Farmer suicides: a qualitative study from Australia

Abstract

Background

Farmers in Australia, in general, have poorer health outcomes, including higher rates of suicide.

Aims

To investigate risk and protective factors and attitudes towards suicide and help-seeking among farmers living and working in New South Wales and Queensland in Australia.

Methods

A qualitative study in which three farming sites were selected in each state to represent an area with a suicide rate equal to, below and above the state average. Focus groups were conducted with men and women separately.

Results

Focus groups involved 30 men and 33 women. Inductive thematic analysis showed three broad themes characterized responses: environment and society; community and relationships; and individual factors. There was considerable overlap and dynamic interaction between themes. A combination of individual factors, as well as social and environmental stressors, was described as most likely to increase risk of suicide death and reduce help-seeking. The vast majority of known farmer suicides described involved men and many of the issues discussed pertained specifically to male farmers. Participants found suicide as an act complex, intertwined with many factors, and hard to fathom. A common belief was that an individual must feel a complete lack of hope and perceive their situation vastly differently from others to contemplate suicide.

Conclusions

Future suicide prevention efforts for farmers should take a biopsycho-ecological approach. Physical, psychological and cultural isolation could be addressed with education and training programmes and public campaigns. These could also improve people’s ability to recognize possible suicidality.

Introduction

People who live and work on a farm have been recognized as having high rates of suicide in Australia [1–6]. A Queensland study [4] found a significantly higher suicide rate for employees in the agriculture, construction and transport industries, particularly for males. In 2011, 72% of farmers in Australia were male [7]. The proportion of female farmers has fallen slightly in recent decades (30% in 1981 to 28% in 2011), as the proportion of women in other occupations has increased (from 37% in 1981 to 47% in 2011) [7]. Australian men are approximately three times more likely to die by suicide than women [8], and the difference between male and female farmers in two of Australia’s largest states, Queensland (QLD) and New South Wales (NSW), is even greater [9]. Between 2000 and 2009, in QLD and NSW, 90 and 96%, respectively, of farmers who died by suicide were men [9]. This trend does not seem to be unique to Australia [10]. Factors identified as related to risk of farmer suicide include occupation-specific conditions, such as overlap of the workplace and home, isolation and working alone; individual factors such as mental illness; social conditions such as intergenerational business partnerships, government regulations and cultural attitudes; as well as environmental factors such as drought, external impositions like gas mining, economic downturns and easy access to firearms [1–6,9–15]. Monk [10], in her examination of the influence of isolation on rural stress and suicide, states that the link between stress, illness, depression and suicide is not linear, but rather a result of a combination of mixed factors and events. The bio-psycho-ecological approach can be used both to understand risk factors for suicide and to address them [16].

Bronfenbrenner’s model of development ecology [16] attempts to describe the complex relationship between the individual and society, outlining five systems operating to influence individual experience within the environment. This model acknowledges that the person is operating and developing within a context, with proximal and distal relationships, in which the individual and their environment change over time, and everything is interrelated. Relationships, both between people and between the different systems (individual, organizational/community and societal), is the focus of this model, showing dynamic interaction to describe the individual in a group and a network [17]. A systematic review and meta-analysis of psychiatric and socioeconomic factors for suicide from individual-level and population-based studies by Li et al. [18] suggested that prevention strategies that focus on more distal factors, such as lower socioeconomic outcomes, lower educational achievement and low occupational status, may have similar population-level effects as strategies which target more proximal factors such as psychiatric disorders. The aim of this study was to explore individual-level factors and attitudes related to suicide in farmers within the context of the person’s inclusive environment, acknowledging the complexity of interrelated phenomena.

Methods

We recruited focus groups via e-mail invitation through rural organizations working with farmers in each of the sites and with telephone follow-up. Focus groups were conducted with people living and/or working on farms in three diverse farming communities in both QLD and NSW. Communities varied in terms of their population, distance from and access to services and farming practices. Male and female focus groups were conducted separately. In some instances, interviews were conducted by telephone. Questions asked were consistent across all groups, however, and responses congruous so that they were easily grouped under consonant themes.

Participants were provided with refreshments and a $50 voucher to reimburse travel costs. Interviews were semi-structured and based around three main themes: risk, protective factors and help-seeking. Interviews were transcribed by a professional company. NVivo software was utilized for data interpretation, following the characteristics of qualitative research as described by Gibbs [19]. Inductive thematic analysis was undertaken, with coding and theme development directed by the content of the data [20,21]. Initial coding of themes was conducted by the first author (M.P.) and further agreement of final themes and subthemes was achieved in discussions with other authors (K.K. and P.R.). This study was approved by both the University of Newcastle (H-2013-0009) and Griffith University Human Research Ethics Committees (OTH/04/12/HREC). Data are securely stored for 7 years as per ethics requirements.

