Help-seeking attitudes and behaviours among humanitarian aid workers

Due to the nature of their work and operating environments, humanitarian aid workers experience higher rates of psychological distress, burnout and mental health conditions than other emergency service worker populations. Fourteen international humanitarian workers were interviewed to examine whether they seek help from others in the context of work-related distress, specifically, their attitudes and behaviors regarding personal help-seeking at such times, their preferred sources of support and factors that enable or constrain effective help-seeking. Thematic analysis of the data derived five superordinate themes: (1) cultural aspects of help-seeking; (2) risks with formal, internal support; (3) lack of shared understanding of humanitarian context; (4) self-censoring and withdrawal; and (5) role maturity. There is high, in principle, support for personal help-seeking but its use is highly selective. Work colleagues are regarded as the most trusted and effective source of help in high stress periods, while barriers that exist with family and friends mean they are rarely sought out at such times. Trust and confidentiality concerns limit the use of internal agency supports and psychosocial services. External psychological services are preferred but are often found to be unsatisfactory. These findings can support aid organisations to address stigma perceptions that are commonly associated with personal help-seeking, particularly among early career practitioners, and normalise its use as a form of occupational self-care.

Humanitarian aid work is a compelling vocation but one characterised by significant exposure to field-based and organisational stressors which can affect the health and wellbeing of practitioners. A review of the mental health status of humanitarian workers (Connorton et al. 2012) found they experience higher rates of anxiety, depression and trauma symptoms than the general population. Prevalence indicators of post-traumatic stress disorder (PTSD) within this population range from 8 to 43% and are higher than those observed among military and police personnel (Rose et al. 2002;Defence Health 2016;Skeoch et al. 2017). While a range of field-based stressors and critical incidents are associated with these outcomes (Nolty et al. 2018), a systematic review by Brooks et al. (2015) highlights that organisational stressors (e.g. poor leadership, poor support, exhaustion and 'daily hassles') may be equally detrimental to the wellbeing of these workers. Within the aid sector, mental health issues are associated with increased staff turnover and health costs, loss of institutional knowledge and less effective programmes (Loquercio et al. 2006;Webster and Walker 2009;Korff et al. 2015). Brooks et al. (2015) argue that to improve the psychological resilience of this workforce, humanitarian agencies need to prioritise strengthening relationships between team members and supervisors and deal effectively with preventable forms of occupational stress.
Despite calls for improved preventive mental health strategies and training for humanitarian roles as part of employer duty-of-care responsibilities (Jachens 2019), these aspects of workforce preparedness continue Page 2 of 18 Stevens et al. Journal of International Humanitarian Action (2022) 7:16 to receive limited attention within aid organisations (Antares Foundation 2012;Cockcroft-McKay and Eiroa-Orosa 2020). This reflects, in part, traditional organisational priorities towards aid beneficiaries, but also limited information regarding stress coping and resilience among humanitarian workers (hereafter 'workers'), particularly those factors amenable to change through training and organisational support programmes (Connorton et al. 2012). Worker altruism, self-sacrifice and job insecurity may also limit their readiness to acknowledge work-related stress and seek assistance when indicated (Tassell 2009;Young et al. 2018). This is significant, as there are established associations between chronic stress within aid roles and occupational 'burnout' . Moreover, burnout is a risk factor for mental health conditions including anxiety, depression and PTSD (Williams and Greenberg 2014). Stress is therefore a key factor impacting occupational mental health and its monitoring and early management, including 'whole-of-organisation' approaches, may substantially reduce such outcomes (Jachens 2019).

Social support
Social support is defined as emotional and instrumental social interactions that provide an individual with actual assistance (received support) or a belief that such assistance is available (perceived support) (Thoits 2011). Within the general population, social support represents an important protective factor against adverse mental health effects, as it appears to provide an effective buffer against stressful experiences (Musa and Hamid 2008;Ager et al. 2012;Cardozo et al. 2012;Eriksson et al. 2013;Brooks et al. 2015). Studies of aid and emergency service workers have observed similar health protective effects (Eriksson et al. 2009;Prati and Pietrantoni 2010;Ager et al. 2012;Cardozo et al, 2012, with several finding that social support moderates the effects of field trauma exposure on PTSD outcomes among humanitarian workers (Eriksson et al. 2001), including within longitudinal research (Lopes Cardozo et al. 2013). Social support positively predicts worker wellbeing, satisfaction (Brooks et al. 2015) and resilience (Eriksson et al. 2009) and is inversely associated with psychological distress and burnout (Musa and Hamid 2008;Cardozo et al. 2012;Brooks et al. 2015), and depression (Eriksson et al. 2013).
Notably, a recent study of aid and emergency workers (Guilaran et al. 2018) found that the perception of available and satisfactory social support ('perceived support') is often more important in building worker resilience than actual 'received support' , because unless received support meets a particular need its effects can be detrimental.

Sources of support
Within the social support and help-seeking literature, sources of support are generally delineated as 'informal' sources (i.e. friends, family and work colleagues) and 'formal' sources such as those within employing organisations (e.g. supervisor, human resource officer or counsellor) and external providers (e.g. psychologist, Employee Assistance Programme) (Deane and Wilson 2007). Eriksson et al. (2009) compared the relative strengths of support sources among aid workers and found perceived supportive relationships were more important to maintaining a positive assessment of one's work role than either support from their organisation or from God, even among faith-based relief workers. Kaspersen et al. (2003) found that a positive social network, as measured by as friends, family, neighbours and colleagues moderated the relationship between trauma exposure and reactions for humanitarian workers. More specifically, they found that the stronger the social network, the less likely workers were to experience adverse reactions after trauma exposure. Several studies highlight the importance of support from an individual's team and colleagues (Curling and Simmons 2010;Cardozo et al. 2012;Young et al. 2018), particularly during international deployments (Eriksson et al. 2009). Ehring et al. (2011) found that the ability of workers to informally de-brief with their colleagues was reported to be one of the most helpful ways of dealing with distressing encounters within their work roles.

