This systematic review was conducted and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020) guidelines [11].

In December 2023, thorough searches of the Scopus, PubMed, and PsycINFO databases were conducted to identify pertinent literature. Study selection was conducted jointly by two authors. Titles, abstracts, and full texts were assessed against the predefined inclusion and exclusion criteria. Any uncertainties regarding study eligibility were discussed among the authors and resolved by consensus. Articles featuring a convergence of two key concepts were evaluated for suitability for inclusion. Concept 1 pertains to mental wellbeing and concept two focuses on the higher education sector. Keywords used for concept one: mental health, stress management. Keywords used for concept two: higher education, HEI, university. Each set of keywords within a concept was separated by using OR, while concepts themselves were linked by using AND. Database-specific filters were applied where appropriate. In Scopus, filters included subject area, publication year, language, and publication stage. In PubMed, publication type, publication year, and English-language filters were applied. In PsycINFO, searches were restricted to peer-reviewed journal publications and publication years 2015–2023. In total, 5997 records were identified, of which 3443 duplicates were removed, leaving 2554 records. These records were refined according to the predefined eligibility criteria. A total of 1833 studies were excluded due to discipline filters, 42 due to language filters, 13 due to publication stage filters, and 221 due to publication date filters. Forty reports could not be retrieved, leaving 405 reports for eligibility assessment. After a careful review of the title and the abstract of the located literature, manual filtering was applied to select the most relevant pieces. This rigorous process yielded a final list of 95 selected articles. The 95 identified research papers were further filtered to studies that discuss stress management interventions at HEIs. The study selection process is illustrated in Figure 1. Systematic reviews and feasibility studies were excluded because the objective of this study was to synthesize evidence from primary empirical studies evaluating stress management interventions. Besides, the review was limited to peer-reviewed publications indexed in the selected databases. Gray literature, including dissertations, conference proceedings, and institutional reports, was excluded to ensure the inclusion of studies that had undergone formal peer review and quality assessment.

Inclusion criteria: 1) published between 2015 and 2023, 2) published in a final publication stage, 3) written in English, 4) discuss stress management interventions at HEIs and explore potential impediments to effective stress management practices. Exclusion criteria: 1) literature published prior to 2015, 2) literature published in languages other than English, 3) systematic reviews, 4) feasibility studies. Based on the selection according to the inclusion and exclusion criteria, 32 [12–25], [26–43] publications were chosen for this systematic literature review.

Regarding the limitation of our review to the period between 2015 and 2023, we selected this timeframe to focus on recent research to provide an up-to-date synthesis of stress management interventions within HEIs. Given the dynamic nature of mental health awareness and interventions in academic settings, focusing on the most recent years allowed us to capture the latest contributions in the field. This timeframe aligns with significant developments in mental health policy and initiatives, including the Joint Action for Mental Health and Wellbeing (JA MH-WB, representing a concerted effort involving 51 partners from 28 EU Member States and several European organizations; and the Comprehensive Mental Health Action Plan ¹ devised by the World Health Organization, both initiated in 2013. Given the launch of these initiatives in 2013 and their subsequent impact on mental health policies and practices, our decision was to commence the review from 2015 onwards. We acknowledge that this timeframe may exclude relevant studies published before 2015, and we have noted this limitation in our review.

Since meta-analysis of the selected articles was not feasible due to differences in research design, such as objectives and measurement methods, this article opts for data synthesis. Data extraction was conducted by the first author and reviewed by the co-authors to ensure consistency and accuracy. Any discrepancies were resolved through discussion. Results from individual studies were tabulated and visually displayed to facilitate comparison and synthesis - detailed in Table 1. The methodological quality of included studies was evaluated by using the Risk of Bias 2.0 (RoB 2.0) ² tool for randomized studies–detailed in Table 2 and the ROBINS-I ³ tool for non-randomized studies–detailed in Table 3.

