The Happy Heels Study: Testing Whether Positive Exercises .

4m ago
1.08 MB
39 Pages
Last View : 4m ago
Last Download : n/a
Upload by : Kaydence Vann

Running head: THE HAPPY HEELS STUDYThe Happy Heels Study: Testing Whether Positive Exercises Improve Mental Health, Happinessand Academics in First-Year StudentsRonald Harris JrSenior Honors ThesisFaculty Advisor: Dr. Barbara FredricksonDepartment of Psychology and NeuroscienceUniversity of North Carolina at Chapel Hill

THE HAPPY HEELS STUDYAbstractCollege students face increasing amounts of stress related to navigating and thriving atuniversities. With increasing levels of anxiety, depressive symptoms and the strain placed oncollege counseling and psychological services, new creative ways are needed to assist studentswith their mental health and well-being. Three Good Things and a variant that focuses on socialconnection, are two exercises shown to improve mental and physical well-being, as well asreduce loneliness. 160 UNC first-year and transfer students participated in a two-weekintervention where they performed Three Good Things (3GT) or an adapted PositivityResonance exercise (3GT-Social). Results showed that while no significant changes in academicperformance or visits to counseling services emerged, students did report significantimprovements in loneliness, positivity resonance with strangers and acquaintances, anddepressive symptoms, with between-group differences approaching significance at one follow-uptime point. The two exercises offer a novel approach to addressing the mental health of manycollege students while also at little costs to university resources.

THE HAPPY HEELS STUDYThe Happy Heels Study: Testing Whether Positive Exercises Improve Mental Health, Happinessand Academics in First-year University StudentsUndergraduate college students are facing higher levels of stress and anxiety than everbefore. From adjusting to college life, classes, exams, making friends, and pressures to succeed,more students are finding it difficult to thrive in the college environment. In 2013, the AmericanCollege Health Association National College Health Assessment found that 61% of students, in asample of over 63,000, had experienced overwhelming anxiety in the previous year. Fifty-threepercent of students reported academics being “very difficult or even traumatic” to handle(American College Health Association, 2013). Since 2013, all measures of mental health forstudents on the assessment have worsened. In the spring of 2018, 55% of students felt hopeless,88.1% felt overwhelmed by what they had to do, 84.6% felt exhausted, 64.4% felt lonely, 69.9%felt very sad, 42.9% felt so depressed it was difficult to function, 64.3% felt overwhelminganxiety, 13% seriously considered suicide, 8.5% committed self-harm, and 1.9% attemptedsuicide (American College Health Association, 2018).Rates of depression in undergraduates have risen concurrently with anxiety (AmericanCollege Health Association, 2013). Over time, more students have begun to seek out mentalhealth and psychological services. There has been a 30% increase in the number of studentsgoing to counseling and psychological services between 2009 and 2015 (Center for CollegiateMental Health, 2016). This trend may point to society becoming more aware of the necessity formental health resources and reduced stigma behind psychological services. However, theincrease in students has overwhelmed university counseling services across the country. Alreadywith limited resources, universities have resorted to a plethora of methods to accommodate the

THE HAPPY HEELS STUDYgrowing number of students. Many have hired additional staff, partnered with private clinics andtherapists, resorted to triage-like methods to reduce appointment wait-times, and assignedclinicians for same-day care (Reilly, 2018). Despite these efforts, the volume of students is stillstaggering, which results in longer wait-times for appointments and the inevitability that theseservices will not get to every student in a timely manner. The students missed may end updeveloping serious mental health complications, needing to withdraw from school or evenhospitalized from self-harm or suicide attempts.Mental Health at UNC-Chapel HillThe University of North Carolina at Chapel Hill has exhibited similar trends in mentalhealth compared to the rest of the nation. An increasing number of students face mental healthchallenges during their undergraduate career and more students are seeking out and utilizingCounseling and Psychological Services (CAPS) on campus. Since 2012, the number of studentsgoing to CAPS, for all appointment types (medical evaluation, medical check, telephone/webappointments, academic interventions, triage, brief therapy, and urgent crisis) has increased(Counseling and Psychological Services Appointment Statistics, 2018). The exhaustive reportsfrom CAPS are provided in Appendix A. Notably, the rise in the number of academicinterventions, urgent crises, and triage appointments have more than doubled between the 20122013 and 2016-2017 academic year ( 113%, 104%, 101% increase respectively) (Counselingand Psychological Services Appointment Statistics, 2018). Between July 1 st, 2017 and June 30th,2018, CAPS had 18,805 attended appointments and saw a total of 4,097 clients (Counseling andPsychological Services Appointment Statistics, 2018) or roughly equivalent to an entireundergraduate first-year class seeking mental health services over the course of a year.