Formal and organisational support
There is a growing literature examining the relationship between the organisational and managerial support of aid workers and their occupational mental health. Supportive organisational policies and practices include training, employee engagement, health evaluation and surveillance, and incentives to support health promotion and protection (Sorensen et al. 2013). At a broader level, Hall et al. (2010) examined organisational culture regarding occupational mental health, which they refer to as psychosocial safety climate (PSC). This reflects the extent to which psychological health and safety within a workplace is valued and prioritised, and how this is expressed through prevailing beliefs, attitudes and behavioural norms in the organisation. Jachens et al. (2019) has highlighted the importance of developing such work cultures, arguing there has been a prevailing focus in the sector on traumatic stressors as the main predictors of worker wellbeing. This medical model perspective, focused on pathology, places the onus on individuals to identify issues and secure treatment, rather than on the organisation to systematically prevent or reduce work stressors and job strain. A multi-faceted approach supporting Page 3 of 18 Stevens et al. Journal of International Humanitarian Action (2022) 7:16 individual strengths and resilience, couched in an organisational culture with high psychosocial safety and process review, likely offers the most beneficial combination to support worker wellbeing (Murta et al. 2007). Several studies have found perceived organisational support increases humanitarian worker resilience on a range of indices (Eriksson et al. 2009;Curling and Simmons 2010;Cardozo et al. 2012;Comoretto et al. 2015). Organisational investment in individuals, such as proactive communication and providing psychosocial information, is associated with higher worker commitment, motivation and appreciation of role (Wilson and Gielissen 2004). The use of psychological services has been shown to reduce burnout and staff turnover (Musa and Hamid 2008). Psychosocial training is also protective against burnout, and this effect increases with training exposures (Young and Pakenham 2021). Duffield (2013) highlights that aid agencies increasingly employ 'resilience training' formats originally developed for military and emergency responders. This training supports skill and asset development such as stress recognition, the need for supportive social networks and developing emotional distance (Southwick and Charney 2012). At the team level, workers who receive positive support from team leaders and colleagues report higher personal accomplishment  and are less likely to experience burnout and other negative health outcomes ). This may be particularly evident in complex and hazardous field environments where well-functioning teams permit feedback, exchange and reflection, and a 'safety net' for worker health and wellbeing (Wilson and Gielissen 2004).
Despite the known benefits of organisational support, Cockcroft-McKay and Eiroa-Orosa (2020) identified significant barriers to aid workers utilising psychosocial services within their organisations (e.g. in-house psychological services, psychosocial education, coaching). Key constraints included a 'tough guy macho culture' , but also a 'martyr culture' . These seemingly reflect an industry that privileges traditional masculine traits (e.g. independence, fearlessness) and encourages workers to sacrifice their own wellbeing to the 'greater good' of the work. The researchers argue that such cultural barriers to workplace psychosocial programmes could be readily reduced or eradicated by employer groups. In this regard, the recent findings of a large UNHCR worker survey (Suzic et al. 2016) are instructive. Half of the respondents (n= 1090, 52.4%) indicated that they did not need to consult a counsellor, while the other half (n= 989, 48%) did express this need. Among this latter group, 548 (26% of total sample) actually spoke to a counsellor. While the limited take-up of 'needed' counselling is of concern, the study also found a significantly higher use of UN internal counsellors than external mental health professionals (88 vs. 12%, respectively). This latter finding suggests that organisational mental health culture, and the trust workers ascribe to their employer regarding this need, are critical factors in the utilisation of internal psychosocial services (Macpherson and Burkle 2021).

Help-seeking
Help-seeking is an interactive process through which the benefits of social support can be realised. It focuses particularly on the willingness and capability of individuals to elicit assistance or support from available sources when needed. Cauce et al. (2002) defines this process as the multiple stages of support-seeking individuals employ when they feel unable to cope effectively on their own, particularly when stress affects their daily functioning and is unlikely to resolve without support from external sources. Help-seeking has both attitudinal components (e.g. perceived effectiveness, social acceptability) and behavioural components, such as the interpersonal skills needed to elicit help . Within general population samples, positive attitudes towards help-seeking are associated with better mental health status (Cornally and McCarthy 2011;Clements 2015), while perceptions of weakness/stigma and threats to self-image reduce help-seeking (Wilson and Deane 2011).
Within organisational research, help-seeking is associated with reduced rates of burnout, increased job satisfaction and performance, and lower staff costs (LaMontagne et al. 2007). Such findings suggest that personal help-seeking may be an important modifiable factor in aid works settings, with the potential to reduce hazardous stress levels and improve occupational mental health (Connorton et al. 2012). While studies have examined perceived and received social support among aid and emergency workers, specific research examining help-seeking among humanitarian workers is scarce (Cockcroft-McKay and Eiroa-Orosa 2020). Young et al. (2018) found positive coping strategies reported by these workers included help-seeking via formal sources (e.g. therapy and psychological support services) and informal sources, such as friends and family. Ehring et al. (2011) found that the ability to informally de-brief and consult with colleagues through 'talking/sharing' was one of the most helpful ways of dealing with distressing encounters and reduced reported levels of anxiety and depression. Among military personnel experiencing mental health issues, seeking help with colleagues reduced feelings of isolation and facilitated decisions to access formal care (Zinzow et al. 2013).
There is evidence that factors such as gender, status and perceived control within an individual's work role may affect help-seeking perceptions and behaviours.  Comoretto et al. (2015) found female aid workers are more likely to experience positive changes to personal resilience in field settings due to their greater tendency to develop social relationships and engage in personal helpseeking in times of need. However, another study (Lopes Cardozo et al. 2013) found no gender differences regarding these factors. One study with emergency workers, Vashdi et al. (2012) examined the relationship between perceived job control (PJC) and help-seeking behaviours among firefighters involved in the 9/11 terror attacks and its aftermath. While firefighters with moderate levels of PJC were able to ask for help when needed, significant reluctance was reported among workers with the lowest and the highest levels of PJC. The authors posited that both less experienced and senior workers feared the potential further loss of control associated with externalising personal support needs in this way.

The present study
Help-seeking has been shown to support occupational mental health, including within military services, but there has been little examination of its use among humanitarian workers, a similarly high-exposure cohort. While Young et al. (2018) found aid workers rated available social connections as a primary element of effective coping, more information is needed about their perceptions and use of personal help-seeking in the context of high work-related stress, and the factors that promote or hinder its use as a coping strategy (Skeoch et al. 2017). Given the limited available data, this study used an exploratory, qualitative approach to examine these issues among professional humanitarian workers. Our primary research questions were as follows: (1) What are worker attitudes and behaviours regarding personal help-seeking in the context of workrelated distress? (2) What are their preferred sources of help (including formal and informal) at such times? (3) What factors support or constrain effective personal help-seeking?

