Understanding university student priorities for mental health and well‐being support: A mixed‐methods exploration using the person‐based approach

Abstract Poor student well‐being at UK universities is overstretching institutional support services, highlighting a need for effective new resources. Despite extensive literature on mental health and well‐being interventions, students' engagement with support remains unexplored. The study aimed to understand students' experience of engagement with well‐being support, identify their well‐being needs and form concrete recommendations for future intervention design and delivery. The Person‐Based Approach to intervention design was followed to centralise users' experience, in turn maximising acceptability and effectiveness of resources. An online survey (N = 52) was followed by three focus groups (N = 14). Survey data were analysed descriptively, and reflexive thematic analysis was performed on qualitative data. Mixed‐methods data integration produced four key student priorities for well‐being resources – ease of access, inclusive and preventative approach, sense of community and a safe space, and applying skills to real‐life contexts. Five actionable guiding principles for intervention design were produced through consultation with expert stakeholders. This work helps understand why and how students engage with support at university. The resulting recommendations can inform future intervention development, leading to more acceptable, engaging and effective student well‐being resources.

is significant; Student suicide rate of 4.7 per 100,000 equates to 95 deaths every year (Office for National Statistics [ONS], 2018).
Most higher education institutions (HEIs) provide student wellbeing support through dedicated welfare teams, and virtually all are reporting an increased demand for well-being services -six in 10 HEIs reported demand increasing by over 25% within a 5-year period (Thorley, 2017). With limited funding available (Universities UK [UUK], 2018), universities are struggling to meet these rising demands, leading to routine long waiting times before students can access help (Gallagher, 2014). Failing to address the root causes of poor well-being and equip students with efficient coping strategies while at university can exacerbate poor mental health (Biasi et al., 2017).
Furthermore, the ongoing COVID-19 pandemic negatively impacted the university experience, including the provision of teaching and well-being services (Burns et al., 2020). This coincided with a marked decrease in student well-being and increased prevalence of clinical-level depressive symptoms among the group to over 30% (Evans et al., 2021). To break this pattern, policymakers have called for HEIs to make student welfare a strategic priority, emphasising early treatment and prevention of mental illness (House of Commons, 2020; Office for Students [OfS], 2019;UUK, 2015UUK, , 2018UUK, , 2020. Meta-analyses suggest that preventative approaches can safeguard against detrimental effects of poor psychological health (Conley et al., 2017;Reavley & Jorm, 2010), however, there is still substantial scope for improved implementation, such as using digital technology to improve access (Conley et al., 2016).
Recent systematic reviews report that a range of psychological interventions can be effective in improving student wellbeing, including cognitive-behavioural and mindfulness-based techniques (Breedvelt et al., 2019;Huang et al., 2018; What Works Centre for Well-being, 2020). However, there is little focus on uptake and effective engagement (i.e., how and why target users engage with the resource; Yardley et al., 2016) -despite insufficient engagement with intervention content reducing the effectiveness of interventions.
Student mental health and well-being interventions often report low engagement across intervention approaches and delivery types, with most distressed individuals even less likely to seek help (the helpnegation effect; Goodwin et al., 2016;Rickwood et al., 2005). Systematic reviews suggest that online resources may improve engagement with student help-seeking, although note that more high-quality evidence is needed (Kauer et al., 2014). This highlights the need for better understanding of the student experience and greater attention to the practical side of intervention design, which is crucial for implementation success and effectiveness (Michie et al., 2018). For example, students like online well-being resources for their accessibility but are sceptical of their potential for communication and human connection (Chan et al., 2016).
In recent decades, the field of behaviour change has recognised the value of target user involvement in developing the most effective and useable interventions (Wicks et al., 2018). The Person-Based Approach (PBA) to intervention design (PBA; Yardley, Morrison, et al., 2015) is a framework for centralising users' experience throughout planning and development of interventions, via systematic and iterative exploration of target users' perspectives and tailoring of the intervention accordingly. It aims to enhance theoryand evidence-based approaches to intervention development, enabling researchers to understand the users' experience of the intervention in a psychosocial context, anticipate (and pre-empt) potential barriers to engagement, and improve clarity in the interpretation of outcomes. Resulting alterations make interventions persuasive, feasible and relevant to users (Yardley et al., 2016), translating into greater effectiveness. In practice, adopting the PBA entails a focus on incorporating qualitative or mixed methods research with a representative sample of the target population. In the first instance, an in-depth understanding of behavioural facilitators and barriers from the users' perspective allows the formation of guiding principles -a concise list of key intervention design objectives paired with the design features through which the intervention will meet them (Yardley, Ainsworth, et al., 2015). The guiding principles provide the foundation for further intervention development and evaluation of complex interventions (Medical Research Council [MRC], 2019). The present study exemplifies this initial part of the PBA process.