THE HAPPY HEELS STUDYPersonal conversations with Carolina Counseling and Psychological Services (CAPS),academic advisors, and admissions staff have corroborated the increases in symptoms of anxietyand depression in students. Staff members at CAPS have noted that making friends on campus,adjusting to college life, imposter syndrome, homesickness, and social anxiety are amongcommon concerns of students. One academic advisor noted that students have anxiety due tofamilial pressure to perform at the perceived standard of what is considered academically“normal” at UNC (D. Robinson, personal communication, September 2018). There is sufficientevidence to suggest that more students are facing challenges while on campus and though moreare seeking help, university resources struggle to assist such a high volume of students.Improving Mental Health in Healthcare Workers (HCW)Students are not the only group to face increasing levels of stress, anxiety, depressivesymptoms, and difficulties balancing their lives. In the field of medicine, healthcare workers arefacing longer hours, increased patient volume, and increased regulation. The rise in demands andexpectations have contributed to the increased prevalence of burnout in physicians (Bakker andDemerouti, 2017). Burnout is a combination of emotional exhaustion, lower productivity at workand over-depersonalization (Perlman and Hartman, 1982). As of 2014, 55% of US physicianshad one or more symptoms of burnout (Shanafelt, T. D., Hasan, O., Dyrbye, L. N., Sinsky, C.,Satele, D., Sloan, J., and West, C. P., 2015). This is a concerning increase given that burnout anddepression, among other stress-related symptoms, are linked to increased medical errors and as aresult, put patients at risk (Shanafelt et al., 2010) (Welp, A., Meier, L. L., and Manser, T. 2015).There has also been a need in medicine to provide cost-effective ways of reducing stress inHCW.

THE HAPPY HEELS STUDYA particular psychological exercise has had documented success in alleviating burnout,depression, improving happiness and work-life balance in HCW. The Three Good Things (3GT)exercise is a short intervention that has been shown to reduce depressive symptoms, burnout, andincrease happiness for six months (Seligman et al., 2005). 3GT is a short exercise in whichparticipants write three things that went well during their day. They are typically prompted withthe question “What went well today?” or slight variations of it. A recent study done at the DukePatient Safety Center in Duke Hospital found that healthcare workers who performed 3GT for 15days had significant improvements in burnout, depressive symptoms, happiness and work-lifebalance six months to a year following the intervention (Sexton & Adair, 2017). The success ofthis exercise in healthcare demonstrates that small interventions can have long-lasting impacts onthe mental health of participants. Additional cohorts of healthcare workers have partaken in 3GTwith similar results, and the Patient Safety Office continues its research into 3GT’s efficacy.Multiple units at Duke Hospital have seen improvements in their staff after partaking in theintervention.Because 3GT has been shown to have marked success as a cognitive exercise and as partof a concrete intervention for hospital personnel, the present research proposes that 3GT canhave similar effects for undergraduate students. Both populations, healthcare workers andundergraduate students, have experienced increases in anxiety, depressive symptoms, burnoutand difficulties balancing work and home in recent years. Both universities and hospitals arecurrently ill-equipped to handle the increase in those who may be seeking mental health services.Consequently, 3GT may be a useful tool for undergraduate students to use, in the same way thehealthcare workers have utilized it in the hospital.