Participants and procedure
All study participants met the requirement of being (i) internationally deployed for at least 40 days within the last 2 years and (ii) deployed principally in field-based roles (i.e. including direct engagement with affected communities). Individuals who were not internationally deployed or whose work involved exclusively administrative or office-based roles were excluded. Convenience sampling was used to recruit participants. All participants were employed within a single, international humanitarian organisation at the time of interview, albeit the majority had worked for multiple international agencies and their responses drew upon career-wide experiences. After obtaining ethics approval (Approval number: H12207), an agency representative in Australia sent email invitations for study participation to 32 Australian-based and international workers who were on a regional deployment roster. From this, 24 workers indicated willingness to be contacted and provided personal details. From these, 17 met study inclusion criteria, were sent an information sheet and were subsequently interviewed during August 2017. As most either resided outside of Australia or were on overseas deployment, 16 interviews were conducted via videoconference and one was conducted face to face, with participants indicating their preferred interview mode. The participants gave verbal consent to take part in the study and agreed for the interview to be electronically recorded for later transcription and analysis. The interviews were audio recorded with the QuickTime Player application. The final sample consisted of seven females and seven males (N=14) who ranged in age from 33 to 62 years (M = 44.71, SD = 9.92). All were deployed as international (expatriate) staff and did not work in their countries of origin as national staff. Overall, they were an experienced group of practitioners. The reported duration working for humanitarian agencies ranged from 4 years to 27 years (M= 13.28, SD = 6.47). Reported number of international deployments ranged from 4 to 35 (M = 12.79, SD = 9.90), with four having completed 25 or more deployments. All participants held university level qualifications: undergraduate (n=10) and masters (n=4) degrees. Their countries of reported nationality were Australia (n=10), USA (n=2), Jordan (n=1) and UK (n=1).
Data collection was conducted to the point of data saturation, that is, where no additional issues are identified, data begin to repeat and further data collection becomes redundant (Kerr et al. 2010). Fourteen interviews had been conducted at this point of data saturation, resulting in the final sample. The two final interviews completed were not transcribed for analysis on this basis. The data of a third participant was excluded, as clarification at interview indicated that their experience was not consistent with study inclusion criteria. Participants were de-identified at the point of manual transcription with pseudonyms used in all subsequent use of the data, including in this paper.

Measures
As little research exists on help-seeking attitudes and behaviours among workers, semi-structured interviews were considered an appropriate strategy to gain a comprehensive understanding of this issue. To support the identification of preferred sources of help, we used the definitions of Deane and Wilson (2007) to delineate help-seeking using 'informal' sources (i.e. intimate partner, friend (non-relative), family member, colleague or peer-support worker) and 'formal' sources within employing organisations (e.g. supervisor, human resource officer or counsellor) and external sources (e.g. psychologist and Employee Assistance Programme).
Questions for the interview were adapted from items within validated questionnaires which assessed workrelated stress and help-seeking or were adapted to the work context. Items about help-seeking attitudes from the Attitudes Towards Seeking Professional Psychological Help Scale (ATSPPHS) (Mackenzie et al. 2004) such as "Someone with an emotional problem is unlikely to solve it alone", informed our item, "How do you feel about seeking help from others when you're feeling distressed?". Preferred sources of help assessed with the APSPPHS item, "I would rather be advised by a close friend than by a psychologist, even for an emotional problem", informed our question, "Do you prefer seeking support [for workrelated distress] from informal sources (e.g. friends, colleagues) or formal sources such as your manager or a counsellor?". The General Help-Seeking Questionnaire (GHSQ)  item, "Have you ever seen a mental health professional (e.g. counsellor, psychologist) to get help for personal problems?", informed our question, "Have you previously sought external support (e.g. counsellor, psychologist) for work-related stress?". We developed open-ended questions regarding factors that 'motivate' or 'restrain' help-seeking e.g. "What factors would motivate you to seek help from others?" Help-seeking questions in the interview related specifically to work-based stress and focused on attitudes, and actual use, during periods of high stress. To examine whether question phrasing and concepts were appropriate for the population of interest, pilot interviews were conducted with individuals familiar with humanitarian aid and development contexts, including two former humanitarian practitioners. Practice interviews were also conducted with lay people of non-English speaking backgrounds to ensure understanding within these populations. Interview questions were refined based on feedback received via these pilot interviews. These groups also provided advice regarding help-seeking definitions, phrasing of interview questions and examples used in the survey instructions. For example, the latter focused participants on help-seeking during "periods or situations that were particularly stressful". Times when workers may directly elicit the help or support of others were framed with the examples, "feeling you were struggling to cope, felt unsafe or wanted to talk/debrief about a particular incident or situation". The interview schedule is available via the research repository ResearchDirect.

Analysis
Thematic analysis was considered the most appropriate analytic method for this study, as it allows in-depth exploration of an individuals' experiences and perceptions and is well suited to research areas which have limited theoretical and empirical background (Braun and Clarke 2006). Braun and Clarke's (2012) approach to thematic analysis was utilised and involved familiarisation with the data through initial readings of transcripts and analysing and grouping similar phrases deemed relevant to the research questions into codes. The analysis was an inductive process as the themes were grounded in, and induced from, the data set (Braun and Clarke 2012). The codes were then grouped together based on their similarities and distinctions, which formed the foundation for the subthemes. An iterative process was employed whereby constant referral to transcripts and literature ensured the validity of the produced themes.
The primary data coder was a female graduate psychology researcher (author two). The work was supervised by a senior academic (author one) with research background regarding the occupational mental health of emergency workers. A validity check was conducted using a corater to provide multiple perspectives of the data (Fereday and Muir-Cochrane 2006). This female researcher (author three) had experience of qualitative methods and research regarding the humanitarian workforce. The corater independently followed the process detailed above for the entire data set. The researcher and co-rater then consulted regarding their initial findings, which resulted in further refinements and the creation of the final overarching themes. This collaborative mapping by the researcher and the co-rater resulted in five superordinate themes that best represented the data and answered the research questions.

