Order 2689148: men and mental health in rural Australian farmers

lable at ScienceDirect

Journal of Rural Studies 46 (2016) 102e110

Contents lists avai

Journal of Rural Studies

journal homepage: www.elsevier.com/locate/jrurstud

“Watching the bank balance build up then blow away and the rain clouds do the same”: A thematic analysis of South Australian farmers’ sources of stress during drought

Kate M. Fennell a, b, c, *, Claire E. Jarrett a, Lisa J. Kettler a, James Dollman c, d, Deborah A. Turnbull a

a School of Psychology, The University of Adelaide, South Australia 5005, Australia b Flinders Centre for Innovation in Cancer, School of Medicine, Flinders University, GPO Box 2100, Adelaide, South Australia 5001, Australia c Sansom Institute for Health Research, Division of Health Sciences, The University of South Australia, City East Campus, Adelaide, South Australia 5001, Australia d Alliance for Research in Exercise, Nutrition and Activity (ARENA), School of Health Sciences, University of South Australia, Adelaide, Australia

a r t i c l e i n f o

Article history: Received 28 November 2015 Received in revised form 1 May 2016 Accepted 15 May 2016 Available online 17 June 2016

Keywords: Agriculture Drought Farm Mental health Rural Stress

* Corresponding author. Sansom Institute for Health Sciences, The University of South Australia, City E Adelaide, South Australia 5001, Australia.

E-mail addresses: [email protected] (K student.adelaide.edu.au (C.E. Jarrett), lisa.kettler@a [email protected] (J. Dollman), debo (D.A. Turnbull).

http://dx.doi.org/10.1016/j.jrurstud.2016.05.005 0743-0167/© 2016 Elsevier Ltd. All rights reserved.

a b s t r a c t

Background/aims/method: To investigate causes of farmers’ stress during drought, 309 drought-affected South Australian farmers and their spouses, mainly from sheep, cattle and/or gain properties, ranging in age from 23 to 85 years (34.6% female) completed a questionnaire. Demographic and work-related details were collected and participants were asked to list their most stressful experiences in the past month and past 12 months. Results: Most (73.1%) of the sources of stress reported were farm-related. A thematic analysis identified drought, financial pressure and uncertainty about the future as dominant stressors. A range of other more specific drought-related (e.g. poor crop yields, unsatisfactory conditions for livestock, overwhelming work- loads) and contextual stressors (e.g. rising input costs, family involvement in the farming business, pressure to take part in community work, lack of understanding of farming from ‘outsiders’) were also identified. Dominant stressors were compounded by non-drought-related issues (e.g. machinery breakdowns) and non-farm related stressors (e.g. illness and death of loved ones). Conclusion: Improved awareness of stressors may help to reduce farmers’ frustration with outsiders’ lack of understanding of their way of life, normalise stressors and thereby reduce stigma and empower farmers to seek help. These findings could also inform the development of targeted mental-health prevention and promotion initiatives for farmers in future periods of drought.

© 2016 Elsevier Ltd. All rights reserved.

1. Introduction

Farming is both a physically (Franklin et al., 2001; Kennedy et al., 2014) and psychologically hazardous occupation (Gregoire, 2002; Edwards et al., 2009; Schirmer et al., 2015), resulting in high levels of stress (Booth and Lloyd, 2000; Walker and Walker, 1987;

Research, Division of Health ast Campus, GPO Box 2471,

.M. Fennell), claire.jarrett@ delaide.edu.au (L.J. Kettler), [email protected]

Schirmer et al., 2015; Brumby et al., 2013), depression and anxi- ety among farmers (Eisner, Neal & Scaife, 1998 cited in Fraser et al., 2005). Previous studies have also found that farmers are more vulnerable to suicide than the general population (Andersen et al., 2010; Berry et al., 2010; Page and Fragar, 2002; Kennedy et al., 2014).

There are several reasons farming is distinct from many other occupations and worthy of specific research. A unique characteristic of this sector is that the farm is generally both a farmer’s workplace and home (Fraser et al., 2005; Gray and Lawrence, 1996; Gregoire, 2002; Lunner Kolstrup et al., 2013; Sartore et al., 2008). This is thought to have implications for farmers’ mental-health as many view farming not only as an occupation but also as a lifestyle, resulting in work, home and family roles being intertwined (Fraser

1 This also increases the risk of bushfires, which add another layer of complexity to stress associated with drought (Whittaker et al., 2012).

2 At the time of research, SAFF was South Australia’s leading lobbying organi- sation representing South Australian farmers. However it has recently been renamed Primary Producers SA, which (like SAFF) lobbies on overarching issues facing South Australian famers (e.g. drought relief). Although the percentage of South Australian farmers that belonged to SAFF was known to SAFF staff, they did not wish to declare this information. However, they were able to provide assurance that their members were representative of cereal and livestock farmers in South Australia at the time this research was conducted.