The data synthesis focuses on the primary and secondary outcomes of the identified studies. Data synthesis focused on study population, sample size, main objectives, type and length of interventions, outcome measures, effectiveness, and methodological rigor.

Data synthesis was conducted by the first author after joint consultation sessions with the corresponding authors. The consultations aimed at ensuring consensus on the synthesis process.

To synthesize the impact of the findings, a narrative approach was adopted. The choice of a narrative synthesis was justified by the need to accommodate diverse study designs and outcome measures, which precluded a quantitative meta-analysis.

The selection of articles comprised 31 quantitative studies [12–22], [24–43] and one qualitative study [23], utilizing various study designs. Predominantly, the designs consisted of randomized controlled trials [12–17], [19–22], [25, 26, 28], [30–33], [35], [37–40], followed by quasi-experimental pre-post assessments [18, 24, 34, 43], and cross-sectional studies [29, 36, 41, 42]. Additionally, the review encompasses one interview-based qualitative study [23] and one questionnaire-based pilot study [27], as indicated.

This section summarizes the methodological characteristics, strengths, and limitations of the included studies and provides context for interpreting the findings.

Study population: The study population comprises students and academic staff in HEIs across various countries, including China [17, 41, 42], the US [20, 21, 25, 28, 32], Nigeria [26, 39], Iran [12], Australia [19, 23, 31, 34, 36], India [35], Brazil [37], Germany [13], Japan [24], Spain [29, 30], Turkey [43], Portugal [16], and the UK. [15, 22, 33]. Additionally, two studies feature an international sample [18, 27]. Among the included studies, the primary focus is on students’ mental health [12–23], [25, 27, 28]^, [30–35], [37, 38, 40–43], with a limited number examining the experiences and perspectives of university staff and lecturers [24, 26, 29, 36, 39]. Sample sizes vary across different groups, ranging from 16 [23] to 11954 [42]. The studies were published across a span of eight years, with the most recent publications from 2023, indicating a sustained interest in the subject matter over time.

The main aim of the studies: The primary objective of the included studies was to evaluate the effectiveness of interventions designed to reduce stress, enhance coping, and promote mental wellbeing among students and staff in HEIs. Through various methodologies and approaches, the studies sought to explore the efficacy of interventions aimed at alleviating stress symptoms, enhancing coping mechanisms, and ultimately fostering a positive psychological state among participants.

The nature of the interventions: Interventions in the reviewed studies are categorized into five groups, each targeting stress management and mental wellbeing through distinct approaches, as outlined in Table 4. These categories encompass traditional therapeutic interventions [12, 24, 39], technology-based interventions [14, 18, 21, 23, 25, 28, 31, 33, 35, 40], lifestyle-based interventions [26, 29, 41, 42], social and community-based interventions [36, 43], and alternative and complementary approaches [13, 15–17, 19, 20, 22, 27], [30–32, 34, 37, 38]. Each category offers unique strategies to address stress and promote mental health across various populations and settings.

The length of the interventions: The duration of interventions varied considerably across the reviewed studies. Interventions ranged from brief one-time questionnaires [41] to more extended programs spanning several months. Some interventions were conducted over short periods, such as a couple of minutes [15, 24, 35], days [27, 37], while others over several weeks [12, 14, 16–18, 20–22, 25], [26, 28, 30–33], [38, 39, 42, 43], or an entire academic semester [13, 19, 23, 34, 40]. Additionally, intervention lengths varied based on the format, with some consisting of weekly sessions over several weeks, while others were condensed into intensive workshops or single sessions.

Outcome measures: Table 5 offers a comprehensive overview of the outcome measures employed throughout the studies under review, encompassing a wide range of categories. These categories include psychological indicators, engagement metrics, mental wellbeing assessments, trust and alliance evaluations, acceptance and competence assessments, university adjustment and coping measures, and various other psychological measures.