Results

In NSW, there were three male groups and three female groups in each site, with a total of six participants in each group. In QLD, recruitment proved more difficult. One facilitator ran all NSW groups; however, in QLD, there were two facilitators and a mix of how groups ran. Few groups reached the target of six participants, and in some instances, the interviews took place by telephone. In total, there were 12 males and 15 females interviewed in QLD. Telephone interviews done individually had a different feel to the group interactions, and groups with the target number of six participants differed from the interaction of groups with fewer participants. Overall, 30 men and 33 women were involved from both states.

For individual-level risk factors, our analysis highlighted the subthemes: combination of factors; identity; isolation, loneliness and withdrawal; and lifestyle factors, which are described below. Attitudes towards suicide were also discussed, and grouped participant quotes reflecting these subthemes are shown in Table 1.

The subtheme relating to a combination of factors revealed that environmental, social, interpersonal and individual factors overlapped considerably to create a complex interrelated suite of constituents for risk. Explicit in most discussions was that it is not one risk factor that leads someone to take their life, but many. A common sentiment expressed was that suicide was the result of a building up of internal and external stresses. Identity emerged as a subtheme. Differing personal traits, including thresholds for stress, and the individual’s response to life’s challenges, were described as making individuals either more or less resilient, with variant (either positive or negative) coping mechanisms. Individual traits were deemed to be important, but equally if not more so, was the individual’s perception of themselves and their life. An individual’s perception of themselves, their circumstances and how they perceived others to be judging them arose in a number of ways across the groups as relevant to the suicidal mind.

Concepts of masculinity became apparent as the male identity, the Australian male identity and the Australian male farming identity were discussed. In particular, challenging circumstances seemed to affect the male farmer’s sense of identity, which was expressed as being strongly enmeshed in their industry. If the farm failed, they failed. They could not see themselves in any other context. Descriptions of rural people, and particularly farming men, as ‘private people’ commonly arose and also denoted notions of masculinity. Being private seemed part of the rural identity, especially for the older and male farmers. In addition, they were described as ‘proud’, often too proud to ask for help, feeling that if they did so it would threaten their identity and make them appear weak or feel like a failure. Across all groups, the fact that men find it difficult to communicate, do not communicate a lot or do not have the opportunity or avenues to facilitate talking was raised and identified as a risk for suicide and seemed to be a part of the Australian male farming identity.

Linked with one’s sense of identity, stories of known suicides sometimes related to people participants knew who had either a physical illness or injury. This affected their sense of identity because they were then not able to work, their circumstances changed or their circumstances were perceived by the person to have changed significantly. All groups spoke about financial pressure as a possible risk factor for suicide in farmers, and this again related to the person’s sense of identity, and how others may perceive them. Loneliness, isolation and withdrawal featured strongly as a subtheme accentuating risk of suicide. Participants spoke about different types of isolation, including the inherent isolation of farming as an occupation; geographically and by the fact that farmers spend many hours working alone; and emotional isolation and loneliness. Participants understood that stress or mental illness can lead to further withdrawal and social isolation. A number of groups spoke about physical and mental isolation in combination as being a risk factor for suicide. Compounding environmental, social and individual circumstances were described as strongly influencing lifestyle factors such as how much time farmers had, hours they worked, their sleep, alcohol and/or drug use and whether or not they exercised. Participants reflected that this may in turn relate to suicide risk or exacerbate physical and/or mental stress.

Attitudes towards suicide reflected incredulousness. Despite freely offering suggestions, answers and opinions in relation to why people may be at risk of, or die by suicide, the words ‘I don’t know’ were used repeatedly. Participants could understand and describe, and many had even experienced themselves, the pressures of farming and even severe mental illness; however, they struggled to understand how a person gets to the point of being able to take their own life. They considered that people who die by suicide must get to a point where they see no other option and have lost all hope. A strikingly common aspect of known suicides referred to was that ‘no one saw it coming’. Known risk factors, for example isolation and relationship breakup, were stated but participants did not link this to the possibility that the person could be suicidal before the event, or even necessarily, after. Participants spoke frequently of ‘having no idea’ that the person was feeling suicidal prior to their death. Consistently across all groups, participants stated that suicide frequently happened in people you may never suspect of being at risk, because they perceived the person as successful, they had masked their true feelings extremely well or no one had recognized any warning signs prior to their death. Hindsight was considered. In some instances, people could put together reasons after the fact and see what they could not see before the person died. Far more often though, participants described being left still searching for answers, with deep regret that different conversations had not taken place.