| Research aims
Given the concerns regarding mental health and well-being at HEIs, there is need for preventative, scalable and student-accepted approaches to help with student well-being (UUK, 2018(UUK, , 2020. Indepth exploration of how and why students effectively engage with support allows for the design of optimally acceptable and effective future interventions (Yardley, Ainsworth, et al., 2015).
The aim of the present project was to inform the development of interventions improving student well-being, through understanding students' experience of engagement with well-being resources, determining students' well-being needs and establishing guiding principles for the design of student well-being interventions.

| Design
The study utilised a parallel mixed-methods design. A series of qualitative focus groups was complemented by a predominantly quantitative online survey, then findings from both methodologies combined to produce recommendations for future intervention development. A mixed-methods approach-previously used in developing well-being resources tailored to specific populations (e.g., Simpson and Mercer's (2020) mindfulness-based intervention for people with multiple sclerosis)-provided detailed insight into students' experience through focus groups, paired with the assurance that the identified priorities were echoed by a larger survey sample.

Stakeholder
consultations were used to ensure that REMSKAR ET AL.
-777 recommendations were appropriate and practical for service developers and providers to implement, in line with best practice for intervention co-development (Pottie et al., 2021). The research was carried out from a critical realist philosophical position-the researchers presumed that students' priorities for efficient well-being services exist and can be identified, albeit through the inevitably constructed understanding of social reality (Maxwell, 2012).

| Setting
The project was carried out at a research-intensive HEI based in a mid-sized city in South-West England. The institution provides students with varied well-being support, ranging from group to individual, online and in-person (pre-pandemic), grounded in approaches including cognitive behavioural therapy, mindfulness-based techniques and social activities (e.g., gardening for well-being; University of Bath, 2020).

| Participants
Participants were current students at the University of Bath over  (Thompson, 2007). Participants rate, on a scale from 1 to 5, to what extent they felt each of 20 sentiments, such as 'Enthusiastic' and 'Afraid', over the past 2 weeks.
10 positive and 10 negative items are summed up separately to give scores of positive and negative affect, respectively. Higher scores indicate stronger affect (range 10-50 each).
The survey aimed to assess the key facilitators and barriers relating to both content and delivery of interventions. Students' inclinations towards different types and characteristics of well-being support offered by the well-being services were explored using a survey developed for the current study (see supporting Information S1). Participants reported most pressing issues they desired support for, as well as the format of support which would most likely promote their engagement-including group size and composition, delivery format, frequency of sessions and proportion of skill development (vs. didactic) content.

| Focus groups
A semi-structured focus group schedule (see supporting Information S1) was developed to guide the discussions. It consisted of three broad sections: well-being and its contributing factors, previous engagement with well-being resources and preferences for future well-being resources. The questions and prompts were used flexibly, accommodating for the natural progression of conversation and allowing the moderator to follow up points deemed relevant to students' experience and the research objectives.

| Data analysis
2.6.1 | Survey -Descriptive statistics Means and standard deviations or frequency counts were described as appropriate for numerical data (i.e., demographics, well-being scales, resource preferences and topic ratings).
2.6.2 | Focus groups -Reflexive thematic analysis Braun and Clarke's (2006; reflexive thematic analysis (TA) approach was adopted for analysis of focus group data, maintaining a critical realist stance throughout. Analysis was predominantly datadriven (i.e., inductive), yet included elements of a deductive approach; Coding itself and construction of themes were inductive, whereas the goals of identifying students' well-being needs and relating the findings back to efficient well-being services were set in advance of analysis (i.e., deductively). Combining analytic approaches in this way is recognised and encouraged in reflexive TA (Braun & Clarke, 2020).
Six stages of TA (see Braun & Clarke, 2006, 2014 were followed, even though the process was iterative rather than linear (Braun & Clarke, 2020). This included transcription and initial coding by the first author using NVivo 12 (QSR International, 2020), development of draft themes and subthemes, review and consultation with the final author resulting in theme restructuring, checking new themes against data extracts and refining themes by all authors. Data interpretation was guided by an intervention development framework, the PBA (Yardley, Morrison, et al., 2015), by setting the focus of analysis on practical aspects of students' engagement with wellbeing support.

