Review of Stress Management Interventions for College Students PDF

Received: 11 November 2019 | Revised: 27 March 2020 | Accepted: 28 March 2020 DOI: 10.1111/cpsp.12342 L I T E R AT U R E R E V I E W Stress management interventions for college students: A systematic re...

Received: 11 November 2019 | Revised: 27 March 2020 | Accepted: 28 March 2020 DOI: 10.1111/cpsp.12342 L I T E R AT U R E R E V I E W Stress management interventions for college students: A systematic review and meta-analysis Yagmur Amanvermez1 | Metta Rahmadiana1 | Eirini Karyotaki1,2 | Leonore de Wit1 | David D. Ebert 1,3 | Ronald C. Kessler4 | Pim Cuijpers1 1 Department of Clinical, Neuro-, and Developmental Psychology, Amsterdam Abstract Public Health Research Institute, Vrije This systematic review and meta-analysis aimed to examine the effects of stress Universiteit Amsterdam, Amsterdam, The management interventions in reducing stress, depression, and anxiety among col- Netherlands 2 lege students. Two separate meta-analyses of randomized controlled trials were con- Department of Global Health and Social Medicine, Harvard University Medical ducted using the random-effects model, for students with high-stress levels (n = 8) School, Boston, MA, USA and for the unselected college student population (n = 46). Overall, main results 3 Department of Clinical Psychology showed moderate intervention effects for stress, depression, and anxiety in both and Psychotherapy, Friedrich-Alexander University Erlangen-Nuremberg, Erlangen, groups. Subgroup analyses yielded significant differences related to the theoretical Germany background, the type of control condition, and the length of the intervention in trials 4 Department of Health Care Policy, Harvard targeting students with high-stress levels. However, these subgroup differences, ex- Medical School, Boston, MA, USA cept for the theoretical background, were not observed in trials including unselected Correspondence samples. Our results suggest that stress management interventions may be effective Yagmur Amanvermez, Department of in reducing distress among college students. Clinical, Neuro-, and Developmental Psychology, Amsterdam Public Health KEYWORDS Research Institute, Vrije Universiteit college students, meta-analysis, stress management Amsterdam, Van der Boechorststraat 7-9, 1081 BT Amsterdam, The Netherlands. Email: [email protected] Funding information Ministry of National Education, Turkey 1 | IN T RO D U C T ION a multitude of stressors might be associated with lower aca- demic achievement (Andrews & Wilding, 2004; Eisenberg, Several studies have shown that between 20% and 45% of col- Golberstein, Golberstein, & Hunt, 2009), college dropout lege students experience at least one mental health disorder (DeBerard, Spielmans, & Julka, 2004), depression (Ebert in a given year (Auerbach et al., 2016; Blanco et al., 2008). et al., 2018), physical illness (Cohen, Janicki-Deverts, & The onset of common mental health disorders mostly occurs Miller, 2007), and suicidal thoughts or behaviors (Eskin during the ages of 20–30 years old and may be related to et al., 2016; Mortier et al., 2017). a series of developmental transitions (Kessler et al., 2005; Despite the considerable prevalence of common mental dis- McGorry, Purcell, Goldstone, & Amminger, 2011). Students orders among college students (Auerbach et al., 2016), a small encounter various stressors, such as academic stress, per- number of this population seek professional help for mental sonal problems, career issues, and economic concerns (The health problems. Reasons for treatment underutilization include American College Health Association, 2009). Encountering perceived stigmatization, negative attitudes toward treatment, © 2020 American Psychological Association. Published by Wiley Periodicals, LLC on behalf of the American Psychological Association. All rights reserved. For permissions, please email: [email protected] Clin Psychol Sci Pract. 2020;00:e12342.  wileyonlinelibrary.com/journal/cpsp | 1 of 27 https://doi.org/10.1111/cpsp.12342  | 14682850, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cpsp.12342 by Fernuniversitaet Hagen, Wiley Online Library on [07/08/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 2 of 27    AMANVERMEZ et al. lack of time, lack of perceived need, preference of self-manage- ment in problem-solving, and cultural values (Ebert et al., 2019; Public Health Statement Eisenberg, Downs, Downs, Golberstein, & Zivin, 2009; Han & Symptoms of stress are common among college stu- Pong, 2015; Hunt & Eisenberg, 2010). Universities can play dents. Stress management programs are provided to a role in preventing mental health problems by reaching stu- help students in reducing stress. The present study dents in distress and offering relevant interventions. Although showed that stress management programs have existing mental health support may substantially differ across the potential to decrease stress and prevent mental universities and countries, university counseling services gen- health problems in higher education. erally provide basic mental health services, such as workshops, psychoeducational programs, and individual or group coun- seling for students with personal, interpersonal, academic, or career-related issues at no or little cost (Brunner, Wallace, complaints (Cohen, Kessler, & Gordon, 1995). Stress can be Reymann, Sellers, & McCabe, 2014; Eisenberg, Golberstein, & measured using stressful events checklists, scales measur- Gollust, 2007; Francis & Horn, 2017; Hunt & Eisenberg, 2010). ing the evaluation of the stressful situation, or physiological Stress management programs might also be a useful way responses including cortisol or cardiovascular responses. to help students cope with stressors in universities. A growing Therefore, in this study, high stress is defined as having an body of research has shown that such programs can be effec- elevated score on one of the above-mentioned measures. In tive in improving the mental health of college students (Davies, addition, given that the stress symptomatology overlaps with Morriss, & Glazebrook, 2014; Harrer et al., 2018; Regehr, the symptoms of depression and anxiety (Slavich, 2016), in Glancy, & Pitts, 2013). However, the existing evidence is still the present study we combined the effects of these overlap- not robust. Previously conducted systematic reviews and me- ping mental health problems. ta-analyses did not focus solely on students with high stress, but also included unselected populations and analyzed results simultaneously. Pooling studies including students with high- 2 | M ETHODS stress levels and those who possibly have high-stress levels (unselected populations) together might be misleading, as the In the present article, we report our meta-analyses in line effects of stress management programs might differ according to with the Preferred Reporting Items for Systematic Review the stress scores of participants at pre-intervention. Participants and Meta-Analysis (PRISMA) Statement (Moher, Liberati, with high-stress scores at pre-intervention might show greater Tetzlaff, & Altman, 2009). This study is preregistered in improvements after receiving a stress management program PROSPERO (PROSPERO ID: CRD42019117427) in order compared to the unselected population with mixed levels of to minimize selective reporting, increase transparency, and stress. Furthermore, there are several differences in the concep- reduce publication bias. tualization and measurement of stress that have been found in the previous literature, thereby restricting the interpretability of the findings. A more comprehensive definition of stress is 2.1 | Identification of studies needed to permit a clearer interpretation of the meta-analytic evidence for stress management interventions. Finally, since the Systematic literature searches were performed in the biblio- literature on stress management interventions is continuously graphic databases of Cochrane Library, PubMed, EMBASE, growing, there is a need to update the current state-of-the-art and PsycINFO. An overview of search strings can be found in this field. in Appendix A. Search strings included index and free terms In the present study, unlike previous systematic reviews, of psychotherapy, stress management, and stress/distress. we used strict inclusion criteria focused only on guided stress Moreover, we examined publications of key researchers in this management programs and excluded interventions that fo- field, reference lists of related studies, and previous systematic cused on areas other than stress. We aimed to investigate the review and meta-analyses on stress and psychological distress. extent to which stress management programs are effective in We conducted systematic searches up to the 11th of June 2018. reducing stress/psychological distress, depression, and anxi- The review started the day that the searches were performed. ety compared to control conditions among college students. We performed this investigation separately in considering interventions aimed at students with high stress and in the 2.2 | Eligibility criteria unselected population of college students. Stress was conceptualized as a process, which is influ- Studies were eligible for these meta-analyses if a stress man- enced by external events or internal evaluations of a situa- agement intervention for any type of college student (under- tion and results in emotional, behavioral, or physiological graduate or graduate) was compared to a control condition.  | 14682850, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cpsp.12342 by Fernuniversitaet Hagen, Wiley Online Library on [07/08/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License AMANVERMEZ et al.    3 of 27 We conducted two separate meta-analyses based on symp- and depression as the outcome(s). Only RCTs published in tomatology. Thus, for the first meta-analysis, we included English were eligible for inclusion. studies that focused on college students with elevated levels We excluded studies if the intervention was unguided or of stress based on cutoff scores on stress/distress scales. In consisted of mixed populations (e.g., university students and the second meta-analysis, we added only stress management academic staff/working population/non–college peers). We interventions targeting the unselected college students. We also excluded studies if the main focus was not stress (e.g., included studies using stress/psychological distress, anxiety, inclusion criteria of the participants were solely based on Records identified through database Additional records identified searching through other sources Identification (n = 4055) (n = 344) Records after duplicates removed (n = 3197) Screening Records screened Records excluded (n = 3197) (n = 2680) Dissertation (n = 37) Full text articles assessed Incalculable effect size (n = 13) for eligibility Mixed or other population (n (n = 517) =130) Eligibility No control condition (n = 31) Studies included in No full text (n = 17) qualitative synthesis Not randomized (n = 4) (N/A) Other inclusion criteria (n = 31) Other intervention (n = 125) Other language (n = 8) Studies included in Included Other outcome (n = 23) quantitative synthesis (meta-analysis) Protocol paper (n = 6) high-stress (n = 8) Results published in another unselected (n = 46) paper (n = 7) Unguided (n = 25) Very brief intervention (n = 6) FIGURE 1 PRISMA flow diagram  | 14682850, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cpsp.12342 by Fernuniversitaet Hagen, Wiley Online Library on [07/08/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 4 of 27    AMANVERMEZ et al. TABLE 1 Result of meta-analysis and subgroup analyses of associations between effect sizes and study characteristics for pooled outcomes (students with high-stress levels) n 95% CI comparisons g 95% CI I2 around I2 p NNT Overall effects 8 0.54 [0.31, 0.78] 54 [0, 79] 3.31 Symptom outcomes Stress 6 0.61 [0.30, 0.93] 72 [25, 87] 2.69 Anxiety 8 0.52 [0.25, 0.78] 64 [24, 83] 3.50 Depression 8 0.46 [0.16, 0.77] 73 [44, 87] 3.91 Subgroup analyses Theoretical background CBT versus 6 0.68 [0.47, 0.88] 20 [0, 65] 0.01 2.70 Other 2 0.18 [−0.11, 0.47] 0 N/A 9.43 Comparison Group Active 4 0.30 [0.08, 0.53] 0 [0, 85] 0.01 5.75 Inactive 4 0.76 [0.52, 1.01] 25 [0, 80] 2.42 Length Brief 3 0.36 [0.12, 0.60] 0.00 [0, 90] 0.01 5 (1–4 week) Moderate 4 0.76 [0.52, 1.01] 25 [0, 71] 2.42 (5–8 week) Quality Criteria > 4 2 0.62 [0.37, 0.86] 0 N/A 0.61 2.96 Criteria ≤ 3 6 0.51 [0.16, 0.85] 65 [16, 85] 3.55 anxiety or depression or the main target of the intervention We classified stress management programs according to was sleep/eating habits). theoretical background. Interventions including cognitive re- structuring and stress inoculation were classified as cognitive behavioral therapy (CBT)–based stress management inter- 2.3 | Quality assessment ventions. We grouped interventions incorporating third-wave (TW) concepts including acceptance, defusion, values, and Two reviewers (Y.A. and M.R.) independently evaluated mindfulness (i.e., mindfulness-based stress reduction; Cash & the risk of bias of the studies using the Cochrane Risk of Whittingham, 2010; Öst, 2008). However, if the mindfulness Bias Assessment Tool (Higgins & Green, 2011). The as- component was solely used as a meditation component (i.e., sessments were conducted in the following sources of risk mindfulness meditation), we included these interventions of bias: (a) random sequence generation, (b) allocation under the mind–body category. We clustered meditation, concealment, (c) blinding of participants and personnel, muscle relaxation, breathing exercises, guided imaginary (d) blinding of outcome assessors, (e) handling of incom- techniques, and biofeedback under the mind–body interven- plete outcome data, (f) selective outcome reporting, and tions, as these interventions’ assumptions are compatible and (g) other risk of bias. Disagreements were discussed with a generally include one another (Astin, Shapiro, Eisenberg, & third reviewer (E.K.) Forys, 2003). Lastly, skills training included interventions aimed at improvement of social, academic, or coping skills. 2.4 | Data extraction 2.5 | Meta-analyses Data were extracted by two reviewers (Y.A. and M.R.). The extracted data were related to the publication year, country, We calculated the overall effect size for continuous outcomes means and standard deviations of age, percentage of female using mean scores, standard deviations, and the number of participants, number of participants in intervention and con- participants from the intervention and the control groups. If trol groups, type of intervention, number of sessions, type of the mean or the standard deviation was not reported, we used control condition, outcomes, the scale used for measuring t test scores or reported effect sizes in the paper. If multiple the outcome and inclusion criteria, and means and standard outcomes or scales were used for measuring stress/psycho- deviations for both groups at post-test. If available, we also logical distress, depression, or anxiety in a study, we first extracted follow-up time and scores. We contacted authors pooled these outcomes within the study. As a result, we had if the necessary data were not reported in the paper. In the one effect size for each study. After calculating the effect case of nonresponse, the study was excluded. size of each study, we pooled them across studies separately  | 14682850, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cpsp.12342 by Fernuniversitaet Hagen, Wiley Online Library on [07/08/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License AMANVERMEZ et al.    5 of 27 TABLE 2 Result of Meta-Analysis and Subgroup Analyses of Associations between Effect Sizes and Study Characteristics for Pooled Outcomes (Unselected Population) n 95% CI comparisons g 95% CI I2 around I2 p NNT Overall effects 52 0.56 [0.44, 0.68] 74 [65, 80] 3.18 Overall effects after 45 0.44 [0.37, 0.51] 0.00 [0, 34] 4.10 removing the outliers Symptom outcomes Stress 40 0.58 [0.44, 0.73] 78 [71, 84] 3.09 Anxiety 27 0.52 [0.36, 0.68] 65 [48, 77] 3.50 Depression 18 0.36 [0.21, 0.51] 53 [21, 73] 5.00 Cortisol 6 1.50 [0.63, 2.37] 82 [63, 92] 1.40 Cortisol 5 0.98 [0.58, 1.38] 0 [0, 79] 1.95 (excluding outlier) Subgroup analyses Theoretical background CBT 11 0.56 [0.34, 0.77] 47 [0, 74] 0.01 3.18 TW 17 0.66 [0.42, 0.91] 85 [77, 90] 2.75 Mind–body 13 0.60 [0.32, 0.88] 71 [48, 83] 2.99 Skills training 11 0.28 [0.16, 0.39] 0 [0, 60] 6.41 Skills training 11 0.28 [0.16, 0.39] 0 [0, 60] 0.001 6.41 versus other 41 0.61 [0.42, 0.76] 76 [68–82] 2.69 Comparison group Active 25 0.65 [0.45, 0.85] 86 [80, 90] 0.13 2.78 Inactive 27 0.48 [0.38, 0.58] 0.00 [0, 43] 3.76 Length Brief 13 0.40 [0.25, 0.55] 23 [0, 60] 0.19 4.50 (1–4 week) Moderate 26 0.59 [0.40, 0.77] 79 [70, 86] 3.05 (5–8 week) Long 13 0.63 [0.34, 0.91] 74 [55, 85] 2.86 (9-week and longer) Quality Criteria ≥ 4 10 0.77 [0.37, 1.18] 92 [87, 95] 0.20 2.39 Criteria < 3 42 0.50 [0.39, 0.61] 52 [31, 66] 3.55 for students with high-stress levels and the unselected stu- Higgins, Bottai, & Buchan, 2013). Publication bias was as- dent populations. We conducted meta-analyses using the sessed by inspection of the funnel plot. We also used Egger's Comprehensive Meta-Analysis software package under the test of the intercept to test whether asymmetry observed in random-effects model for pooling outcomes, as we expected the funnel plot was significant (Egger, Smith, Schneider, & considerable heterogeneity between studies. Effect sizes Minder, 1997). Moreover, we conducted Duval and Tweedie's were calculated using Hedges’ g, as it corrects for small sam- trim and fill procedure to detect the number of missing stud- ple size bias (Cuijpers, 2016). A Hedges’ g value of 0.2, 0.5, ies and impute them (Duval & Tweedie, 2000). and 0.8 is considered small, moderate, and large, respectively We conducted subgroup analyses according to the (Cohen, 1988). mixed-effect model. We conducted subgroup analyses in To interpret the clinical meaning of the effect size, we also terms of the theoretical background, namely CBT, TW, calculated number needed to treat (NNT) using the formula mind–body interventions, and skills training. In addition, of Kraemer and Kupfer (2006). We examined heterogeneity we conducted a series of subgroup analyses in terms of the using I2 (Ioannidis, Patsopoulos, & Evangelou, 2007). An control condition (active and inactive control group) and I2 value of 0% is interpreted as nonobserved heterogeneity. duration of the intervention (brief: 1 to 4 weeks; medium: Heterogeneity is considered to be low, moderate, and high 5 to 8 weeks; long: 9 weeks or longer). We also conducted when I2 equals 25%, 50%, and 75%, respectively (Higgins, a subgroup analysis in terms of the quality of studies. We Thompson, Deeks, & Altman, 2003). We used the heterogi classified studies as high quality if they fulfilled at least module of Stata to calculate 95% CI around I2 (Orsini, four criteria of the Cochrane Risk of Bias Assessment Tool.  | 14682850, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cpsp.12342 by Fernuniversitaet Hagen, Wiley Online Library on [07/08/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 6 of 27    AMANVERMEZ et al. Low quality was defined as having fulfilled less than four with students at private universities. Eighteen studies did items. not report the type of university. Studies with the unselected population predominantly were conducted in the United States (14 studies). Other studies were conducted mostly 3 | R ES U LTS in European countries, namely Austria, Belgium, Finland, Germany, Greece, Norway, Spain, Sweden, Switzerland, 3.1 | Selection and inclusion of studies and the Netherlands. There were also studies conducted in Argentina, Australia, Canada, China, India, Iran, Korea, The database searches resulted in 4,399 studies. After re- Malaysia, Saudi Arabia, Thailand, and the United Kingdom. moving the duplicates, 3,197 studies remained of which we The characteristics of the studies with the unselected popula- screened the titles and abstracts and excluded 2,680 studies. tion can be seen in Table B2. We retrieved the full text of 517 studies for further inves- tigation. We excluded 463 studies as they did not meet the inclusion criteria. We had eight studies targeting students 3.3 | Risk of bias with high-stress levels, and 46 studies conducted with the unselected population. Details of the selection process and Risk of bias was substantial in most studies. All studies had at the reasons for exclusion of studies can be seen in Figure 1. least some risk of bias. Five out of eight studies with students with high-stress levels clearly indicated adequate random se- quence generation, whereas three studies indicated adequate 3.2 | Characteristics of included studies allocation concealment. Only one study indicated a low risk of bias for blinding of the participants and personnel, which There were eight studies with 685 students with high- is not possible for almost all psychological interventions. stress levels (intervention groups: N = 346; control groups: Because all studies except one used self-report scales for out- N = 339). The sample size of the studies ranged from 26 to come measurement, we assessed them as at high risk of bias. 150. Sixty-seven percent of participants were female. Four Four studies handled incomplete outcome data adequately. studies were conducted at public universities. One study was Only two studies fulfilled the criteria for selective reporting, conducted at a private university. The type of university was and we could not find a protocol paper of the remaining six not reported in three studies. The included studies mostly studies. In total, two studies fulfilled four or more criteria of used CBT as their theoretical background (n = 6). Four stud- the Cochrane Risk of Bias Assessment Tool. Details of the ies used an inactive control (i.e., waiting list) and four studies risk of biased assessment can be seen in Table B1. used an active group (i.e., no treatment, only self-help in- Similarly, we found the risk of bias in all studies with the formation). Studies used various scales for inclusion criteria. unselected population. Nineteen out of 46 studies showed Two studies used the Kessler Psychological Distress Scale adequate random sequence generation. Allocation conceal- (Kessler et al., 2003). Other scales used for the inclusion cri- ment reported adequately in 10 studies. Blinding of partic- teria were the Perceived Stress Scale-4 (Cohen, Kamarck, & ipants was not possible for almost all studies; therefore, we Mermelstein, 1983), the Depression Anxiety and Stress Scale- found only three studies as being at low risk of bias for this 21 (Lovibond & Lovibond, 1995), the Perceived Control criterion. Only eight studies used physiological assessment Over Stressful Events Scale (Frazier et al., 2011), the General (cortisol or blood pressure) in addition to self-reported mea- Health Questionnaire-30 (Goldberg, 1978), the Revised Life sures; therefore, we rated these studies as being at low risk of Stress Scale for College Students (Chon, Kim, & Yi, 2000), performance bias. Twenty-six studies adequately handled in- the Symptom Check List-90-Revised (SCL-90-R; Derogatis, complete data. We could only retrieve protocol papers of five Rickels, & Rock, 1976), and the Life Experiences Survey studies. Details can be seen in Table B2. In total, we could (LES; Sarason, Johnson, & Siegel, 1978). Two studies were only find ten studies being at low risk across all bias domains. conducted in Australia and two in the United States. Others were conducted in Canada, Germany, Korea, and Scotland. The characteristics of the studies with students with high 3.4 | The overall effect of stress management stress can be found in Table B1. interventions We included 46 studies with the unselected population yielding a total number of 4,800 participants (intervention The overall analysis of effect sizes yielded a significant mod- groups: N = 2,457; control groups: N = 2,343). The sample erate effect size for stress management interventions for stu- size of the studies ranged from 28 to 616. Seventy-one per- dents with high-stress levels (g = 0.54; 95% CI [0.31, 0.78]; cent of the participants were female. Twenty-six studies were p <.001; n = 8; Figure C1). Heterogeneity was moderate I2 conducted at public universities. Two studies were conducted = 54% (95% CI [0, 79%]). We found no outliers based on  | 14682850, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cpsp.12342 by Fernuniversitaet Hagen, Wiley Online Library on [07/08/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License AMANVERMEZ et al.    7 of 27 visual inspection of the forest plot. We also found a small- Three studies used salivary cortisol (Gaab et al., 2003, 2006; to-moderate effect size for depression (g = 0.46; 95% CI Iglesias et al., 2012), and one study measured blood corti- [0.16, 0.77]; p =.003; n = 8), a moderate-to-large effect for sol (Gopal et al., 2011). Various cortisol indices were used stress (g = 0.61; 95% CI [0.30, 0.93]; p <.001, n = 6), and in the studies, including cortisol awakening response, short a moderate effect for anxiety (g = 0.52; 95% CI [0.25, 0.78]; circadian salivary cortisol profile, and serum cortisol level. p <.001; n = 8) with moderate-to-high heterogeneity (I2 for In our meta-analysis, we pooled all cortisol indices to exam- stress: 72%, 95% CI [25, 87%]; depression: 73%, 95% CI ine change scores between intervention and control groups in [44, 87%]; anxiety: 64%, 95% CI [24%–83%]). Details can the studies with the unselected population. The meta-analysis be seen in Table 1. We found no indication for publication yielded a large effect size with high heterogeneity (g = 1.50; bias (Figure C2). 95% CI [0.63, 2.37]; p =.001; I2 = 82%). We detected one Forty-six studies with 52 comparison groups were in- outlier (Gopal et al., 2011) based on the visual inspection of cluded in our meta-analysis for the unselected population. As the forest plot and ran the analysis after excluding the outlier. can be seen in Figure C3, the overall effect size was found Similarly, results showed a large effect size with zero hetero- to be moderate and significant (g = 0.56; 95% CI [0.44, geneity (g = 0.98; 95% CI [0.58, 1.38]; p <.001; I2 = 0%). 0.68]; p <.001) with high heterogeneity (I2 = 74%; 95% Although heterogeneity was zero after the exclusion of the CI [65, 80%]). Since heterogeneity was high, we excluded outlier, there was still evidence that high heterogeneity might seven outliers (Aboalshamat, Hou, & Strodl, 2015; Baghurst exist based on a wide range of confidence intervals around I2 & Kelley, 2014; Frögéli, Djordjevic, Rudman, Livheim, & (95% CI [0, 79%]). Results can be seen in Table 2. Gustavsson, 2016; Galante et al., 2018; Kanji, White, & Ernst, 2006; Kim, 2014; Sampl, Maran, & Furtner, 2017) based on the visual inspection of the forest plot. We excluded 3.6 | Subgroup analyses studies if their 95% CI did not overlap with the 95% CI of the pooled effect size. After excluding outliers, we observed a Subgroup analyses of studies targeting students with small-to-moderate effect size (g = 0.44; 95% CI [0.37, 0.51]; high-stress levels showed significant effects in favor of p <.001; n = 45) with zero heterogeneity (I2 = 0%; 95% CI CBT-based interventions versus other types of theoretical [0%–34%]). We also found a moderate effect size for anxiety background approaches, inactive control groups versus ac- (g = 0.52; 95% CI [0.36, 0.68]; p <.001; n = 27) and stress tive controls, and moderately lengthy interventions versus (g = 0.58; 95% CI [0.44, 0.73]; p <.001; n = 40), while brief interventions. However, there was no significant dif- a small-to-moderate effect size was found for depression ference between groups for high versus low risk of bias. (g = 0.36; 95% CI [0.21, 0.51]; p <.001; n = 18). Details can Results for students with high-stress levels can be seen in be seen in Table 2. With regard to publication bias, the in- Table 1. Subgroup analyses for the unselected population spection of the funnel plot indicated some risk for publication yielded no significant differences in terms of the type of bias for studies with the unselected population (Figure C4). control condition, length of intervention, and risk of bias. Egger's test was significant (p =.03; intercept: 1.31; 95% Skills training programs yielded a significantly lower effect CI [0.07, 2.53]). However, Duval and Tweedie's trim and fill size than CBT, TW, and mind–body interventions. Results procedure showed no missing studies. can be seen in Table 2. 