Results
The study found that there is high, in principle, support for personal help-seeking in the context of work-related distress, but that its use is highly selective and directed to key groups and processes. Work colleagues were regarded as the most trusted sources of help and support at such times. Their help was perceived to be effective, with respondents explaining that it provided acknowledgement of stressors, supported acceptance of circumstances and dissipated elements of their stress. Reported barriers experienced with family and friends mean they are rarely sought out for help with significant stressors. A perceived lack confidentiality and trust concerns limit the use of internal agency services and supports. While Page 6 of 18 Stevens et al. Journal of International Humanitarian Action (2022) 7:16 professional support services external to the agency are generally preferred (e.g. counselling and EAP), and accessed more often, issues identified with these services often limited their perceived value and use. The thematic analysis identified five themes that were prominent within the data: (1) cultural aspects of helpseeking; (2) risks with formal and internal support; (3) lack of shared understanding of humanitarian context; (4) self-censoring and withdrawal; and (5) role maturity. Relevant interview extracts were used as supporting evidence in establishing the identified themes (see code chart, Table 1). The themes in this section are presented under categories related to the study research questions regarding help-seeking attitudes, preferred sources and enabling factors, albeit there is a degree of overlap across these categories.

Attitudes towards personal help-seeking Cultural aspects of help-seeking
Aid cowboy culture Respondents spoke about perceived organisational expectations of being "resilient" in their work roles, including high stress situations. While no specific questions were asked about personal resilience, 7/14 workers mentioned this when asked about stress management training.
Michelle (37, Australia): I sort of snapped and I lost it in Afghanistan. I realized I wasn't coping with the stress as well as I should and the stress was getting to me too much. But I think because I was in somewhat of a senior role, there was a sense of I needed to be strong, resilient and keep it together, you know, like an aid cowboy.
Beyond its common association with training, several workers also linked resilience with being like 'aid cowboys' . This suggests a phrase in common usage, conveying a cultural aspect and expectations within their work environment. This construction of resilience portrays an individual as stoic, head strong and able to manage difficult situations independently and silently; a view consistent with the 'machoism' of aid work described in recent literature.
There was a perception that the 'cowboy' identity is more common among older staff and relates to the more limited focus on mental health that previously existed within the sector.
Erika (33, Australia): It's like, if you're coming into this industry, you should be the toughest and deal with the stress on your own, but that's also some-thing that comes more from the older workers who have been doing it longer … but it's professionalising. I think there's less and less stigma because we're more aware of the impacts.
Some older staff indicated that operating according to these more traditional expectations had contributed to adverse health outcomes over time.
Kyle (50, Australia): Look I've done this job for a while now, and I've tried to deal with everything on my own, you know, tried to be that expected aid cowboy, and I've had my breakdowns.
Some respondents noted that young, recent entrants to the field also often have, 'alpha dominant personali-ties… and are too big for their own britches' (Chloe, 62, Australia). This suggests that the tough, 'resilient cowboy' persona may not simply fall on generational lines and persists despite increasing awareness of the potential health risks of this work.
Organisational culture (help-seeking seen as poor coping) While all respondents (14/14) reported positive perceptions regarding help-seeking in the context of work-related distress, their reported behaviours were often not consistent with this view. Most (8/14) indicated a preference for 'managing on their own' during stressful periods on deployment. This was largely due to concerns of raising their organisation's awareness of any workrelated distress they may be experiencing. The hesitance in seeking help, particularly from within the organisation, stemmed from the widely held belief that help-seeking indicates poor coping and that such perceptions can have a detrimental impact on one's reputation and career.

Jerry (49, United States): 'When it came time to renew your contract, my organization wouldn't because they see you as crazy or damaged goods or not being able to perform at your previous level' .
Eleven of the fourteen respondents held similar beliefs, indicating that stigma regarding mental health, linked to performance perceptions, is common among these workers. Some felt such views were undergoing 'generational' change within the sector, with newer workers more focused on the mental health and selfcare aspects of their work (4/14). Several perceived that the agencies themselves were 'professionalising' (4/14) and 'building into their structures more psychosocial support and more ability to care for their staff. ' (Selina 33, United Kingdom). At the same time, those in positions of influence such as managers and team leaders were seen as more likely to retain traditional views of The value of help-seeking may be learned through time and experience and can support personal coping and wellbeing. East/West cultural views Twelve of the fourteen workers identified cultural mores in workers region of origin, such as 'East/West' culture, as an important factor affecting help-seeking, including their own approaches while deployed overseas. Some (7/14) expressed the view that workers with North American or European cultural backgrounds were 'more likely to openly access supports' (Erika, 33, Australia). Six workers indicated that cultural mores affected the way personal help-seeking was conceived and whether other coping approaches may be preferred. For example, Brian (46, Australia) perceived that his international colleagues, such as colleagues in Africa, were often more independent in their management of work pressure. This included situations he perceived may be distressing but where these colleagues, 'would not even consider seeking help [formal or informal sources] as their last option, it's not even an option' .
Yusuf (42), who was born and resides in Jordan, described his ambivalence regarding support-seeking in work settings, particularly from sources within his organisation. This was despite significant experiences with workrelated distress, which he attributed to inadequate rest periods between deployments. He believed this had resulted in burnout and ultimately a stress-related 'breakdown' . He also indicated that because of unaddressed traumas he had experienced post-traumatic stress disorder. Despite this background, when asked about the strategies he now employs in times of stress, Yusuf stated, 'So I don't know what they are. So I can say, where I am standing now, I don't have any real mechanisms that will help me relieve my stress' . When asked whether seeking help from outside his organisation (e.g. a counsellor) may assist him in identifying such coping strategies, he did not see this as a feasible option: Yusuf ( Yusuf acknowledged that both cumulative stress and specific trauma impacts, and having limited coping resources to manage these, were having detrimental effects on his wellbeing and work functioning. Even so, he remained reluctant to utilise internal or external supports due to stigma concerns; being seen as, 'mad, crazy or insane' , along with apprehension about becoming dependent upon such support. At the same time, he reported that support and training from his organisation concerning stress management was lacking. In addition to personal cultural background, most of these international workers (8/14) reported that their help-seeking was influenced by the attitudes and behaviours of national staff in the overseas postings to which they had been deployed.

Michelle (37, Australia): If I'm deployed to a region where people really keep to themselves in a stressful time, obviously even if I'm not doing so well, I can't say much either because then I'll be seen as this person who can't do their job properly… I just sort of have to deal with it on my own…until I get home.
When asked about attitudes towards help-seeking, respondents (10/14) spoke about the potentially damaging consequences of internalising stress, particularly chronic forms, and the risks of not actively managing its effects.