| Data integration
Survey and focus group data were first analysed separately, then analytically combined following Guetterman and colleagues' (2015) recommendations for convergent mixed-methods designs. Qualitative themes identified through reflexive TA served as basis for student priorities because the data set provided sufficient level of detail to establish them. This was then iteratively cross-referenced with the quantitative data set. Through this process, key student priorities (i.e., those sufficiently present in both data sets) were extracted and interpreted into guiding principles for student well-being intervention design. Draft guiding principles were reviewed with stakeholders familiar with provision of well-being support in the HE sector until deemed practically viable (version 3; see also page 7).
Employing multiple data collection methods provided crossvalidation of findings and allowed drawing upon the strengths of each approach; For example, the survey determined the preference that well-being provision settings limit their group size (i.e., assurance that a finding is echoed by the population). Discussion in the focus group uncovered the reasons behind it and its implications-smaller groups feel less intimidating to students, foster trust and facilitate sharing of personal experience, which is deemed important for efficient participation in well-being provision (in-depth understanding of a phenomenon). Finally, consulting stakeholders offered a provider's perspective and ensured that the resulting recommendations were scientifically and practically sound, such as emphasising the balance between available resources and impact of group size on effectiveness (readiness for implementation). A multi-method approach is essential for nuanced understanding pursued by applied research in complex systems (e.g., Johnson et al., 2017), which includes university-wide well-being provision.

| Summary of qualitative findings
The main four themes generated through reflexive TA of focus group data are presented in Figure 1. Each theme is elaborated on in the following sections. There was a recurrent notion-expressed even by students with positive personal experience-that existing well-being support for students is scarce and rarely easily accessible. This scarcity creates an atmosphere of competitiveness around accessing mental health help.
They're often, like, under resourced and they're first come, first serve or, like, for those who need it most 'cause it's not that much, they can't give it to everyone.
The idea of inaccessibility was a result of hearsay for some and direct contact with the well-being team for others. It encompassed practical barriers to participation, such as a prolonged referral process, and inadequate structural support, such as staffing levels.  (Clement et al., 2015). Indeed, the uncovered normative belief may obstruct students' sense of capability (i.e., being allowed and able to engage with well-being support), which poses a barrier to successful enactment of this behaviour (Michie et al., 2011).
Regardless, the notion is relatively novel, having not been previously identified in relation to inclusivity of services in qualitative literature.

| Theme 3: Sense of community and a safe space
The third theme details participants' desire for a supportive environment where people benefit from sharing personal experience. Nearly all participants felt that the sense of community and 'safe space' was only possible within small groups. This was underpinned by personal experience of larger groups where the sense of community was lost, and sharing felt less comfortable or even 'a little scary' (PB3).
Students have two distinct, interrelated priorities which should be balanced: a community and a safe space. Relating to others in a similar situation can reduce feelings of isolation with regards to personal struggles (Burlingame et al., 2004). Feeling safe enables the sharing of intimate experiences, which aligns to the capability component of effective behaviour change (Michie et al., 2011). These ideas chime with previous research into therapeutic benefits of groups. Participants of a group therapy for depression reported that a group setting encouraged self-disclosure and made them feel understood, benefitting well-being and progress (Schuster et al., 2018).