3.5 | The overall effect of stress management 3.7 | Follow-up interventions on cortisol Five studies with students with high-stress levels assessed the Only one study (Kim, Lee, Kim, Noh, & Lee, 2016) assessed long-term effects of stress management programs. Results the cortisol level for students with high-stress levels as an showed that the effect of stress management programs re- outcome. Four studies with six intervention groups investi- mained up to 12-month follow-up (g = 0.40; 95% CI [0.21, gated the effects of the stress management program for cor- 0.60]; p <.001; I2 = 25%). We classified the follow-up tisol levels for the unselected population (Gaab et al., 2003; length into three groups as 1 to 3 months, 4 to 6 months, and Gaab, Sonderegger, Scherrer, & Ehlert, 2006; Gopal, Mondal, 6 months and longer. A moderate effect size was found for Gandhi, Arora, & Bhattacharjee, 2011; Iglesias et al., 2012). studies assessing the effects of stress management interven- Two studies measured cortisol during a natural stressor (i.e., tions for 1 to 3 months (g = 0.50, 95% CI [0.27, 0.73]; p examination day; Gaab et al., 2006; Gopal et al., 2011), and <.001; n = 2; I2 = 0%) and 4 to 6 months (g = 0.45; 95% one study used psychosocial stress and a cognitive stress CI [0.14, 0.77]; p <.001; n = 3, I2 = 20%). We found a test (i.e., The Trier Social Stress Test and mental arithme- small-to-moderate effect size up to 20-month follow-up for tic test) during measurement of cortisol (Gaab et al., 2006). the unselected population (g = 0.41; 95% CI [0.22, 0.59];  | 14682850, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cpsp.12342 by Fernuniversitaet Hagen, Wiley Online Library on [07/08/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 8 of 27    AMANVERMEZ et al. p <.001; n = 16; I2 = 68%). For the unselected population, In the present study, we also calculated the effect size results did not yield a significant effect size for 1- to 3-month for stress, depression, and anxiety separately. We found that follow-up (g = 0.43; 95% CI [−0.02, 0.89]; p =.06; n = 6; stress management programs had a large effect size in reduc- I2 = 85%). However, a large effect size was found for 3- to ing stress (students with high stress: g = 0.61; unselected 6-month follow-up (g = 0.71; 95% CI [0.34, 1.09]; p <.001; population: g = 0.58), a moderate effect size on anxiety (both n = 3; I2 = 0%). We also found a low effect size for 6-month groups: g = 0.52), and a small-to-moderate effect on depres- and longer follow-up (g = 0.23; 95% CI [0.02, 0.44]; p =.03; sion (students with high stress: g = 0.46; unselected group: n = 8; I2 = 56%). 0.36). In a meta-analysis, stress management programs for college students had a considerably large effect in improving anxiety (g = 0.77; 95% CI [0.58, 0.88]) and an even higher 4 | D IS C U SS ION effect on depression (g = 0.81; 95% CI [0.13, 1.49]; Regehr et al., 2013). A possible reason for this difference might be The purpose of this systematic review was to investigate the related to different characteristics of included stress manage- effectiveness of stress management interventions for college ment programs. In our meta-analyses, we might have used students with regard to stress/psychological distress, depres- more strict inclusion criteria. For instance, unlike the me- sion, and anxiety. We conducted two separate meta-analyses ta-analysis of Regehr et al. (2013), we did not include studies to summarize the literature for college students with high- on test anxiety. On the other hand, similar to our findings, stress levels and the unselected student group. We found that another meta-analysis demonstrated that preventative inter- stress management programs were effective when compared ventions for college students had a higher effect on anxiety with control conditions. Subgroup analyses showed that (g = 0.40; 95% CI [0.33, 0.46]) and stress (g = 0.32; 95% CI CBT, TW, and mind–body interventions had higher effects [0.25, 0.40]) when compared to depression (g = 0.20; 95% than skills training programs in the unselected population. CI [0.11, 0.30]; Conley, Durlak, & Kirsch, 2015). This find- In addition, there were no significant differences in terms of ing is also in line with another meta-analysis examining the the length of the intervention, the risk of bias, and the type effects of psychological interventions targeting depression of control condition for the unselected population. Results prevention in young adults (Breedvelt et al., 2018). Results of the comparison between type of control, length of the in- of this meta-analysis indicated that stress reduction programs tervention, and theoretical background yielded significant had a smaller effect on depression (g = 0.30; 95% CI [0.07, differences for students with high-stress levels. Stress man- 0.54]), whereas interventions with a depression focus yielded agement programs had a large effect size for cortisol. a higher effect size (g = 0.54; 95% CI [0.24, 0.83]). Although Our findings are in line with the evidence from previ- stress is an important predictor of depression, interventions ous meta-analyses on stress management intervention in targeting depression might possibly benefit more from com- non–college student populations (Heber et al., 2017; van der ponents that are related to depression itself. Klink, Blonk, Schene, & Van Dijk, 2001; Regehr et al., 2013; In subgroup analyses, we found that there was no signif- Richardson & Rothstein, 2008). In a recent meta-analysis, icant difference between CBT- and TW-based stress man- stress management interventions delivered via the Internet agement programs, and they both produced a higher effect yielded a similar effect size for stress (d = 0.43; 95% CI [0.1, size when compared with skills training interventions, sim- 0.54]) but a lower effect size for depression (d = 0.34; 95% CI ilar to previous studies (Harrer et al., 2018; van der Klink [0.21, 0.48]) and anxiety (d = 0.32; 95% CI [0.17, 0.47]; Heber et al., 2001; Regehr et al., 2013). Generally, CBT- and TW- et al., 2017). Similarly, a meta-analysis of stress management based stress management programs incorporate psychoedu- programs for the working population found a moderate effect cation on stress accompanied by information and activities size on symptoms of anxiety, depression, and stress (d = 0.44; on coping skills. Evidence from a previous meta-analysis in- 95% CI [0.36, 0.52]; van der Klink et al., 2001). Moreover, in dicated that psychoeducation was an important component of line with our findings, van der Klink et al. (2001) showed that the stress management interventions (Van Daele, Hermans, the effects of the stress management programs were higher Van Audenhove, & Van den Bergh, 2011). In addition, for participants with elevated stress levels compared to the CBT- and TW-based stress management programs might be unselected population. The higher effect size for the prese- successful by changing the appraisal of the students toward lected population compared to the unselected population was stressors and enable them to improve their repertoire of cop- also found by previous meta-analyses in this field (Harrer ing skills. Nevertheless, it should be noted that subgroup anal- et al., 2018; Zoogman, Goldberg, Hoyt, & Miller, 2015). The yses do not imply causality; thus, more evidence is needed difference between the effects of stress management inter- to draw firm conclusions regarding the relative effectiveness ventions for people with elevated scores and the unselected of different psychotherapeutic types in stress management. population may be attributed to the differences in baseline Another finding of our study is that mind–body interventions scores of participants. showed similar effects to those of CBT-based interventions in  | 14682850, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cpsp.12342 by Fernuniversitaet Hagen, Wiley Online Library on [07/08/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License AMANVERMEZ et al.    9 of 27 the unselected college student population. Findings from pre- targeting the students with high-stress levels, there was no vious systematic reviews showed that mind–body interven- consensus on the inclusion criteria in terms of the scales and tions showed mixed results by yielding low (van der Klink cutoff scores to assess the high levels of stress. One possi- et al., 2001) to moderate effect sizes (Breedvelt et al., 2019; ble explanation for this might be the ongoing discussions on Conley