Kyle (50, Australia): The bigger danger is not dealing with the day-to-day stress, just banking it or putting it on the shelf and saying, you know, that's okay I'm tough, to the point that it becomes too much … that's when I think it becomes dangerous.
While positive attitudes were conveyed towards seeking help at such times, whether this group actually sought support was influenced by the approach taken by national staff and the cultural context. This mismatch between perceptions and behaviours related primarily to apprehension about being seen as 'not coping' . It generally resulted in the suppression of help-seeking behaviours when compared to other cultural contexts or where larger contingents of international workers may be present.

Risks with formal, internal support
Distrust and lack of confidentiality When asked about the extent to which participants approach formal sources of support within their agency (i.e. supervisors and human resource staff officers), most (11/14) indicated that a perceived lack of understanding and trust in these staff allowed them to only disclose 'appropriate' workrelated issues; those more instrumental in nature, such as managing resources and task allocation, rather than work-related stress.

Leah (43, Australia): I had two bosses, one on the field and one in the office. One of the bosses, I would not have gone to for anything and the other I would have gone for professional concerns, not really per-sonal…There was no trust relationship and trust is something very important to me.
Most workers emphasised the importance of separating work-related matters and "personal" work stress. A perceived lack of confidentiality was cited as a key restraint in disclosing to supervisors and human resources staff (10/14). However, some also indicated that the nature of these relationships was often not conducive to this process, with Leah describing, "the power relation supervisors have" over workers.
In contrast, confidentiality was described as sound and "total" in relationships with professional mentors (2/14) and peer-to-peer support workers (3/14). As Brian (46) described, 'With [my peer support worker] it's totally confidential. No problem' . All reported contacts of this kind were internally sourced, as agency-supported programmes, and were seen as valued sources of help and support by participants that mentioned them.
Respondents also expressed a lack of trust and confidence in internal psychosocial support services (i.e. counsellors and psychologists) during and post deployment (8/14). During times of distress, many avoided seeking help from these services and chose instead to be self-sufficient and, 'deal with it' (Jerry, 49, United States) until post-deployment.
Leah (43, Australia): I think in some organizations there are suspicions by staff that speaking to an inhouse counselling service will be reported to the senior management … and that the confidentiality is not there.
Discrimination While lack of confidentiality within agencies was raised as a common concern regarding support-seeking for mental health issues, and thought to have possible career repercussions, only 2/14 workers had observed discrimination against those who had sought help within their organisation. Nine reported that while this was a concern, they had no direct or indirect experiences of discrimination against those who had sought help. Beyond hearsay, most were unable to explain the association they perceived between help-seeking and its supposed deleterious effects on employment.

Kyle (50, Australia): You know, you'd hear that so and so didn't get their contract renewed because he couldn't handle it. But I think this was all made up. In my whole career, never have I ever seen this happen. It's just something that was created in thin air.
Most workers (9/14) reported that the perceived consequences of help-seeking related to mental health stigma that is prevalent within their organisation and the wider sector. When asked whether they felt their organisation had taken steps to address this stigma, half (7/14) perceived that no substantial action had been taken. Thus, it remains the case that seeking emotional or psychological support is commonly perceived to portray a worker as not coping with field role requirements. This perceived association acts as a largely implicit barrier to help-seeking.

Lack of shared understanding of humanitarian context
Friends Friends who work outside the sector were regarded as the least preferred source of specific help by all 14 workers. This was primarily due to due to their lack of understanding and insight about the humanitarian work role and its stressors.

David (51, United States): My friends, again, they don't understand the context of what it's like. Some of the friends I've made in the field, however, do get it.
Unlike the more limited expectations of family, some workers seemed to begin with higher expectations of what they may be able to share with close friends outside the sector, but expressed disappointment that they could not connect in a meaningful way. This lack of understanding, and sometimes interest, appeared to create social barriers in some non-work friendships and often led to greater reliance on relationships formed within the industry. Colleagues Half of the participants (7/14) reported that a lack of genuine supportive interactions, particularly post-deployment, resulted in a strong reliance on colleagues within the sector. Importantly, perceived understanding and trust in their colleagues provided all 14 workers with meaningful relationships, enabling them to fully disclose work-related distress and confidently seek help when needed. We observed a clear preference and rationale for help-seeking with colleagues compared to any other single source.

Mathew (48, Australia): You tend to reach out to them [workmates] because there's that kind of unwritten understanding of what you do and what you see and what you experience.
Jerry (49,United States): I guess reflective conversations tend to happen with your friends and colleagues in the sector itself. Because I guess there's a greater level of empathy and understanding and you are able to digest and break apart the issue a little easier and get useful advice, also chances are, you're all going through the same thing so we can have a 'bitch session ' [venting complaints one-toone or small group] without being worried that the word will somehow go to management.
The first comment highlights an implicit understanding of the working context and its issues. This is unlike the kind of 'framing' explanation and effort often needed with family or non-work friends. The latter comment conveys that this understanding with colleagues then enables practical processes to occur within the exchange which contribute to it being experienced as helpful ('useful'); a knowledgeable and trusted source allows you to deconstruct and analyse ('break apart' , 'reflect') an issue of concern and assimilate ('digest') some of its mental and emotional burden.
External professional support A perceived lack of confidentiality, and duty-of-care issues with managers and internal psychosocial services meant that most preferred formal support services independent of the organisation (i.e. mental health workers, counsellors, psychologists) and 9/14 had previously used such services. Doing so allowed them the freedom to disclose any information without concerns of organisational consequences.

Selina (33, United Kingdom): [External services] … are more confidential, it's not necessarily feeling like you have to go to your line manager who may have a duty of care to report things, which then is put on your record so then everyone might know.
Five workers reported never having accessed formal support services for work-related stress. However, when asked for their preference, all associated support services independent of their organisation with a greater level of comfort and trust. Although external services were preferred by users and non-users, a common concern (6/9 service users) was finding counsellors and psychologists with an adequate understanding of the aid worker role.

David (51, United States): [external counselling] … might not lead to the result that I need because chances are they might not necessarily understand what we go through.
Erika (33, Australia): She was very focused on the impact of the stresses of the field on me as opposed to the things that were really stressing me out, which was work related stress…I tried a couple of times but she kept bringing it back to the field so I just found it pointless to go back to the same topic over and over again… this psychologist didn't have any experience in the humanitarian field, she did what she knew best but I never went back.
While accessing external providers allowed workers to freely disclose information without organisational consequences, their relative lack of experience regarding aid Page 12 of 18 Stevens et al. Journal of International Humanitarian Action (2022) 7:16 worker roles left two thirds of service users feeling that this process was unsatisfactory, ultimately limiting its use.