A detailed overview of the outcome assessment instruments and assessors of the included studies is presented in Table 1. Most studies assessed outcomes using validated self-report surveys. Frequently used instruments included the Perceived Stress Scale, Warwick-Edinburgh Mental Wellbeing Scale, Depression Anxiety Stress Scale, General Help-Seeking Questionnaire, and Five Facet Mindfulness Questionnaire. Only a limited number of studies supplemented self-reported outcomes with other, non-self-reported measures. These include salivary cortisol assessment, plasma cortisol analysis, computerized performance tasks, laboratory-based physiological measures, or qualitative interview analysis.

While the selected studies employ varying research designs–explored in Table 1, including randomized controlled trials [12–17, 19–22, 25, 26, 28], [30–33], [35], [37–40] quasi-experimental pre-post assessments [18, 24, 34, 43], cross-sectional studies [29, 36, 41, 42], one questionnaire-based pilot study [27], and one interview-based qualitative study [23], there are notable differences in methodological rigor across the literature. Some studies lack experimental designs [27, 36, 41], relying instead on one-time questionnaires or surveys, which may limit the ability to establish causal relationships or assess long-term intervention effects. Additionally, the use of self-report measures [17, 18, 20–22, 25, 26], [28–30, 38, 39, 42] without validation through factor analysis or physiological measures may introduce reporting biases and limit the robustness of findings. Moreover, several studies exhibit unreasonably small sample sizes [16, 23, 27, 32, 35, 37], which may reduce the generalizability and statistical power of results. Furthermore, the absence of control groups [19, 23, 24, 27, 29, 34, 36, 41, 42] or the use of waitlist controls [18, 26, 28, 30, 31] in some studies may introduce potential biases in the assessment of intervention effects. Gender imbalances in samples are also noted in more studies [16, 20, 23, 25, 30], which could impact the validity of the results. Furthermore, the absence of follow-up assessments in many studies prevented the evaluation of sustained intervention effects.

Summarizing the quality of study design across the included literature reveals variability in methodological rigor. Studies employing rigorous designs such as randomized controlled trials were deemed to have higher methodological quality due to their ability to establish causality and minimize bias. Quasi-experimental pre-post assessments and cross-sectional studies were also considered, albeit with some limitations in terms of causal inference and generalizability. Studies with larger sample sizes were considered to have higher methodological quality, as they are more likely to detect significant effects and provide more representative insights into the target population. Similarly, studies incorporating control groups were perceived to have higher methodological quality, as they allow for comparisons between intervention and control conditions, thereby strengthening causal inference. Furthermore, studies conducting follow-up assessments were regarded as having higher methodological quality, as they provide insights into the durability of intervention effects beyond the immediate post-intervention period.

To assess the risk of bias, randomized controlled trials were assessed using the Risk of Bias 2.0 (RoB 2.0) tool, developed by Cochrane. The RoB 2.0 is designed to evaluate the risk of bias of randomized studies in 5 domains, including the randomization process, deviations from intended interventions, missing outcome data, measurement of the outcome, and selection of the reported results. In each domain several questions must be answered by selecting one of the five potential answers: yes, probably yes, no, probably no and not enough information. Overall, each domain is evaluated as low risk of bias, some concerns or high risk of bias. Based on the evaluation of the five domains, a final judgement of overall risk of bias is calculated by the algorithm.

For non-randomized studies the risk of bias was assessed by using the Risk of Bias in Non-Randomized Studies of Interventions (ROBINS-I) tool, developed by Cochrane. This tool evaluates the risk of bias in 7 domains, including bias due to confounding, bias due to selection, bias due to the classification of the intervention, bias due to deviations from the intended intervention, bias due to missing data, bias in outcome measurement, and bias in the selection of reported results. Each of these domains is assessed as low risk, moderate risk, serious risk or critical risk of bias.