THE HAPPY HEELS STUDYPositivity Resonance Improves Mental HealthResearch has shown that positive emotions affect a person’s mental and physical health.In particular, shared positive emotions are integral for general well-being, social interaction andhealth (Fredrickson, 2016). Based on this research, Dr. Barbara Fredrickson proposed positivityresonance theory, which states that these shared positive interactions can result in increases inpsychological well-being, mental flourishing, and lower levels of loneliness (Major, Lundberg,& Fredrickson 2019). Positivity resonance is a brief experience that occurs when two or moreindividuals share an interpersonal connection with: shared positive affect, mutual care andconcern, and behavioral or biological synchrony (Major et al., 2019). A recent study by Majorand colleagues revealed these findings in a cohort of 81 undergraduate students at UNC-ChapelHill. Students in the treatment condition underwent a two-week social intervention, where theywere encouraged to create more positive interpersonal connections with others (Major et al.,2019). Because the study demonstrated that behaviors promoting positivity resonance canimprove well-being in students, it is hypothesized that first-year students who experience morepositivity resonance during their first semester will have higher levels of happiness, flourishingand general well-being.Current StudyRecent data and research show that both universities and hospitals are currently illequipped to handle the increase in individuals seeking mental health services. Both healthcareworkers and undergraduate students have experienced increases in anxiety, depressivesymptoms, burnout and difficulties balancing work and home. However, two simple, freeinterventions have provided measurable improvements to many mental health issues inindividuals. 3GT displays impressive success as a cognitive reappraisal exercise and as part of a

THE HAPPY HEELS STUDYconcrete intervention for healthcare workers. As a result, it may be a useful tool forundergraduate students to use, in the same way the healthcare workers have utilized it in thehospital. In addition, in Fredrickson, Major and Lundberg’s study with undergraduates, thosestudents who cultivated more positivity resonance not only showed lower levels of lonelinessfollowing the two-week intervention, but also displayed higher levels of flourishing mentalhealth and psychological well-being. This positivity resonance exercise follows the same twoweek duration as 3GT. As a result, in the present study, the exercise was adapted to mimic thedaily reminders of 3GT, while also encouraging students to focus specifically on cultivatingmoments of positivity resonance with others. In maintaining the format of 3GT, participantswere asked “What social interactions went well today?” While slightly different from theoriginal positivity resonance exercise, this offers a more direct comparison between 3GT andpositivity resonance. The new condition is termed 3GT-Social throughout the study.The present study aimed to investigate if 3GT and 3GT-Social would have positiveeffects on first-year students embarking on their first year of college. It tested whether either ofthe exercises would improve the overall mental health of first-years and also affect academicperformance. Improvements in students within the 3GT group stood to increase thegeneralizability of the 3GT exercise to populations outside of healthcare. Improvements instudents within the 3GT-Social group would further test whether increases in positivityresonance lead to better health and well-being.Both exercises were tested to examine their effects on a variety of mental states observedwithin the realm of a college campus. They included: burnout, depressive symptoms, anxiety,school-life balance, subjective happiness, gratitude, satisfaction with life, loneliness, positiveempathy, social curiosity, sense of belonging and perceived positivity resonance.

THE HAPPY HEELS STUDYIt was hypothesized that students in the 3GT and 3GT-Social conditions, in comparisonto an active control group, would show improved scores on a core set of measures representingmental health: burnout, depressive symptoms, school-life balance, subjective happiness andanxiety, at the conclusion of the semester. It was also hypothesized that these improvementswould have observable effects on students’ academic performance as well—manifesting in ahigher cumulative GPA for the fall semester. Thus, students in the treatment conditions wouldhave a higher cumulative GPA than students in the active control group. Finally, it washypothesized that the two treatment groups would also have significantly fewer visits to CAPSthan those in the active control condition.These exercises may provide incoming first-year undergraduate students with a new wayto cope with the stress of adjusting to college life in the same way that healthcare professionalshave used 3GT in the hospital. Given that the exercises are free and require minimal time tocomplete, they stand to serve as an additional tool for university counseling services to use withstudents who require mental health support. They have the potential to be used as large-scaleinterventions by on-campus counseling services.MethodsParticipantsParticipants were recruited via email through the Office of New Student and FamilyPrograms listserv at The University of North Carolina at Chapel Hill. Eligible participants were18 and older and classified as first-year or transfer students. Those providing consent (N 206)were directed to complete a pre-survey of mental health measures, as well as to provide theiremail (and optional phone number) in order to be contacted during the study. Participants were