Self-censoring and withdrawal
Protect oneself and family Most respondents (12/14) reported efforts to protect their family members by concealing or limiting information about their exposure to distressing situations. Typically, this was a felt obligation to protect their family's wellbeing by not adding additional stress beyond the decision to deploy and general nature of the deployment itself.

Yusuf (42, Jordan): Usually I'd say, especially with my wife and my family, I say things that doesn't make them worry and, you know, just like my normal worries.
All 14 participants reported help-seeking with family members was problematic and often, 'more stress than it's worth' (David, 51, United States). None nominated a family member as their preferred source of help at such times. Most indicated that telling family members about work-related stressors had two unhelpful consequences. Firstly, family members generally provided minimal emotional or practical support at such times. As such, there was little incentive to engage with them in-depth about issues causing distress. Secondly, when workers did confide in family, they reported feeling burdened by the nature of these communications, which often shifted to focusing on family concerns for their safety.
Daniel (39, Australia): They're [family] dealing with a lot of day to day stress of having you there any-way… and if you actually tell them 'Hi, I almost died today' , they can't do much about it, so then they will be constantly calling you and seeing if you're okay, which is nice and I get it but can get very annoying and eventually takes some sort of toll on you. So not telling them is so much easier for me.
Ironically, participants reported having to reassure and comfort family members after such disclosures, rather than feeling the process had supported their own wellbeing. Consequently, limiting disclosure to family is seen to be mutually beneficial to both workers and their family members.
Beyond inclinations to protect family members, two workers explained that their lack of understanding also created a barrier to confiding in them, particularly when recounting specific incidents. In withholding information at such times, workers were mindful of the need to protect themselves.

Leah (43, Australia): I think one of the more difficult things for aid workers and humanitarian
workers is trying to describe to family…you know living in a brick house with a dog and a swimming pool and they have everything that they want in life, so when you're trying to explain that you're living in a shipping container and it's 50 degrees Celsius and you just watched somebody get killed or a bomb went off, they just can't relate to it. So then what's the point in reliving the horrible incident with no useful outcome?.
Trying to relate the qualitatively different nature of their experiences could be difficult and unsatisfying and risked further distress. For Leah, it risks a focused sense of disquiet and alienation from her family and their 'normal' experiences. These factors commonly lead to censoring of such information with family members. Several reported that due to such experiences, they had adopted this strategy at an early stage of their field careers.
Withdrawal from toxic environments/conversations While colleagues were regarded as the most reliable source of support, some noted contexts where discussing stressful issues with them could have undesirable effects. This was most evident in relation to small group discussions, particularly where these became ritualised (e.g. after work drinks). Several workers (4/14) reported that gatherings like this could become counter-productive and cause further stress due to the negative ('toxic') environment they could create. In some cases, participants restricted these contacts to limit such effects.
Mathew (48, Australia): A number of us were having problems with management and we would sit around and kind of support one another… But that also became quite counterproductive because you were living it during work hours and then after work hours you were kind of still talking about it. So you weren't having downtime. To be protective, sometimes I wouldn't go around to somebody's apartment or space to have a drink to unwind because you knew it was invariably going to lead to somebody talking about that shitty day they had, and sometimes I thought I need to protect myself.
While workers prefer to share their concerns and stress with colleagues and see this as constructive, in some settings, it has the potential to become a burden in itself-to simply re-engage stress experiences without resolving them or otherwise allowing 'downtime' from them.

Role maturity
Resilience was frequently mentioned by participants in relation to stress management themes. Within the coping literature, it is generally defined as the ability to adapt effectively to a range of adversities. While there were largely common understandings of what constituted resilience, and the associated stigma of 'not coping' , there was also evidence that these views change over time. One sub-theme, 'Learning to be Resilient' describes learning about the work role and demonstrating over time that you could reliably handle its pressures (i.e. 'prove yourself '). The second subtheme, 'Learning to Seek Help' relates to reappraisals of the habitual internalising of work-related stress that this often entails and that reaching out for help when needed is beneficial and a 'strength' .
Learning to be resilient (prove yourself ) Participants described how they came to understand organisational expectations of how to manage their work and cope with its rigors. Particularly for new workers, internalising emotional or psychological concerns was often seen as a key aspect of demonstrating control and competence, and a means by which an individual could establish their field credibility.
Daniel (39, Australia): You don't want to demonstrate any weaknesses emotionally and whatnot, particularly when you're new to the field … you have to feel, you have to prove yourself. And then there is that concern as a newbie you know how competitive the spaces are and you display signs of weakness psychologically or your mental health and you wonder will that have a detrimental effect on your career?
Expressing emotional concerns or disquiet did not usually lead to review of the circumstances or its health and safety issues. Instead, it risked questioning of the individual's coping abilities. Several respondents (5/14) indicated that a reluctance to seek help was generally instilled from the beginning of their careers, essentially as a foundational expectation of employees.
Being able to remain strong despite adversities and personally manage work stress was seen to offer personal and professional benefits.
Caitlin (41, Australia): When I first started out in this field, we were always taught to be resilient in the face of difficulty and that this will benefit us in the future. So, as someone who is young, doesn't know much and wants to show everyone how good I am at what I do, I just took everything in and in and in and not a word came out. Everyone always complimented me on what a hard worker I was, but no one asked how I dealt with everything I'd seen.
Learning to seek help While a respondent's presentation and image in front of colleagues and supervisors was identified as a core element of field credibility, several (4/14) indicated that over time they had learned the importance of seeking support from others in managing work-related stress. Experiencing mental health impacts from their work was a key factor in their reappraisal of help-seeking and developing greater readiness to 'reach out' to others when they needed help. This credibility appeared to act as a form of reputational 'capital' built over time. Having proven themselves professionally, these workers felt they had earned the Page 14 of 18 Stevens et al. Journal of International Humanitarian Action (2022) 7:16 right to seek support when needed and not to be judged according to the stigma often associated with this form of self-care. While they may feel less vulnerable to such judgement, or simply care less about its consequences, the comment also suggests that this culture persists and colleagues may still have, 'something to lose' .

Discussion
Humanitarian workers are a high-risk occupational group and there are growing calls to improve their mental health training and support options. Accessing formal and informal sources of help when needed can critically support occupational wellbeing, but these stress management resources are commonly underutilised. Study participants indicated a willingness to seek help with work-related stressors but detailed significant constraints on when, and in what context, it could be sought and the sources that were most effective in providing such help.