The risk-of-bias assessment was initially conducted by the first author using the RoB 2.0 and ROBINS-I tools. Assessments that were considered uncertain or difficult to classify were subsequently reviewed and discussed with a co-author, and final judgments were reached through consensus. Out of the twenty-two randomized controlled trials evaluated, ten [16, 17, 21, 22, 26, 30, 32, 33, 37, 39] were found to have a low risk of bias. Five studies had some concerns regarding bias [14, 20, 25, 28, 40], while seven [12, 13, 15, 19, 31, 35, 38] were deemed to have a high risk of bias, necessitating cautious interpretation of their findings. Out of the ten non-randomized studies evaluated, seven [18, 24, 27, 29, 34, 36, 41] were found to have a serious risk of bias, which may affect confidence in their findings. Three studies [23, 42, 43] were assessed as having a moderate risk of bias, suggesting some concerns in the devised methodology. An overview of the risk of bias assessments can be found in Table 1.

Interventions targeting stress and mental wellbeing at HEIs have been subject to rigorous evaluation through randomized controlled trials and empirical studies. Xu and Choi [41] uncovered a significant positive correlation between engagement in cultural and artistic activities and heightened life satisfaction among university students, highlighting unconventional avenues for bolstering psychological resilience. Frazier et al [21] demonstrated the efficacy of web-based stress management interventions during the COVID-19 pandemic, underscoring the pivotal role of accessible online platforms in alleviating stress symptoms among college students. Similarly, Igu et al [26] emphasized the value of holistic approaches by integrating cognitive behavioral therapy (CBT) with yoga to reduce job stress among university lecturers, showcasing promising strategies for addressing occupational stressors within academic environments. Abbasi et al [12] investigated stress management training for medical students, which yielded notable reductions in both physical complaints and psychological distress, reaffirming the importance of targeted interventions in fostering mental wellbeing within educational contexts. Complementing these findings, Gilmore et al [23] considered the challenges faced by participants in a telephone crisis support training program, emphasizing the necessity of comprehensive support mechanisms for those involved in crisis intervention roles. These studies, alongside others such as Hashimoto et al [24] and Sousa et al, [37] collectively highlight the diverse yet interconnected approaches and positive outcomes of interventions, underscoring the ongoing imperative for research and implementation efforts in promoting mental health within university communities.

Overall, mindfulness-based interventions and CBT-based approaches demonstrated the most consistent positive outcomes across the reviewed studies. Interventions supported by randomized controlled trials and lower risk of bias assessments provided stronger evidence of effectiveness in reducing stress, anxiety, and depression. In contrast, interventions assessed through cross-sectional or non-randomized designs provided less conclusive evidence regarding effectiveness. These findings suggest that mindfulness-oriented and CBT-based interventions currently represent the most promising approaches for promoting mental wellbeing in HEIs, although further longitudinal research is required to confirm sustained effectiveness.

A notable observation emerging from the assessment of intervention impact is the prevalent focus on students within HEIs, with a lack of interventions tailored specifically for academic and administrative staff. The available evidence also suggests differences in the focus of interventions across population groups. Student-targeted interventions primarily addressed individual psychological outcomes, including stress, anxiety, depression, mindfulness, wellbeing, resilience, and help-seeking behaviours. In contrast, the smaller number of studies involving university staff tended to focus on occupational outcomes such as job stress, psychological strain, burnout, job satisfaction, organizational commitment, and workplace coping. These findings suggest that students and staff may experience different stressors and support needs, highlighting the importance of developing and evaluating interventions tailored to the specific challenges faced by each group. Given that staff wellbeing is equally important to the overall functioning of HEIs, future research should prioritize the development and evaluation of interventions for academic and administrative staff to support a more comprehensive and inclusive approach to mental health promotion within higher education settings.