| Theme 4: Applying skills to real-world contexts
The final theme recounts students' quest for translating the wellbeing management skills learned in sessions into real-life contexts.
Specifically, participants wanted more explicit support to bridge this gap, since their needs were not met by current well-being resources.
A number of students distinguished between in-session skill practice and translation into daily life. In-session help was useful but did not guarantee improvements beyond it. PB4 illustrated the mismatch by acknowledging that 'if you will do this on your own terms is kind of another issue'. Interventions were more valuable when they explicitly addressed day-to-day situations influencing their wellbeing. Participants' ability to transfer the newly acquired well-being skills to everyday situations is vital for long-term sustained change in the face of limited availability of support (Gallagher, 2014).
Furthermore, quantitative evidence on effective student mental health interventions indicates that skill-oriented resources were seven times more likely to produce clinically significant improvements in anxiety, depression and emotional well-being compared to information only (Conley et al., 2013). This provides a moral and economic case for expanding the skill-oriented content of well-being resources over and above students' recommendations.

| Online survey: Resource preferences
Participants' interest and format preferences are summarised in Table 2. Students predominantly reported being interested in the described resource (see discussion for how this may have been impacted by contextual factors). In terms of specific preferences for the resource in development, survey respondents' views largely aligned with focus group participants. Students preferred a blended approach (i.e., combined in-person and virtual settings) and online workshops over those taking place in-person. This was possibly due to current COVID-related concerns. A majority of participants voted for small or medium group settings, as seen in Theme 3. The sentiment was not as clear -nor as strong -with regards to group composition. Participants wished to receive support relatively frequently, either weekly or fortnightly. Finally, respondents mirrored the desire of focus group participants for predominantly skill-based resources (Theme 4).

| Guiding principles for student well-being interventions
The survey and focus groups provided an understanding of students' priorities for engagement with well-being support at university. From this, the research team followed the PBA to highlight main behavioural issues standing in the way of students seeking support (Table 3).
These provided a basis for key design objectives, which are met with recommendations for intervention features (Yardley, Ainsworth, et al., 2015). The translation into guiding principles integrated the present projects' multi-methods results, existing literature and consultations with stakeholders familiar with well-being service provision (see also Data integration section).
For instance, the issue of 'students find [ing] well-being support at universities hard to access and insufficient for their needs' (behavioural issue 1) has been identified both in our data and in previous work (Burns et al., 2020;Thorley, 2017 information on WB support more readily available (e.g., include leaflets in fresher's week activities, send out weekly/fortnightly emails to all students)' (intervention features 1). In contrast, the authors were not aware of any previous findings of 'students feel [ing] that well-being support available through universities is aimed at (and exclusively available to) those experiencing severe distress' (behavioural issue 2). This was discussed at length by our qualitative sample, as well as welcomed by the expert stakeholder panel, so we deemed it significant enough to translate into corresponding design objectives and intervention features. This process was repeated for all behavioural issues listed in Table 3. The resulting guiding principles for the development of student well-being resources are summarised in Table 4.

| DISCUSSION
The present study informs the development of student well-being interventions. A qualitative exploration of students' well-being needs identified four core priorities for effective resources: Ease of access, inclusive and preventative approach, community and a safe space, and applying skills to real-life contexts. Along with findings of a broader quantitative survey and consultations with stakeholders, the findings were distilled into a set of recommendations for intervention development. Such concise and concrete guidelines will guide future intervention design to make student well-being interventions for acceptable and efficient. The PBA framework ensured that the process of producing the guidelines was systematic and centred around the target group's experience, enhancing its validity and the credibility of outcomes.