Discussion

Individual factors intertwined with social and environmental stressors were described as most likely to increase the risk of suicide and reduce help-seeking in Australian farmers. A person’s sense of identity, isolation, loneliness and lifestyle factors emerged as important elements of individual risk. The inability to recognize people who had died by suicide as suicidal prior to their death, and suicide as an unfathomable act, emerged. This study supports findings of others in terms of individual-level risk factors for suicide; however, it places these in the context of the Australian male farmer. Being a qualitative study, it adds to our qualitative knowledge. Its weaknesses include the fact that data collection was staggered, and that due to recruitment issues in QLD some individual interviews were conducted. The findings of interviews were not noticeably different from those of focus groups as questions were consistent and so were the themes identified. In addition, sites were diverse in terms of health care facilities. Joiner [22], commenting on research showing that males are particularly over-represented in suicide death rates, describes men as having more trouble seeking help due to greater perceptions of stigma related to mental health. Others have considered stigma as a possible explanation for the higher suicide rates among men within the Australian rural context [11–15]. Alston and Kent [15], like Monk [10], describe a combination of risk factors for farmer suicide including low socioeconomic status, isolation, access to firearms and climatic pressures such as drought, and in particular speaks of ‘a dominant form of masculine hegemony that lauds stoicism in the face of adversity’ [15]. Our participants described risk relating specifically to the Australian male farming identity, including their reluctance to seek help. ‘Pride’ was offered as a major reason for this. The Australian male farming identity was described as highly enmeshed in their industry, and financial pressures, relationship breakdown and physical injury or illness could all be interpreted by the farming male as personal failures. Participants described a ‘culture’ of not seeking help for fear of being seen as ‘weak’ or it only being for ‘sissies’. Sites were diverse in terms of health care facilities, the more remote communities having few primary health care or mental health services available, and access was described as difficult due to great distances involved. The larger communities had better availability and ease of access to services. Despite this, asking for help seemed to be the major barrier to receiving it consistently across sites.

Loneliness can also be a major reason for the disproportionate rate of male suicide [22]. Hawkley and Cacioppo [23] write of loneliness as distress that results from discrepancies between ideal and perceived social relationships. Joiner [22] also distinguishes the perception of loneliness as an internal emotional experience and objective external social connections. Participants in our study spoke about both forms of isolation and subsequent loneliness. Chabon [24] found perceived social support can be as, if not more, important than the actual support available in predicting mental and physical health indicators. Evident in our study was that an individual’s perception may vary vastly from those around them. While people around them may think the person ‘has it all’, including strong social supports, the at-risk farmer may feel extremely alone, and indeed suicidal.

Implicit in participant responses was the continuing search for answers despite their significant accumulated knowledge. Multiple risk factors were discussed and to a trained ear, possible warning signs described. Despite this, participants spoke frequently of ‘having no idea’ that people were suicidal prior to their death. This is significant, and something we should address in future prevention efforts. Education and training programmes may equip farmers in the community, and people who have a lot of contact with them, with more knowledge about what potential suicide warning signs may be, and how to address them. Within this may also lie an opportunity for the cultural shift necessary to enable male farmers not to feel weak if they communicate and ask for help.

In developing prevention initiatives, individual risk factors should be acknowledged and acted upon, but so too the fact that individuals are affected by their immediate surroundings, relationships, community, society and environment. Bronfenbrenner’s [16] model of development ecology may be relevant, in that the individual is dynamically interacting with different systems including community and societal doctrines [17]. Ecological counselling [25], integrating personal and environmental factors and focusing on their interaction could be particularly useful with this population. Ecological counselling addresses issues and their interdependency at the personal and environmental level and incorporates Bronfenbrenner’s ecological systems theory.

In conclusion, future suicide prevention efforts for this population should take a biopsycho-ecological approach [26], helping people to see themselves, and others, within a framework of interacting relationships and systems. Physical, psychological and cultural isolation could also be addressed within such programmes. Cultural specificity within the male farming context is vital.

Funding

The current project was funded by the Australian Research Council (project no. LP120100021).

Conflicts of interest

None declared.

Acknowledgements

The authors would like to acknowledge our partners: the Queensland Mental Health Commission; the Queensland Department of Justice, Office of the State Coroner (Queensland); the Australasian Centre for Rural and Remote Mental Health; the Hunter New England Local Health Network; and the Office of the State Coroners Court of New South Wales. We would also like to acknowledge Professor Brian Kelly, School of Medicine and Public Health, University of Newcastle as a Chief Investigator of the study. We would like to thank Lisa Kunde and Dr Leanne Craze for conducting focus groups in Queensland.

References

Author notes