Another observation worth mentioning is the limited geographical diversity in the locations where the interventions were conducted and evaluated. While the selected studies encompass a range of countries, most of the interventions are still predominantly concentrated in Western contexts, with no representation from regions such as Central and Eastern Europe. This geographical bias raises questions about the generalizability of findings and the applicability of intervention strategies across diverse cultural and socio-economic contexts. The lack of representation from regions with distinct educational systems, cultural norms, and socioeconomic conditions underscores the need for more inclusive and culturally sensitive approaches to intervention development and evaluation. Future research should strive to incorporate a more diverse range of settings and populations, including those from Central and Eastern Europe, to ensure that interventions are effective and relevant across different contexts, ultimately contributing to more equitable and inclusive mental health support within higher education globally.

This review synthesized evidence on stress management interventions implemented within HEIs. The findings suggest that a wide range of intervention approaches have been applied, including mindfulness-based programmes, cognitive-behavioural interventions, psychoeducational approaches, and skills-development programmes. Overall, the reviewed studies suggest that such interventions may contribute to improved stress management and mental wellbeing among participants. However, considerable heterogeneity was observed across intervention types, study designs, intervention duration, outcome measures, and sample characteristics, limiting direct comparisons across studies. Predominantly, studies utilized randomized controlled trials. Intervention lengths varied widely, from brief single-session programmes to extended interventions. Sample sizes also varied substantially, with some studies including relatively small samples that may limit generalizability and statistical power. Furthermore, many studies relied on self-report measures, which may affect the robustness of findings. Most studies relied exclusively on self-reported outcomes, whereas only a small number of incorporated non-self-reported measures. Among these studies, three were rated as having a low overall risk of bias, one as moderate risk, and one as high risk. Although this may suggest a tendency toward stronger methodological rigor among some studies employing more diverse assessment approaches, the small number of such studies does not allow for firm conclusions.

The findings of the present review are broadly consistent with previous systematic reviews conducted in higher education settings. Regehr et al. reported that cognitive, behavioral, and mindfulness-based interventions were effective in reducing student stress [44]. Amanvermez et al. found moderate effects of stress management interventions on stress, depression, and anxiety among college students [45]. Similarly, Worsley et al. highlighted the positive effects of mindfulness- and cognitive behavioural-based approaches on student mental health and wellbeing [46]. The consistency of findings across previous reviews and the present review suggests that mindfulness-based and cognitive-behavioral approaches represent the most promising evidence-based strategies currently available for stress management within higher education settings.

This review identified a notable lack of studies targeting academic and administrative staff. This highlights an important gap in the current evidence base and suggests that future intervention research should extend beyond student populations. The available evidence also suggests differences in the focus of interventions across population groups. Student-targeted interventions primarily addressed individual psychological outcomes, including stress, anxiety, depression, mindfulness, wellbeing, resilience, and help-seeking behaviours. In contrast, the smaller number of studies involving university staff tended to focus on occupational outcomes such as job stress, psychological strain, burnout, job satisfaction, organizational commitment, and workplace coping. These findings indicate that students and staff experience different stressors and support needs, highlighting the importance of developing and evaluating interventions tailored to the specific challenges faced by each group.

This systematic literature review has limitations. The selected databases may not have captured all relevant literature and primarily index peer-reviewed publications; consequently, grey literature (e.g., conference proceedings, dissertations, unpublished studies, and institutional reports) was excluded, potentially increasing publication bias. In addition, only English-language studies were included, and the keyword-based search strategy may have overlooked relevant studies using alternative terminology. Besides, the risk-of-bias assessments were initially conducted by one author, although uncertain cases were reviewed and discussed with the co-authors to reach consensus, the initial assessment by a single author, may have increased the risk of subjective judgment. Furthermore, substantial heterogeneity was observed across intervention types, study designs, outcome measures, outcome assessment instruments, outcome assessors, and reporting practices. Information regarding intervention delivery, participant adherence, intervention fidelity, and assessment time points was often inconsistently reported across the included studies. As a result, direct comparisons across studies were limited, and a quantitative meta-analysis was not feasible. The findings of this review are based on a narrative synthesis of the reported outcomes.