Help-seeking attitudes and behaviours
Participants uniformly reported positive attitudes towards personal help-seeking in the context of high stress periods and incidents related to their work. This related to high recognition of the potentially damaging effects of work stress, particularly chronic forms that were not managed or 'processed' in some fashion. In this regard, there was a notable mental health literacy among the study participants. This was informed by specific psychoeducation or resilience training they had undertaken and increased mental health awareness in the sector and initiatives to 'professionalise' this aspect of worker support. Participants' own experiences of mental health impacts, and those shared with colleagues, also informed these views. This contrasts with other studies (e.g. Cockcroft-McKay and Eiroa-Orosa 2020) which found aid workers lacked self-awareness (i.e. knowing when 'too stressed' and needing help). Participants reported awareness of stress that was at a personal threshold for coping and were generally mindful of management decisions they had taken. There was a notable gap between expressed positive attitudes towards help-seeking and participants actual behaviours during periods of high stress. Most 'managed on their own' at such times, despite effective sources of help being available, particularly with peers. The reasons for this were multi-faceted but mainly related to expectations in their workplace regarding self-reliance and concerns that expressing help needs would present oneself as being incapable. Similar concerns restrain help-seeking in the general population, particularly among males (Clements 2015), and those from 'machismo' cultures (Eisenberg et al. 2009), and drive self-management of mental health issues among military personnel (Hom et al. 2017).
A study of humanitarian trainees (Skeoch et al. 2017) found respondents believed help-seeking was good practice, but mainly in relation to others and not themselves. Our participants perceived that mental health stigma remains common within humanitarian organisations and could have tangible consequences, such as questioning a worker's suitability and non-renewal of contract.
The relationship between mental health stigma and help-seeking among humanitarian workers has not been researched, but a recent meta-analysis of first responders (Haugen et al. 2017) found one third of participants reported experiences of mental health stigma in their workplace, with 10% experiencing related barriers to health care. Fears regarding confidentiality and negative career impacts were the most cited concerns regarding the use of mental health services. Both concerns were frequently raised by our participants and the reasons most avoided internal counselling services and preferred external providers. Paradoxically, external providers were often seen as unsatisfactory due to their limited understanding of the aid worker role. A practical solution could involve a cost-shared psychosocial support programme across Aid NGO's where suitable providers are vetted, and confidentiality is assured since the providers are not 'inside' the agency (Cockcroft-McKay and Eiroa-Orosa 2020).
While most participants perceived mental stigma was common in the sector, few could identify specific instances where discrimination related to mental health issues had occurred. One highly experienced practitioner posited that this widely held view did not reflect the reality of how agencies managed such issues; he had 'never' seen work tenure affected by mental health issues. Although their reality may be contested, these perceptions had the practical effect of restricting access to internal mental health services, as also observed with first responders (Haugen et al. 2017). The common nature of these views indicates the relative lack of a 'safety climate' for occupational mental health and practices such as help-seeking. Hall et al. (2013) found workers in organisations where psychological health is valued and prioritised (i.e. high PSC) report greater engagement and lower levels of work stress. In contrast, Cockcroft-McKay and Eiroa-Orosa 2020) found that 'macho' and 'martyr' cultures within humanitarian organisations were the greatest barriers to workers using in-house psychosocial services. Similar cultural expectations were evident in this study; notably, the 'aid cowboy' culture participants saw as being common in the sector. While often characterised as an anachronistic attitude held by 'older' or 'alpha' workers, it remained an influential reference point and affected behaviours; 'I needed to be strong, resilient and keep it together, you know, like an aid cowboy'  (2020) found that while few workers (21%) personally held 'macho/martyr' beliefs, most (58%) said they felt the need to be tougher in the humanitarian sector than they actually felt. Our study found that it was older and more established practitioners who were more likely to openly elicit help and to challenge workplace expectations that marginalised help-seeking. This has similarities with the transition Cross (2001) describes from 'brash' to 'mature' perspectives of aid work and self-care. Aid workers' own desire to 'put on a brave face' perpetuates social and self-stigma regarding mental health and restricts help-seeking (Gritti 2015). Ben-Zeev et al. (2012) detail observations of military personnel where self-stigma initially involves awareness of stereotypes (e.g. mental health issues or help needs = 'weakness') and where any potential application to oneself triggers internalised devaluation (shame, anxiety). 'Label avoidance' is a common reaction to this; purposefully not acknowledging symptoms or help needs to avoid public and self-stigma and their perceived consequences. Such appraisals risk isolating workers during vulnerable periods and are unlikely to change without intervention at the organisational level. Ben-Zeev et al. (2012) highlight need for 'stigma-change' programmes within military organisations, citing their high rates of PTSD. Cockcroft-McKay and Eiroa-Orosa (2020) contend that regular communication about mental health, and psychosocial programmes that include skills to identify and support those needing help (e.g. Psychological First Aid) can change organisational culture, particularly where these are standardised and mandatory (opt-out rather than opt-in). Highly experienced workers could be invited to leadership roles in such programmes, helping to normalise help-seeking and re-cast it as routine occupational health practice. Jachens (2019) argues persuasively that agencies, and the sector, have the power and 'duty of care' to bring about such changes.

