moderator and researcher. The goal is to customize MB-College to the unique needs of this population. NVivo software (QSR International, Burlington, MA) will be utilized to analyze results in collaboration with Dr. Harrison.