THE HAPPY HEELS STUDYexcluded if they were not 18 years of age, were not a first-year or transfer student, did notconsent, or did not complete the pre-survey in its entirety (see CONSORT diagram in Figure 1).Following these exclusion criteria, the remaining participants (N 160) were randomized into anactive control (n 53) or one of two treatment groups: Three Good things (3GT; n 54) orPositivity Resonance (3GT-Social; n 53). The average age of participants was (M 18.5; SD 2.31). The final sample contained 150 first-year students and 10 transfer students (see Table 1).130 out of 160 students in the final sample consented to allow their fall cumulative GPA to beaccessed by the researchers in the spring. Participants were incentivized to complete the dailysurveys by having the option to receive a customized study-specific stress ball, free of charge, atthe beginning of the study, as well as be entered into a random drawing for 1 of 20 50 Visa giftcards. Drawings for gift cards were done at random among participants who consented andprovided an email to receive daily reminders throughout the study. Drawings occurred followingthe last follow-up survey in January. Stress balls were delivered to students via the on-campuspackaging centers. Students living off campus were able to pick up their stress ball from DavieHall at their earliest convenience.MeasuresMeasures of burnout, depressive symptoms, anxiety, school-life balance, subjectivehappiness, gratitude, satisfaction with life, loneliness, positive empathy, social curiosity, sense ofbelonging, and positivity resonance were administered to participants before, during, and aftercompleting the intervention period. The intervention period was for two weeks (15 days) duringthe fall semester in November of 2018. Fall semester GPA’s of consenting participants (n 130)were collected from the university registrar’s office in January 2019.

THE HAPPY HEELS STUDYAnxiety & Depressive symptoms: To assess anxiety and depressive symptoms instudents, short forms of the Patient-Reported Outcomes and Measurement System (PROMIS )items for anxiety and depression were used (Pilkonis et al., 2011). Participants responded toeight and seven items respectively for depressive symptoms and anxiety, indicating thefrequency that they experienced each item on a 5-level ordinal scale (never, rarely, sometimes,often, always). Example items included: “I felt worthless” (depressive symptoms) or “I feltworried” (anxiety). Cronbach’s alpha for anxiety and depressive symptoms at the pre-surveysurvey for the current sample were .892. and .930 respectively.Burnout: The Maslach Burnout Inventory is the most widely used instrument to assessburnout. A 5-item variant of the 9-item scale was used to measure emotional exhaustion inparticipants (Schaufeli, Bakker, Hoogduin, Schaap, & Kladler, 2001). Participants responded ona scale of 1-5 (1 strongly disagree; 5 strongly agree) to items such as “I feel burned out frommy work”. Cronbach’s alpha at pre-survey for the current sample was .802Gratitude: Gratitude/grateful disposition was assessed using the GratitudeQuestionnaire—6 (McCullough, Emmons, and Tsang, 2002). Participants responded to six itemson a seven-point scale (1 strongly disagree to 7 strongly agree). Example items included: “Ihave so much in my life to be thankful for” and “I am grateful to a wide variety of people”.Cronbach’s alpha at pre-survey for the current sample was .770.Loneliness: Loneliness was assessed with the UCLA Loneliness Scale (version 3)(Russell, 1996). Participants responded on a five-point scale (1 never to 5 always) to 20items regarding their sentiments over the previous week. Example items included: “How oftendo you feel left out?” and “How often do you feel outgoing and friendly?” Cronbach’s alpha atpre-survey for the current sample was .927.

THE HAPPY HEELS STUDYSchool-life balance: Students were assessed on how well they balanced various aspectsof college life using the College Activities and Behaviors Questionnaire (CABQ) (Pennebaker,2013). Students indicated the number of times they had experienced each item over the pastweek. Example items include: “Number of times had difficulty falling asleep ” and “Numberof times exercised strenuously ”. Raw scores were aggregated into four larger categories:unhealthy behaviors, unhealthy consumption, self-care and social. Final scores were computedby subtracting the sum of the unhealthy habits (consumption and behaviors) from the positiveones (self-care and social).Sense of belonging: Students’ sense of belonging while on campus was measured usingthe 12-item General Belongingness Scale (GBS) (Malone et al., 2012). Participants responded ona seven-point Likert scale to items such as: “When I am with other people, I feel included” and“I feel accepted by others”. Cronbach’s alpha at pre-survey for the current sample was .926Subjective Happiness: Measures of participants’ subjective happiness were done usingthe Lyubomirsky and Lepper’s subjective happiness scale (SHS) (Lin J-D, Lin P-Y and Wu CL., 2010; Sexton and Adair, 2017; Lyubomirsky and Lepper, 1999). Participants responded toitems such as “In general I consider myself ” (1 not a very happy person to 7 a very happyperson). Cronbach’s alpha at pre-survey for the current sample was .866Perceived Positivity Resonance: Positivity resonance was measured using a 7-item scaledeveloped by Major et al., 2018. Participants were asked: “Considering the single longestinteraction you had with one or more strangers or acquaintances this week. For what proportionof time during this episode (from 0 to 100 percent) ” to measure positivity resonance withstrangers and acquaintances and “Considering the single longest interaction you had with one ormore close friends or romantic partner this week. For what proportion of time during this episode