et al., 2015). It is well-documented that redirect- the conceptualization of stress. Unlike depression or anxiety, ing one's attention to the present moment, body sensations, stress is not defined as a diagnostic concept; therefore, the or relaxing images might be helpful through replacing the definition of high stress is somehow ambiguous. Moreover, physiological response of stress with relaxation response stress management programs can also be conducted with in- (Baer, 2003; Wieland, Manheimer, & Berman, 2011). This dividuals who have not been clinically diagnosed. Third, few process might also lead to prevent ruminative thinking by studies had a low risk of bias. Although subgroup analysis redirecting the attention from stress-invoking thoughts and for risk of bias revealed no significant difference between reduce the alarm reaction of the body caused by stressors. low- and high-quality studies, a considerable number of Moreover, mind–body interventions might be effective par- studies reported either insufficient information or high risk ticularly for the nonclinical population who have already on the examined domains of risk of bias. This phenomenon acquired certain adaptive coping strategies. Our finding of is called “garbage-in-garbage-out” in systematic reviews. skills training stress management interventions with a low As a result, this might affect the credibility of our results. effect is also consistent with previous meta-analyses (Harrer Therefore, our results should be interpreted with caution. et al., 2018; Richardson & Rothstein, 2008). Fourth, most of the participants in our meta-analysis were In the present meta-analysis, a large effect size was found female students. Therefore, the generalizability of the results for cortisol levels of students. In the previous literature on to males is limited. We are also uncertain whether the find- this topic, mixed results were reported. For instance, Regehr ings are generalized to private universities, as many studies et al. (2013) found that stress management programs for col- did not report the type of university and we could not re- lege students yielded a moderate effect for salivary cortisol trieve this information from other sources. In addition, the (g = 0.52; 95% CI [0.20, 0.83]; Regehr et al., 2013). On the generalizability of our findings to individuals outside the other hand, in a recent meta-analysis, psychological interven- universities such as non–college-attending peers, or adults in tions for depression revealed no significant effect on corti- general populations are equivocal due to specific stressors/ sol (g = −0.19; 95% CI [−0.45, 0.06]; Cristea, Karyotaki, protective factors related to university life, and differences Hollon, Cuijpers, & Gentili, 2019). In our study, there were in social contexts (Hanel & Vione, 2016; Kenney, Anderson, only four articles available for cortisol in which various indi- & Stein, 2018; Kovess-Masfety et al., 2016; Peterson, 2001). ces and conditions were used for the assessment. Therefore, Therefore, future studies should examine the effect of stress it is difficult to draw concrete conclusions based on such a management in general populations. Moreover, our results small number of studies. cannot be generalized to college students in different cultural Overall, previously conducted systematic reviews pooled and social settings (e.g., in low- and middle-income coun- stress management programs without differentiating between tries), since included studies have been mostly conducted in the unselected populations and students with high stress. The high-income Western countries. Future studies should test current findings have demonstrated that there are differences the appropriateness of stress management interventions in between the target groups on effect sizes for stress, anxiety, different cultures. Fifth, only a few studies used cortisol as a and depression for two separate college student populations. measurement of stress, and in these studies, various indices Results also revealed that effect sizes were somewhat higher were used to measure cortisol. This might lead to high het- for students with high stress than unselected populations. erogeneity. Sixth, only a few studies provided follow-up data; And yet, findings suggest that the unselected populations can therefore, we lack knowledge about the duration of effects of also benefit from the stress management programs. Our re- the stress management programs. Lastly, negative attitudes sults can be informative for planning the stress management and beliefs toward mental health problems such as fear of programs in tertiary education. However, these results should stigmatization can cause low uptake of mental health ser- be interpreted with caution due to several limitations. First, vices (Andrade et al., 2014; Eisenberg, Downs, et al., 2009). we included a wide range of studies in terms of the inter- However, it is not clear whether the impeding effect of atti- vention characteristics. This might be a possible explanation tudinal factors on help-seeking is substantial in the included for the high number of outliers according to visual inspec- studies. Because college students inevitably face several tion of the forest plot. Second, studies varied widely also in stressors during these years, they generally regard stress as a terms of the screening and outcome measures. Most studies normal part of university life (Acharya, Jin, & Collins, 2018; did not indicate any inclusion criteria for participants and in- Downs & Eisenberg, 2012). In addition, stress management stead included all students who were willing to participate programs are offered as preventative approaches by nature, in a stress management program. Moreover, within studies which aim to promote coping skills and improve individual  | 14682850, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cpsp.12342 by Fernuniversitaet Hagen, Wiley Online Library on [07/08/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 10 of 27    AMANVERMEZ et al. resources before the stressors lead to more severe problems ORCID (Archer, 1986). Therefore, stress management programs Yagmur Amanvermez https://orcid. might be less stigmatized and may be accepted more easily org/0000-0002-5555-193X than interventions directly targeting depression, anxiety, or Metta Rahmadiana https://orcid. other mental health problems (Billings, Cook, Hendrickson, org/0000-0002-4150-7559 & Dove, 2008). However, evidence related to the relationship Eirini Karyotaki https://orcid.org/0000-0002-0071-2599 between attitudes toward mental health problems and utiliza- Leonore de Wit https://orcid.org/0000-0002-2745-3806 tion from stress management programs remains unclear. David D. Ebert https://orcid.org/0000-0001-6820-0146 Despite these limitations, evidence suggests that stress Ronald C. Kessler https://orcid. management programs at universities have the potential to org/0000-0003-4831-2305 decrease psychological distress, depression, and anxiety by Pim Cuijpers https://orcid.org/0000-0001-5497-2743 helping students change their maladaptive coping responses. CBT-based stress management programs can be effective for R E F E R E NC E S students with high stress. College students could specifically Aboalshamat, K., Hou, X. Y., & Strodl, E. (2015). The impact of a benefit from CBT- and TW-based stress management pro- self-development coaching programme on medical and dental grams even if they do not experience high stress. Universities students’ psychological health and academic performance: A ran- domised controlled trial. BMC Medical Education, 15(1), 1–13. could provide stress management interventions such as https://doi.org/10.1186/s1290​9-015-0412-4 mind–body interventions without requiring massive re- Acharya, L., Jin, L., & Collins, W. (2018). College life is stressful sources (i.e., mental health professionals) for general univer- today–Emerging stressors and depressive symptoms in college stu- sity student populations. As our findings suggested, college dents. Journal of American College Health, 66(7), 655–664. https:// students might benefit more from medium- and longer-term doi.org/10.1080/07448​481.2018.1451869 interventions compared to brief interventions. In general, American College Health Association. (2009). American College Health stress management interventions can reduce stress levels in Association-National college health assessment Spring 2008 refer- college students and possibly the symptoms of mental health ence group data report (abridged): The American College Health Association. Journal of American College Health, 57(5), 477–488. problems that are associated with stress, such as anxiety and https://doi.org/10.3200/JACH.57.5.477-488 depression. In addition to offering evidence-based stress Andrade, L. H., Alonso, J., Mneimneh, Z., Wells, J. E., Al-Hamzawi, management programs on-campus, universities can develop A., Borges, G., Bromet, E., Bruffaerts, T., de Girolamo, G., de multiple strategies to engage students in these programs by Graaf, R., Florescu, S., Gureje, O., Hinkov, H. R., Hu, C., Huang, helping them overcome the barriers to help-seeking. Campus- Y., Hwang, I., Jin, R., Karam, E. G., Kovess-Masfety, B., Levinson, wide campaigns or educational programs can be organized to