Key design objectives Intervention features
To improve accessibility of WB support for university students.
• Provide content through widely accessible means (e.g., online or 'blended' approach) • Make information on WB support more readily available (e.g., include leaflets in fresher's week activities, send out weekly/fortnightly emails to all students) To challenge the belief that WB support is aimed only at those who struggle.
• Ensure students are aware of preventative WB support (vs. only high intensity/clinical) • Minimise practical barriers to participation (e.g., make self-referral sufficient) To present WB as a state to be maintained rather than 'fixed' when poor.
• Advertise the importance of WB maintenance for health and academic outcomes • Stress the benefits of early and preventative engagement with WB support (i.e., the idea of building a coping 'toolkit' in advance) To provide a supportive environment conducive to sharing personal experience.
• Employ teaching tactics to build community within larger groups (e.g., breakout rooms) • Ensure the physical environment sessions are held in is comfortable and relaxing (if delivered in person) To offer WB support acknowledging and directly addressing vday-today issues.
• Explicitly apply WB skills (e.g., mindfulness, compassion) to real-life contexts in-session • Frame the intervention itself and its content as skill-based and relevant to common student concerns • Offer clear guidance and support about integrating skills into daily life during sessions.
T A B L E 3 Behavioural issues students face when engaging with university well-being support Key behavioural issues 1. Students find well-being support at universities hard to access and insufficient for their needs.
2. Students feel that well-being support available through universities is aimed at (and exclusively available to) those experiencing severe distress.
3. Students do not acknowledge that mental health and well-being require (or can benefit from) maintenance, instead only seeking support when the situation is serious.
4. Students feel unable to meaningfully engage with well-being support that feels impersonal and does not offer a supportive (figurative and physical) environment.
5. Students do not engage with well-being support that cannot be easily applied to real-life issues.
REMSKAR ET AL. -785 The context in which research is carried out inevitably shapes its conclusions (Phillippi & Lauderdale, 2018). Present research was conducted early on during the COVID-19 pandemic, Most participants indicated that their well-being was negatively impacted by the pandemic, which is in line with existing research on this (Evans et al., 2021). As a result, opinions on well-being support were altered -for example, over half of all survey respondents reported increased interest in well-being support. There was also greater emphasis on physical safety (i.e., minimising the risk of transmission), which was not a concern in previous research.

| Strengths, limitations and future directions
Present research is methodologically robust -qualitative exploration of students' experience provided an in-depth insight into engagement with well-being help at HEIs. Integrating this with a broader survey assessment, despite its relatively small sample size, enabled a more wholesome evaluation of student well-being needs than a purely qualitative study, since qualitative research does not (intend to) provide generalisability of findings (Silverman, 2013). Thus, the study added to the limited literature-exploring motives behind students' engagement, rather than quantitatively evaluating intervention effects with little consideration of why those effects were (or were not) present.
The adoption of the PBA, a renowed intervention development framework (Yardley, Ainsworth, et al., 2015), enhanced the work's methodological rigour. It provided a blueprint for exploring the target group's well-being needs, ensuring that the process was systematic and detailed. It also mandated the translation of findings into readyto-use guidelines, increasing the project's impact potential. Nevertheless, the PBA is designed as an iterative process, whereas the present study only consulted the target group once. Future research can build on this foundation by producing a prototype intervention based on the guiding principles and continuing the approach.
In addition, the PBA requires a representative sample of the whole target user population for input, which this study cannot claim to have achieved. Our sample only included students without current psychological diagnoses. This constricts the generalisability of findings (i.e., our guiding principles) to low-intensity well-being support but is less applicable to acute psychological services-the need for which unfortunately also increased due to the pandemic (Evans et al., 2021).
Finally, the study consulted students as one homogenic group, which does not reflect the group's diverse reality (Higher Education Statistics Agency [HESA], 2020). It did not recruit a significant number of students who may be structurally predisposed to experiencing poorer well-being (e.g., Black, Asian, and minority ethnic [BAME] or lesbian, gay, bisexual or transgender [LGBT+] students; OfS, 2019) and whose circumstances differ from those of most students (e.g., part-time students, those with caring responsibilities).
While the findings present a valuable starting point-given the paucity of research on the subject-future studies should aim to distinguish student well-being needs and priorities for different demographic profiles and circumstances. To achieve this, studies could focus on hard-to-reach student groups and tailor their approach accordingly (Ellard-Gray et al., 2015).

| CONCLUSION
The present mixed-method investigation determined student priorities for acceptable and efficient well-being support at university.
Several of its conclusions further existing findings, whereas other ideas are novel -most notably, students' assertion of the need for an inclusive and preventative approach to well-being, which has so far only been expressed by service providers (rather than users).
The findings themselves-particularly distilled into guiding principles (see Table 4)-are direct recommendations for more efficient future student well-being support. This work amplifies the calls for a sector-wide shift towards a more systemic and preventative management of student well-being (e.g., Hughes & Spanner, 2019).
Formalised findings and the engagement of stakeholders in the process contribute towards a stronger, more legitimate case for this. By adopting the PBA, current findings were elevated above the passive speculation that research outcomes are too often confined to; Instead, they were translated into the 'language of intervention and implementation' (Sandelowski & Leeman, 2012, p. 1404), which lends itself to application and promises tangible impact.