Sources of help and factors affecting usage
Five main support networks were examined regarding factors that encourage or deter work-related help-seeking; family, close friends outside the sector, work colleagues, managers and internal counsellors, and external mental health providers. While all five sources were seen as notionally available, most were not routinely used as they had been found to be unsatisfactory or presented specific risks. The exception to this was work colleagues. They were the preferred help source above all other options due to the trust, understanding and empathy workers experienced in these relationships. Help-seeking with peer support workers was valued for similar reasons. Their 'unwritten' understanding of the work and its rigors meant that concerns could be broached more easily and processed in ways that supported reflection and accommodation of stressors. Similar help-seeking preferences with colleagues have been observed in the U.S. military populations where normalisation of symptoms with peers (i.e., knowing they were experiencing similar struggles) reduced feelings of isolation (Murphy et al. 2014) and facilitated decisions to seek formal care (Zinzow et al. 2013). Ben-Zeev et al. (2012) suggests programmes that connect workers with peers who have experienced mental health problems may be an effective, targeted strategy to reduce help-seeking stigma.
One notable issue with peer support involved situations during deployments where routinely 'venting' work stress could, for some, become counter-productive or even 'toxic' . These more ritualised social contacts seem to differ from focused approaches to colleagues to review or 'debrief ' specific stressors. The former may relate to situational constraints such as curfews and security restrictions to compounds, but risk prolonging stress exposure and/or the avoidance of social contact that may otherwise be sustaining.
There have been mixed results in the humanitarian literature regarding the effectiveness of support from family and friends (Brooks et al. 2015). With regard to specific help for work-related concerns, our respondents generally saw these as their least preferred sources. Family members could provide support for general or 'background' stress, but the effort required to disclose significant stressors (i.e. time, vulnerability, 're-living' situations, managing resulting family worry) was usually seen as greater than the value of help received. This is consistent with what Nadler (1997) calls the 'helpseeking dilemma' , where the anticipated psychological costs of help seeking hinders its actual use. While the avoidance of help from family could risk intensifying specific concerns, participants generally saw colleagues and peers as preferred alternatives in any case. Having such options likely contributed to the finding that many resolved not to rely on family as a main help source from early stages in their career. This contrasts with reports of emergency service workers, who generally prefer seeking help from family and friends (Cardozo et al. 2005;Benkel et al. 2009). This disparate finding may reflect the different nature of humanitarian roles and exposures. These include long periods of separation during deployments where communication with home may be both irregular and stressful (Cardozo et al. 2005;Hearns and Deeny 2007). These may result in distinctly different communications cycles and outlets.
Page 16 of 18 Stevens et al. Journal of International Humanitarian Action (2022) 7:16 While the cost/benefit of help-seeking with family was generally determined early, the lack of support from close, non-work friends was often less anticipated. Friends were the least preferred of all help sources. Dissatisfaction was particularly felt post-deployment and contributed to isolation, maladjustment and the need to self-manage the transition to home life. This may be a vulnerability factor that warrants greater attention in pre-deployment training, such as preparing workers for potential issues in eliciting help from friends, or strategies to so more effectively. Brooks et al. (2015) reports most humanitarian workers desire some form of postdeployment support, but are often trying to re-adjust at a time when colleague support is less available. The authors suggest deploying organisations could extend psychoeducation programmes to family and friends to increase their understanding, particularly regarding post-deployment adjustment and reintegration.
The avoidance of formal help sources within agencies such as line managers and in-house counsellors related to concerns previously noted; mental health stigma, lack of confidentiality and risks to reputation. Related issues could be seen in management power relations and the often-limited mental health support skills of supervisors. In some instances, workplace leaders were seen to perpetuate the 'aid cowboy' attitude, signalling a lack of safety regarding help needs. At the same time, participants themselves conflated ideas of worker resilience (often via 'resilience training') with the cowboy persona. While resilience training typically addresses the use of social support networks and developing emotional distance (Southwick and Charney 2012; Duffield 2013), the 'take home' message for participants appeared to privilege the latter; toughness, cognitive hardiness and self-managing work stress. The functional benefits of self-management in aid settings are self-evident, but this common characterisation of 'resilience' risks locating the responsibility for high level stress almost exclusively with individual workers-potentially increasing their isolation at hazardous times. By extension, organisational 'duty of care' may be seen to be diminished.
While workplace stigma regarding mental health persists, self-management of distress is likely to remain the primary response; it simply offers more immediate safety through stigma-avoidance. Such issues reflect what Macpherson and Burkle 2021 call, 'an industrywide failure' to address aid worker psychological health. They argue that while comprehensive psychosocial frameworks are available, their effective implementation requires the creation of an open and non-judgemental work culture. At a basic level, workers need clear assurances about confidentiality and that seeking help from internal services will not threaten their employment. At a service level, improving mental health awareness and management skills, particularly among workplace leaders, appears critical to creating attitudinal change and a more engaged 'culture of resilience' (Macpherson and Burkle 2021). For example, Mental Health First Aid training has been shown to equip individuals with appropriate skills to support others in need, to increase the likelihood of advising others to seek help and to improve the mental health of trainees themselves (Kitchener and Jorm 2004;Jensen et al. 2016 found that initiatives to foster team cohesion within aid agencies resulted in more supportive professional relationships and collaboration within teams, and greater reported trust and understanding from field colleagues, team managers and head office staff. This study has several limitations. The findings reflect views of a group of expatriate international workers. The attitudes and behaviours of national workers regarding personal help-seeking may differ and could be explored through further research. The study sample was drawn from a single, large international humanitarian organisation. While participants were highly experienced and their views reflect 'whole-of career' perspectives, typically across multiple agencies, their feedback could relate primarily to their employing organisation at the time of interview. The inclusion in the dataset of a single faceto-face interview risked introducing bias within the interview process. However, all participants were offered face-to-face and online options and the duration of the interview (56 min) was consistent with average interview length (52 min). While observations were made regarding cross-cultural influences on help-seeking, these draw on a relatively small sample with limited cultural diversity and where respondents sometimes inferred the views of other cultural groups. Similarly, discussion of how helpseeking approaches of national staff may affect expatriate workers may fail to reflect the complex dynamics that can exist in these relationships. This can include substantial differences in resource access, remuneration and power within organisations, potentially affecting how these relationships are formed and maintained. Further research of cultural factors affecting help-seeking among international and national workers is warranted.

Conclusion
A key strength of this study is that it provides detailed understanding of attitudes and workplace culture affecting help-seeking among humanitarian workers and their preferred sources of help, areas not previously examined in this population. Positive attitudes towards personal help-seeking were common and often founded on clear mental health literacy, but there was substantial discord between these attitudes and actual help-seeking Page 17 of 18 Stevens et al. Journal of International Humanitarian Action (2022) 7:16 behaviours. Specific issues limit help-seeking with family and friends and although colleagues are trusted sources and frequently used, workers more often default to managing on their own at times of high stress. Perceived mental health stigma and job insecurity within organisations present substantial barriers to help-seeking involving formal agency resources. Resilience training is a positive development within the sector but was commonly interpreted by our participants as reinforcing self-management of distress as a primary strategy. This characterisation of resilience risks isolating workers at times when assistance from others can mitigate high stress and enhance coping. Recent research highlights that building team cohesion can be a vehicle for individual resilience, promoting trust and engagement at team, line-management, and organisational levels. Such initiatives may offer the kind of systemic changes in workplace culture that are needed to better support worker wellbeing, including enabling environments for help-seeking-places where it's 'OK to not be OK' .