D., … Kessler, R. C. (2014). Barriers to mental health treatment: create awareness for stressors in higher education and reduce Results from the WHO world mental health surveys. Psychological mental health stigma. College students should be notified via Medicine, 44(6), 1303–1317. https://doi.org/10.1017/S0033​29171​ 3001943 numerous channels (e.g., study advisors, faculty and staff, Andrews, B., & Wilding, J. M. (2004). The relation of depression and and counseling services) about the on-campus resources. anxiety to life- stress and achievement in students. British Journal Universities can conduct screening for stress symptoms, pro- of Psychology, 95, 509–521. https://doi.org/10.1348/00071​ 26042​ vide individualized feedback, and refer students to relevant 369802 services based on their stressors and needs. The impact of Archer, J. (1986). Stress management: Evaluating a preventive approach the different engagement strategies on the effects of stress for college students. Journal of American College Health, 34(4), management programs also needs to be examined in future 157–160. https://doi.org/10.1080/07448​481.1986.9939629 Astin, J. A., Shapiro, S. L., Eisenberg, D. M., & Forys, K. L. (2003). research. In addition, for further studies, we recommend ex- Mind-body medicine: State of the science, implications for practice. amining moderators and mediators in order to gain a deeper The Journal of the American Board of Family Practice, 16(2), 131– understanding of who benefits more from which stress man- 147. https://doi.org/10.3122/jabfm.16.2.131 agement program. We also recommend that rigorous report- Auerbach, R. P., Alonso, J., Axinn, W. G., Cuijpers, P., Ebert, D. D., ing of the RCTs is required to improve the assessment of the Green, J. G., Hwang, I., Kessler, R., Liu, H., Mortier, P., Nock, M., validity of evidence. Overall, based on the current findings, Pinder-Amaker, S., Sampson, N., Aguilar-Gaxiola, A., Al-Hamzawi, stress management interventions appear to be beneficial for A., Andrade, L., Benjet, C., Caldas-de-Almeida, M., Demyttenaere, both students with elevated stress levels and those who might K., Florescu, S., … Bruffaerts, R. (2016). Mental disorders among college students in the World Health Organization world mental or might not suffer from stress. health surveys. Psychological Medicine, 46(14), 2955–2970. https:// doi.org/10.1017/S0033​29171​6001665 ACKNOWLEDGMENTS Baer, R. A. (2003). Mindfulness training as a clinical intervention: Ronal Kessler was a consultant for Datastat, Inc, Sage A conceptual and empirical review. Clinical Psychology: Science Pharmaceuticals, and Takeda for the past 3 years. We would and Practice, 10(2), 125–143. https://doi.org/10.1093/clips​ y/ like to thank David Turner for proofreading the manuscript. bpg015  | 14682850, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cpsp.12342 by Fernuniversitaet Hagen, Wiley Online Library on [07/08/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License AMANVERMEZ et al.    11 of 27 Baghurst, T., & Kelley, B. C. (2014). An examination of stress in college Davies, E. B., Morriss, R., & Glazebrook, C. (2014). Computer- students over the course of a semester. Health Promotion Practice, delivered and web-based interventions to improve depression, 15(3), 438–447. https://doi.org/10.1177/15248​39913​510316 anxiety, and psychological well-being of university students: A Billings, D. W., Cook, R. F., Hendrickson, A., & Dove, D. C. (2008). A systematic review and meta-analysis. Journal of Medical Internet web-based approach to managing stress and mood disorders in the Research, 16(5), 1–22. https://doi.org/10.2196/jmir.3142 workforce. Journal of Occupational and Environmental Medicine, DeBerard, M. S., Spielmans, G. I., & Julka, D. L. (2004). Predictors 50(8), 960. https://doi.org/10.1097/JOM.0b013​e3181​6c435b of academic achievement and retention among college freshmen: A Blanco, C., Okuda, M., Wright, C., Hasin, D. S., Grant, B. F., Liu, S.- longitudinal study. College Student Journal, 38(1), 66–80. M., & Olfson, M. (2008). Mental health of college students and Derogatis, L. R., Rickels, K., & Rock, A. F. (1976). The SCL 90 and the their non-college-attending peers: Results from the national epi- MMPI: A step in the validation of a new self report scale. British demiologic study on alcohol and related conditions. Archives of Journal of Psychiatry, 128(3), 280–289. https://doi.org/10.1192/ General Psychiatry, 65(12), 1429. https://doi.org/10.1001/archp​ bjp.128.3.280 syc.65.12.1429 Downs, M. F., & Eisenberg, D. (2012). Help seeking and treat- Breedvelt, J. J., Amanvermez, Y., Harrer, M., Karyotaki, E., Gilbody, S., ment use among suicidal college students. Journal of American Bockting, C. L., … Ebert, D. D. (2019). The effects of meditation, College Health, 60(2), 104–114. https://doi.org/10.1080/07448​ yoga, and mindfulness on depression, anxiety, and stress in tertiary 481.2011.619611 education students: A meta-analysis. Frontiers in Psychiatry, 10, Duval, S., & Tweedie, R. (2000). Trim and fill: A simple fun- https://doi.org/10.3389/fpsyt.2019.00193 nel-plot–based method of testing and adjusting for publica- Breedvelt, J. J. F., Kandola, A., Kousoulis, A. A., Brouwer, M. E., tion bias in meta-analysis. Biometrics, 56, 455–463. https://doi. Karyotaki, E., Bockting, C. L. H., & Cuijpers, P. (2018). What are org/10.1111/j.0006-341x.2000.00455.x the effects of preventative interventions on major depressive disor- Ebert, D. D., Buntrock, C., Mortier, P., Auerbach, R., Weisel, K. K., der (MDD) in young adults? A systematic review and meta-analy- Kessler, R. C., … Bruffaerts, R. (2018). Prediction of major de- sis of randomized controlled trials. Journal of Affective Disorders, pressive disorder onset in college students. Depression and Anxiety, 239(June), 18–29. https://doi.org/10.1016/j.jad.2018.05.010 36(4), 294–304. https://doi.org/10.1002/da.22867 Brunner, J. L., Wallace, D. L., Reymann, L. S., Sellers, J. J., & McCabe, Ebert, D. D., Mortier, P., Kaehlke, F., Bruffaerts, R., Baumeister, A. G. (2014). College counseling today: Contemporary students and H., Auerbach, R. P., … Kessler, R. C. (2019). Barriers of men- how counseling centers meet their needs. Journal of College Student tal health treatment utilization among first-year college students: Psychotherapy, 28(4), 257–324. https://doi.org/10.1080/87568​ First cross-national results from the WHO World Mental Health 225.2014.948770 International College Student Initiative. International Journal Cash, M., & Whittingham, K. (2010). What facets of mindfulness con- of Methods in Psychiatric Research, 28(2), 1–14. https://doi. tribute to psychological well-being and depressive, anxious, and org/10.1002/mpr.1782 stress-related symptomatology? Mindfulness, 1(3), 177–182. https:// Egger, M., Smith, G. D., Schneider, M., & Minder, C. (1997). Bias in doi.org/10.1007/s1267​1-010-0023-4 meta-analysis detected by a simple, graphical test. British Medical Chon, K. K., Kim, K. H., & Yi, J. S. (2000). Development of the re- Journal, 315, 629–634. https://doi.org/10.1136/bmj.315.7109.629 vised life stress scale for college students. Korean Journal of Health Eisenberg, D., Downs, M. F., Golberstein, E., & Zivin, K. (2009). Psychology, 5(2), 316–335. Stigma and help seeking for mental health among college students. Cohen, J. (1988). Statistical power analysis for the behavioral sciences Medical Care Research and Review, 66(5), 522–541. https://doi. (2nd ed.). Routledge. org/10.1177/10775​58709​335173 Cohen, S., Janicki-Deverts, D., & Miller, G. (2007). Psychological stress Eisenberg, D., Golberstein, E., & Gollust, S. E. (2007). Help-seeking and disease. Journal of the American Medical Association, 298(14), and access to mental health care in a university student population. 1685–1687. https://doi.org/10.1001/jama.298.14.1685 Medical Care, 45(7), 594–601. https://doi.org/10.1097/MLR.0b013​ Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure e3180​3bb4c1 of perceived stress. Journal of Health and Social Behavior, 24(4), Eisenberg, D., Golberstein, E., & Hunt, J. B. (2009). Mental health and 385–396. https://doi.org/10.2307/2136404 academic success in college. Journal of. Economic Analysis and Cohen, S., Kessler, R. C., & Gordon, L. U. (1995). Strategies for mea- Policy, 9(1), https://doi.org/10.2202/1935-1682.2191 suring stress in studies of psychiatric and physical disorders. In S. Eskin, M., Sun, J., Abuidhail, J., Yoshimasu, K., Janghorbani, M., Cohen, R. C. Kessler, & L. U. Gordon (Eds.), Measuring stress: Flood, C., … Yoshimasu, K. (2016). Suicidal behavior and psycho- A guide for health and social scientists (pp. 3–26). Oxford, UK: logical distress in university students : A 12-nation study. Archives Oxford University Press. of Suicide Research, 20(3), 369–388. https://doi.org/10.1080/13811​ Conley, C. S., Durlak, J. A., & Kirsch, A. C. (2015). A meta-analy- 118.2015.1054055 sis of universal mental health prevention programs for higher ed- Francis, P. C., & Horn, A. S. (2017). Mental health issues and coun- ucation students. Prevention Science, 16(4), 487–507. https://doi. seling services in US higher education: An overview of recent re- org/10.1007/s1112​1-015-0543-1 search and recommended practices. Higher Education Policy, 30(2), Cristea, I. A., Karyotaki, E., Hollon, S. D., Cuijpers, P., & Gentili, C. 