THE HAPPY HEELS STUDY(from 0 to 100 percent).” for positivity resonance with close friends or romantic partners.Participants indicated how much they experienced each item as a percentage ranging from 0-100percent. Responses from each item were averaged to make an individual score. PositivityResonance for romantic partners and close friends was denoted as PosRes RC. PositivityResonance for strangers/acquaintances was denoted as PosRes SA. Cronbach’s alpha at the presurvey for the current sample were .962 for PosRes SA and .978 for PosRes RC.Social Curiosity: Social curiosity was measured using the 10-item Social Curiosity Scale(SCS) (Renner, 2006). Participants responded to items such as “I like to learn about the habits ofothers” and “When people quarrel, I like to know what’s going on” on a 5-point Likert scale (1 Does not describe me to 5 Describes me extremely well). Cronbach’s alpha at pre-survey forthe current sample was .760.Positive Empathy: Positive empathy was measured using the Positive Empathy Scale(Morelli, Lieberman, & Zaki, 2015). Participants responded on a 5-point Likert scale to itemssuch as “When someone is else is enthusiastic, I can’t help but be enthusiastic too”. Cronbach’salpha at pre-survey for the current sample was .853.Satisfaction with Life: Satisfaction with life was measured using the RiversideSatisfaction with Life Scale (Margolis, Schwitzgebel, Ozer, & Lyubomirsky, 2018). Participantsresponded on a 7-point Likert scale (1 Strongly Disagree to 7 Strongly Agree) to items such as“I like how my life is going” and “I am content with my life.” Cronbach’s alpha at pre-survey forthe current sample was .836.DesignAll study procedures were approved by the University of North Carolina at Chapel Hill’sInstitutional Review Board. Potential participants were recruited through email and directed to a

THE HAPPY HEELS STUDYQualtrics survey containing the consent form, option to receive a stress ball as incentive, and thepre-survey. In the consent form, students could choose to: consent to the study and share theirfall semester GPA at the end of the study; consent to the study and NOT share their fall semesterGPA at the end of the study; or not consent to be in the study. Students who were not 18 or didnot classify as first-years or transfer students, were screened out, while those meeting therequirements continued to the pre-survey. Participants received all correspondences throughemail and text and all data was collected via Qualtrics.Consenting participants completed the pre-survey in Qualtrics assessing initial levels ofanxiety, burnout, depressive symptoms, school-life balance, gratitude, loneliness, sense ofbelonging, subjective happiness, positivity resonance, and positive empathy. Following the presurvey, they were randomly assigned into an active control or one of two treatment conditions:Three Good Things (3GT) or Positivity Resonance (3GT-Social). All groups began theirrespective intervention on November 1 st, during which they received an email and text message(if participants opted in to provide a phone number) each evening, asking them to complete theirdaily exercise. Messages were sent at 7pm daily.Participants in the active control condition were asked “What did you do today?”, thosein the Three Good Things (3GT) condition were asked “What went well today?”, and those inthe positivity resonance condition (3GT-Social) were asked “What social interactions went welltoday?”. In each condition, participants were encouraged to provide a minimum of threeresponses but could list up to six. The intervention period lasted 15 days—concluding onNovember 15th. An error in the survey distribution also resulted in participants in the activecontrol condition not receiving the loneliness scale for Day 15.