263–277. https://doi.org/10.1057/s4130​7-016-0036-2 (2019). Biological markers evaluated in randomized trials of psy- Frazier, P., Keenan, N., Anders, S., Perera, S., Shallcross, S., & Hintz, S. chological treatments for depression: A systematic review and me- (2011). Perceived past, present, and future control and adjustment to ta-analysis. Neuroscience & Biobehavioral Reviews, 101, 32–44. stressful life events. Journal of Personality and Social Psychology, https://doi.org/10.1016/j.neubi​orev.2019.03.022 100(4), 749–765. https://doi.org/10.1037/a0022405 Cuijpers, P. (2016). Meta-analyses in mental health research: A practi- Frögéli, E., Djordjevic, A., Rudman, A., Livheim, F., & Gustavsson, cal guide. Vrije Universiteit Amsterdam. P. (2016). A randomized controlled pilot trial of acceptance and  | 14682850, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cpsp.12342 by Fernuniversitaet Hagen, Wiley Online Library on [07/08/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 12 of 27    AMANVERMEZ et al. commitment training (ACT) for preventing stress-related ill health Medical Journal, 335(7626), 914–916. https://doi.org/10.1136/ among future nurses. Anxiety, Stress and Coping, 29(2), 202–218. bmj.39343.408449.80 https://doi.org/10.1080/10615​806.2015.1025765 Kanji, N., White, A., & Ernst, E. (2006). Autogenic training to re- Gaab, J., Blättler, N., Menzi, T., Pabst, B., Stoyer, S., & Ehlert, U. duce anxiety in nursing students: Randomized controlled trial. (2003). Randomized controlled evaluation of the effects of cogni- Journal of Advanced Nursing, 53(6), 729–735. https://doi. tive-behavioral stress management on cortisol responses to acute org/10.1111/j.1365-2648.2006.03779.x stress in healthy subjects. Psychoneuroendocrinology, 28(6), 767– Kenney, S. R., Anderson, B. J., & Stein, M. D. (2018). Drinking to 779. https://doi.org/10.1016/S0306​-4530(02)00069​-0 cope mediates the relationship between depression and alcohol Gaab, J., Sonderegger, L., Scherrer, S., & Ehlert, U. (2006). risk: Different pathways for college and non-college young adults. Psychoneuroendocrine effects of cognitive-behavioral stress Addictive Behaviors, 80, 116–123. https://doi.org/10.1016/j. management in a naturalistic setting – A randomized controlled addbeh.2018.01.023 trial. Psychoneuroendocrinology, 31(4), 428–438. https://doi. Kessler, R. C., Barker, P. R., Colpe, L. J., Epstein, J. F., Gfroerer, J. C., org/10.1016/j.psyne​uen.2005.10.005 Hiripi, E., … Zaslavsky, A. M. (2003). Screening for serious mental Galante, J., Dufour, G., Vainre, M., Wagner, A. P., Stochl, J., Benton, illness in the general population. Archives of General Psychiatry, A., … Jones, P. B. (2018). A mindfulness-based intervention to 60(2), 184–189. https://doi.org/10.1001/archp​syc.60.2.184 increase resilience to stress in university students (the Mindful Kessler, R. C., Berglund, P., Demler, O., Jin, R., Merikangas, K. R., & Student Study): A pragmatic randomised controlled trial. The Walters, E. E. (2005). Lifetime prevalence and age-of-onset distri- Lancet Public Health, 3(2), e72–e81. https://doi.org/10.1016/S2468​ butions of DSM-IV disorders in the National Comorbidity Survey -2667(17)30231​-1 Replication. Archives of General Psychiatry, 62(6), 593–602. Goldberg, D. (1978). Manual of the general health questionnaire. Nfer https://doi.org/10.1001/archp​syc.62.6.593 Nelson. Kim, S. D. (2014). Effects of yogic exercises on life stress and blood Gopal, A., Mondal, S., Gandhi, A., Arora, S., & Bhattacharjee, J. (2011). glucose levels in nursing students. Journal of Physical Therapy Effect of integrated yoga practices on immune responses in exam- Science, 26(12), 2003–2006. https://doi.org/10.1589/jpts.26.2003 ination stress: A preliminary study. International Journal of Yoga, Kim, S., Lee, H., Kim, H., Noh, D., & Lee, H. (2016). Effects of an 4(1), 26–32. https://doi.org/10.4103/0973-6131.78178 integrated stress management program (ISMP) for psychologically Han, M., & Pong, H. (2015). Mental health help-seeking behaviors distressed students: A randomized controlled trial. Perspectives among Asian American community college students: The effect of in Psychiatric Care, 52(3), 178–185. https://doi.org/10.1111/ stigma, cultural barriers, and acculturation mental health help-seek- ppc.12114 ing behaviors among Asian American community college students. Kovess-Masfety, V., Leray, E., Denis, L., Husky, M., Pitrou, I., & Journal of College Student Development, 56(1), 1–14. https://doi. Bodeau-Livinec, F. (2016). Mental health of college students and org/10.1353/csd.2015.0001 their non-college-attending peers: Results from a large French Hanel, P. H., & Vione, K. C. (2016). Do student samples provide an ac- cross-sectional survey. BMC Psychology, 4(1), 20. https://doi. curate estimate of the general public? PLoS ONE, 11(12), e0168354. org/10.1186/s4035​9-016-0124-5 https://doi.org/10.1371/journ​al.pone.0168354 Kraemer, H. C., & Kupfer, D. J. (2006). Size of treatment effects Harrer, M., Adam, S. H., Baumeister, H., Cuijpers, P., Karyotaki, E., and their importance to clinical research and practice. Biological Auerbach, R. P., … Ebert, D. D. (2018). Internet interventions Psychiatry, 59(11), 990–996. https://doi.org/10.1016/j.biops​ ych. for mental health in university students: A systematic review and 2005.09.014 meta-analysis. International Journal of Methods in Psychiatric Lovibond, P. F., & Lovibond, S. H. (1995). The structure of negative Research, 28, e1759. https://doi.org/10.1002/mpr.1759 emotional states: Comparison of the Depression Anxiety Stress Heber, E., Ebert, D. D., Lehr, D., Cuijpers, P., Berking, M., Nobis, S., Scales (DASS) with the Beck Depression and Anxiety Inventories. & Riper, H. (2017). The benefit of web- and computer-based inter- Behaviour Research and Therapy, 33(3), 335–343. https://doi. ventions for stress: A systematic review and meta-analysis. Journal org/10.1016/0005-7967(94)00075​-U of Medical Internet Research, 19(2), e32. https://doi.org/10.2196/ McGorry, P. D., Purcell, R., Goldstone, S., & Amminger, G. P. (2011). jmir.5774 Age of onset and timing of treatment for mental and substance use Higgins, J., & Green, S. (2011). Cochrane handbook for systematic re- disorders: Implications for preventive intervention strategies and views of interventions (Vol. 4). John Wiley & Sons. models of care. Current Opinion in Psychiatry, 24(4), 301–306. Higgins, J. P. T., Thompson, S. G., Deeks, J. J., & Altman, D. G. https://doi.org/10.1097/YCO.0b013​e3283​477a09 (2003). Measuring inconsistency in meta-analyses. British Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G. & The PRISMA Medical Journal, 327(7414), 557–560. https://doi.org/10.1136/ Group (2009). Academia and clinic annals of internal medicine bmj.327.7414.557 preferred reporting items for systematic reviews and meta-anal- Hunt, J., & Eisenberg, D. (2010). Mental health problems and help-seek- yses. Annals of Internal Medicine, 151(4), 264–269. https://doi. ing behavior among college students. Journal of Adolescent Health, org/10.1371/journ​al.pmed1​000097 46(1), 3–10. https://doi.org/10.1016/j.jadoh​ealth.2009.08.008 Mortier, P., Cuijpers, P., Kiekens, G., Auerbach, R. P., Demyttenaere, Iglesias, S. L., Azzara, S., Argibay, J. C., Arnaiz, M. L., Carpineta, M. K., Green, J. G., … Bruffaerts, R. (2017). The prevalence of suicidal D. V., Granchetti, H., & Lagomarsino, E. (2012). Psychological and thoughts and behaviours among college students: A meta-analysis. physiological response of students to different types of stress man- Psychological Medicine, 48(4), 554–565. https://doi.org/10.1017/ agement programs. American Journal of Health Promotion, 26(6), s0033​29171​7002215 149–159. https://doi.org/10.4278/ajhp.11051​6-QUAL-199 Orsini, N., Higgins, J., Bottai, M., & Buchan, I. (2013). Heterogi: Stata Ioannidis, J. P. A., Patsopoulos, N. A., & Evangelou, E. (2007). module to quantify heterogeneity in a meta-analysis. Retrieved from Uncertainty in heterogeneity estimates in meta-analyses. British http://fmwww.bc.edu/repec​/bocod​e/h/heter​ogi.ado  | 14682850, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cpsp.12342 by Fernuniversitaet Hagen, Wiley Online Library on [07/08/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License AMANVERMEZ et al.    13 of 27 Öst, L. G. (2008). Efficacy of the third wave of behavioral therapies: Van Daele, T., Hermans, D., Van Audenhove, C., & Van den Bergh, O. A systematic review and meta-analysis. Behaviour Research and (2011). Stress reduction through psychoeducation. Health Education Therapy, 46(3), 296–321. https://doi.org/10.1016/j.brat.2007.12.005 & Behavior, 39(4), 474–485. https://doi.org/10.1177/10901​98111​ Peterson, R. A. (2001). On the use of college students in social sci- 419202 ence research: Insights from a second-order meta-analysis. van der Klink, J. J. L., Blonk, R. W. B., Schene, A. H., & Van Dijk, F. Journal of Consumer Research, 28(3), 450–461. https://doi. J. H. (2001). The benefits of interventions for work-related stress. org/10.1086/323732 American Journal of Public Health, 91(2), 270–276. https://doi. Regehr, C., Glancy, D., & Pitts, A. (2013). Interventions to reduce org/10.2105/AJPH.91.2.270 stress in university students: A review and meta-analysis. Journal Wieland, L. S., Manheimer, E., & Berman, B. M. (2011). Development of Affective Disorders, 148(1), 1–11. https://doi.org/10.1016/j. and classification of an operational definition of complementary and jad.2012.11.026 alternative medicine for the Cochrane Collaboration. Alternative Richardson, K. M., & Rothstein, H. R. (2008). Effects of occupational Therapies in Health and Medicine, 17(2), 50–59. stress management intervention programs: A meta-analysis. Journal Zoogman, S., Goldberg, S. B., Hoyt, W. T., & Miller, L. (2015). of Occupational Health Psychology., 13(1), 69–93. https://doi. Mindfulness interventions with youth: A meta-analysis. Mindfulness, org/10.1037/1076-8998.13.1.69 6(2), 290–302. https://doi.org/10.1007/s1267​1-013-0260-4 Sampl, J., Maran, T., & Furtner, M. R. (2017). A randomized controlled pilot intervention study of a mindfulness-based self-leadership train- ing (MBSLT) on stress and performance. Mindfulness, 8(5), 1393– How to cite this article: Amanvermez Y, 1407. https://doi.org/10.1007/s1267​1-017-0715-0 Rahmadiana M, Karyotaki E, et al. Stress management Sarason, I. G., Johnson, J. H., & Siegel, J. M. (1978). Assessing the interventions for college students: A systematic impact of life changes: Development of the Life Experiences Survey. review and meta-analysis. Clin Psychol Sci Pract. Journal of Consulting and Clinical Psychology, 46(5), 932–946. https://doi.org/10.1037/0022-006X.46.5.932 2020;00:e12342. https://doi.org/10.1111/cpsp.12342 Slavich, G. M. (2016). Life stress and health. Teaching of Psychology, 43(4), 346–355. https://doi.org/10.1177/00986​28316​662768 APPENDIX A Participants S E A RCH STR INGS FO R T HE "students, dental"[MeSH Terms] OR "Students, SYSTE M ATIC L IT E R AT U R E Medical"[MeSH Terms] OR "Students, Nursing"[MeSH R EV IEW Terms] OR "Students, Pharmacy"[MeSH Terms] OR "Students, Premedical"[MeSH Terms] OR "Students, Public Search Strings for PubMed Health"[MeSH Terms] OR "undergraduate students"[All fields] OR "undergraduate student"[All fields] OR "undergraduates"[All fields] OR "undergraduate"[All Intervention fields] OR "university students"[All fields] OR "uni- versity student"[All fields] OR "college students"[All "stress"[All Fields] OR "Stress, Psychological"[MeSH Terms] fields] OR "college student"[All fields] OR "professional AND ("counseling"[All Fields] OR "development"[All students"[All Fields] OR "professional student"[All fields] Fields] OR "handling"[All Fields] OR "intervention"[All OR "health professional students"[All Fields] OR "health Fields] OR "management"[All Fields] OR "managing"[All professional student"[All fields] OR "dental students"[All Fields] OR "modification"[All Fields] OR "modifying"[All fields] OR "dental student"[All fields] OR "medical Fields] OR "program"[All Fields] OR "programme"[All students"[All fields] OR "medical student"[All fields] OR Fields] OR "psychoeducation"[All Fields] OR "reducing"[All "nursing students"[All fields] OR "nursing student"[All Fields] OR "reduction"[All Fields] OR "skill"[All Fields] fields] OR "pharmacy students"[All fields] OR "pharmacy OR "skills"[All Fields] OR "skills training"[All Fields] student"[All fields] OR "premedical students"[All fields] OR "teaching"[All Fields] OR "technique"[All Fields] OR OR "premedical student"[All fields] OR "public health "techniques"[All Fields] OR "training"[All Fields] OR students"[All fields] OR "public health student"[All fields] "treating"[All Fields] OR "treatment"[All Fields]) OR "engineering students"[All fields] OR "engineering  | 14682850, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cpsp.12342 by Fernuniversitaet Hagen, Wiley Online Library on [07/08/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 14 of 27    AMANVERMEZ et al. student"[All fields] OR "mental health students"[All fields] "pharmacy students" OR "pharmacy student" OR "premed- OR "mental health student"[All fields] OR "bachelor's de- ical students" OR "premedical student" OR "public health gree students"[All fields] OR "master students"[All fields] students" OR "public health student" OR "engineering stu- OR "phd students"[All fields] OR "phd student"[All fields] dents" OR "engineering student" OR "mental health stu- OR "postgraduate students"[All fields] OR "postgradu- dents" OR "mental health student" OR "bachelor's degree ate student"[All fields] OR "postgraduates"[All fields] students" OR "master students" OR "phd students" OR OR "postgraduate"[All fields] OR "higher education "phd student" OR "postgraduate students" OR "postgradu- students"[All fields] OR "higher education student"[All ate student" OR "postgraduates" OR "postgraduate" OR fields] OR "tertiary education students"[All fields] OR "higher education students" OR "higher education student" "tertiary education student"[All fields] OR "college OR "tertiary education students" OR "tertiary education population"[All Fields] OR "university population"[All student" OR "college population" OR "university popu- Fields] OR "college sample"[All Fields] OR "univer- lation" OR "college sample" OR "university sample" OR sity sample"[All Fields] OR "freshman"[All Fields] OR "freshman" OR "freshmen" OR "sophomores" OR "sopho- "freshmen"[All Fields] OR "sophomores"[All Fields] more" OR "first year students" OR "first year student" OR OR "sophomore"[All Fields] OR "first year students"[All "first-year students" OR "first-year student" OR "young Fields] OR "first year student"[All Fields] OR "first-year adult" OR "young adults" OR "social work student" OR students"[All Fields] OR "first-year student"[All Fields] "social work students" OR "foreign student" OR "foreign OR "young adult"[All Fields] OR "young adults"[All students" OR "international student" OR "international Fields] OR "social work student"[All Fields] OR "so- students" OR "junior students" OR "junior student" OR cial work students"[All Fields] OR "foreign student"[All "senior students" OR "senior student" OR DE "undergrad- Fields] OR "foreign students"[All Fields] OR "interna- uate students" OR DE "college students" OR DE "dental tional student"[All Fields] OR "international students"[All students" OR DE "medical students" OR DE "nursing stu- Fields] OR "junior students"[All Fields] OR "junior dents" OR DE "postgraduate students" OR DE "foreign student"[All Fields] OR "senior students"[All Fields] OR students" OR DE "international students" OR DE "busi- "senior student"[All Fields] OR "business students"[All ness students" OR "business students" OR "business stu- Fields] OR "business student"[All Fields] OR "graduate dent" OR DE "graduate students" OR "graduate students" student"[All Fields] OR "graduate students"[All Fields] OR OR "graduate student" OR DE "law students" OR "law stu- "law students"[All Fields] OR "law student"[All Fields] dents" OR "law student" AND filtered by RCT, Species: Humans, Languages: AND filtered by English, adulthood and young adult- English, Ages: Adult: 19+, Young Adult:19–24, Adult: hood, human, clinical trial, treatment outcome, exclude 19–44 dissertations Search Strings for Cochrane Search Strings for PsycInfo #1 stress, emotional Intervention #2 stress, psychological "stress" OR DE "stress" AND ("counseling" OR "develop- #3 student* ment" OR "handling" OR "intervention" OR "management" #4 Counseling OR "managing" OR "modification" OR "modifying" OR #5 Developing "program" OR "programme" OR "psychoeducation" OR "re- #6 Development ducing" OR "reduction" OR "skill" OR "skills" OR "skills #7 Handling training" OR "teaching" OR "technique" OR "techniques" #8 Intervention OR "training" OR "treating" OR "treatment") #9Management #10 Managing Participants #11Modification "undergraduate students" OR "undergraduate student" OR #12 Modifying "undergraduates" OR "undergraduate" OR "university stu- #13Program dents" OR "university student" OR "college students" OR #14Programme "college student" OR "professional students" OR "pro- #15Psychoeducation fessional student" OR "health professional students" OR #16Reducing "health professional student" OR "dental students" OR #17Reduction "dental student" OR "medical students" OR "medical stu- #18Skill dent" OR "nursing students" OR "nursing student" OR #19Skills  | 14682850, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cpsp.12342 by Fernuniversitaet Hagen, Wiley Online Library on [07/08/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License AMANVERMEZ et al.    15 of 27 #20Skills training student' OR 'nursing students'/exp OR 'nursing students' OR #21Teaching 'pharmacy student'/exp OR 'pharmacy student' OR 'phar- #22Technique macy students'/exp OR 'premedical students' OR 'premedi- #23Techniques cal student'/exp OR 'premedical student' OR 'premedical #24Training students'/exp OR 'premedical students' OR 'Public Health #25Treating student'/exp OR 'Public Health student' OR 'Public Health #26Treatment students' OR 'undergraduate student'/exp OR 'undergradu- ( #1 or #2) AND #3 AND (#4 or #5 or #6 or #7 or #8 or ate student' OR 'undergraduate students' OR 'undergradu- #9 or #10 or #11 or #12 or #13 or #14 or #15 or #16 or #17 ates' OR 'undergraduate'/exp OR 'undergraduate' OR 'health or #18 or #19 or #20 or #21 or #22 or #23 or #24 or #25 professional students' OR 'health professional student' OR or #26) 'engineering students' OR 'engineering student' OR 'mental Limits: Trials