THE HAPPY HEELS STUDYWithin the intervention period, all participants were presented with repeat assessments ofanxiety, burnout, depressive symptoms, school-life balance, gratitude, loneliness, sense ofbelonging, subjective happiness, positivity resonance, and positive empathy on days 7 and 15.After the intervention period, follow up measures of all mental health measures were collectedtwo (Follow-up 1), four (Follow-up 2) and eight (Follow-up 3) weeks after the intervention.Participants were also asked if they had visited Counseling and Psychological Services (CAPS)throughout the semester. In January, fall 2018 cumulative GPAs of consenting participants (n 130) were obtained through the UNC Registrar’s Office. All students were sent a debriefingemail and offered to attend an additional debriefing session to learn the results of the study.ResultsStatistical AnalysisData were analyzed using SPSS Statistics Software v. 25. Chi-squares assessed whetherattrition during the intervention differentially affected the three conditions. Scores for all mentalhealth measures were calculated at each time point based on published scoring procedures andused for analysis. Scores for positivity resonance were averaged across all responses, while allother scores, excluding school-life-balance, were summed. Between-group differences for mentalhealth measures were assessed with One-way ANCOVAs. These were also used to control forpre-survey results when analyzing post-intervention measures. Post-hoc tests further analyzeddifferences between conditions. A One-way ANOVA was used to test differences in GPA foracademic performance, as well as test for differences in GPA between those who were retainedand dropped out of the study.

THE HAPPY HEELS STUDYAttritionCollapsing across conditions over the intervention period, attrition rates through theintervention period were 69.1% compared to Sexton and Adair’s, 2017 3GT study, which hadattritions rates averaging 53% following their intervention. Table 2 shows the sample size ofeach condition as the study progressed. A Chi Square analysis determined that attrition rates ateach time point were not significantly different across conditions. However, because there was alarge amount of participant drop-out, additional analysis of pre-survey results examined whetherstudents who dropped out of the study had markedly different scores on pre-survey mental healthmeasures. A one-way ANOVA tested all pre-survey measures against whether or not participantsdropped out at Day 15. Results determined that students who dropped out had higher levels ofanxiety than those who completed the intervention period (F 5.892 p .05). For all othermeasures, there was no difference in pre-survey scores between those who dropped out and thosewho remained in the study. Thus, the group of students who did not complete the interventionwere more anxious prior to its start (M 18.13) compared to those who were retained (M 16.03), but did not differ in other measures assessing mental health.Student Mental Health Pre-InterventionParticipants did not show significant between-group differences in any mental healthmeasures before the intervention period. However, all students were assessed for concerninglevels of key measures. Students passing a certain threshold for each measure were considered tomeet a “percent concerning” criteria. Table 3 notes the number of students (collapsed acrossconditions) meeting a percent concerning score at the pre-survey. Percent concerning scores forkey measures, excluding school-life balance, were determined using published precedents. Tocalculate percent concerning for burnout, scores were converted from a 50-point scale to a 100-

THE HAPPY HEELS STUDYpoint scale. A threshold of 50 was used to mark moderate levels of burnout and consideredpercent concerning (Schaufeli et al., 2001). Raw scores for depressive symptoms were convertedto t-scores and scores beyond 1 standard deviation (60 ) were considered moderate levels ofdepressive symptoms and marked as percent concerning (Choi, Schalet, Cook & Cella, 2014).Percent concerning for sense of belonging was determined by comparing the scores for inclusionand rejection questions within the scale; inclusion scores were summed separately fromrejection. A difference score was calculated by subtracting inclusion scores from rejectionscores, with a positive difference score indicating higher feelings of rejection and isolation. Adifference score of 10 or higher was considered a percent concerning level of isolation.Individual items for happiness were averaged across all responses. A score below 5 markedpercent concerning in line with data on various populations (Lyubomirsky, 1999). A score of 50or higher for loneliness was considered a moderate level of loneliness by Perry and was used asthe threshold of percent concerning for students (Perry, 1990). A score below -2 for school-lifebalance was marked as percent concerning and indicated that students were engaging in greaterproportion of negative habits than positive ones on campus.More than half of participants met the percent concerning criteria for happiness andloneliness (64.4% and 62.5%) and 31.1% of students had difficulties with school

Positivity Resonance Improves Mental Health Research has shown that positive emotions affect a person’s mental and physical health. In particular, shared positive emotions are integral for general well-being, social interaction and health (Fredrickson, 2016). Based on this research, Dr. Barbara