K.M. Fennell et al. / Journal of Rural Studies 46 (2016) 102e110 103

et al., 2005; Gregoire, 2002). In addition, farm work is physically strenuous (Fraser et al., 2005). It often requires working long hours in a variety of weather and light conditions (Gerrard, 1998), the use of dangerous machinery (Roufeil and Lipzker, 2007) and frequent exposure to chemicals (Gerrard, 1998), often in the absence of adequate safety measures (Carruth and Logan, 2002). Farm work is also complex, involving substantial financial investments and risks (Walker and Walker, 1987), large amounts of paperwork (Gregoire, 2002) and a number of forces beyond farmers’ control such as changes in weather, commodity markets and government regula- tions (Fraser et al., 2005; Roufeil and Lipzker, 2007). Further, as most farms are family-owned and operated businesses (Australian Bureau of Statistics, 2006; Carruth and Logan, 2002; Barclay et al., 2011) and often involve family members from across different generations (Rosenblatt and Anderson, 1981 cited in Walker and Walker, 1987), business and personal issues frequently fuse (Ben- nett,1982 and Kohl,1976 cited in Weigel et al.,1987). Of the 129 934 individual farming businesses in Australia, 99% are family-owned and operated (Heffernan et al., 2008). Therefore, intergenera- tional family conflict over roles and problems with the transition of farm ownership are thought to be common causes of stress among farmers (Wheeler et al., 2012; Barclay et al., 2011).

In South Australia where the present research was conducted, during the 2012e2013 financial year approximately 13 000 farms were in operation, producing goods worth $5.2 billion to the local economy (Australian Bureau of Statistics, 2014). At the time of data collection (2008),14 900 farms were in operation in South Australia and 33 000 South Australians were employed in direct farm jobs (National Farmers Federation, 2008).

The reasons farming is different from other occupations (out- lined previously) are well documented in the literature and these issues clearly affect a significant number of people. However, despite acknowledgment that causes of farm-related stress are dynamic and appear to be different during drought (Berry et al., 2011; Berry et al., 2008), less attention has been paid to the spe- cific causes of farmers’ stress during these difficult periods. Drought is thought to be associated with mental health problems as well as predisposing adversities such as financial pressure, which further contribute to poor mental health (Berry et al., 2008; Sartore et al., 2008; Stanke et al., 2013). A link between drought in New South Wales (NSW) and suicide of farmers has also been identified (Berry et al., 2008). While it is evident that drought has enduring negative consequences on farmers’ mental and social health, there is currently a lack of research detailing exactly why this is so (Alston, 2004; Caldwell and Boyd, 2009; Dijk et al., 2013). Previous research that has identified farmers’ stressors has predominantly been un- dertaken outside Australia and generally not in times of drought. Page and Fragar (2002) note the differences between Australian, North American and European agriculture (for example in pro- duction systems and economic subsidization), and warn that care should be taken when applying trends found elsewhere to Australian farmers. The obvious exceptions to the lack of local, recent research are our own report on coping strategies farmers use during drought (Gunn et al., 2012) collected from the same participant pool as that reported on in this paper, a qualitative study on what makes some farmers resilient during drought (Greenhill et al., 2009), a study on South Australian citrus growers (Staniford et al., 2009) and the study on the effects of the drought on two rural communities in NSW (Sartore et al., 2007). However, the present study focuses more specifically on the effects of drought on farmers who are mainly involved in grain, sheep and cattle production and sources of stress rather than coping strategies. It is necessary to address this gap in the literature to better un- derstand and mitigate the negative effects of stress on farmers’ health during future droughts. As van Dijk et al. (2013) explain, “Key

causative pathways from physical drought to the degradation of ecological, economic, and social health remain poorly understood and quantified … this means future droughts may well break re- cords in ever new ways and not necessarily be managed better than past ones” (p. 1040).

Therefore, the purpose of the present study is to identify the factors that specifically cause farmers stress in times of drought. For the purposes of this study, a farmer is considered anyone who owns and takes an active role in the operation of a farming or pastoral enterprise in South Australia, or someone who is the partner of a person fitting this description. This broad definition has been adopted as stress associated with farm life affects not only farmers, but also their families (Carruth and Logan, 2002; Fraser et al., 2005). A recent systematic review of the health effects of drought recently concluded that there is a need for more systematic reporting of the impacts of drought (Stanke et al., 2013); a gap this paper helps to fill.

It is anticipated that these findings will assist in the develop- ment of practical interventions to support this unique population when the need arises, as well as interventions aimed at ‘drought proofing’ farmers in readiness for future droughts. Identifying specific sources of stress may also inform evidence-based ap- proaches to help improve mental health (Polain et al., 2011; Sartore et al., 2008) and prevent suicide in this at-risk group (Lunner Kolstrup et al., 2013). Increases in extreme weather events, spe- cifically frequent, severe droughts, are expected in the future1

(Berry et al., 2010; Berry et al., 2008; Costello et al., 2009; Pachauri et al., 2014), underscoring the urgency of these research outcomes.

2. Method

2.1. Procedure

Data were collected via a self-report questionnaire that was available in printed, faxed and online form. This data collection method was selected based on feedback from farmers who met the selection criteria, that workloads were exceptionally heavy, they had to prioritise farm work and asking them to take time away from work, family and community commitments to participate in in- terviews would add to their levels of stress. However, providing them with a questionnaire to complete at their leisure was considered more acceptable, as well as more feasible given our very limited budget.

Questionnaires were sent to all members of the South Australian Farmers Federation (SAFF)2 by post, fax or email depending on members’ preferred mode of communication. Four weeks prior to this, a forewarning notice was placed in the monthly SAFF Member Update newsletter in an effort to improve response rates (Cole et al., 2007). After ethics approval was granted from The University of Adelaide’s School of Psychology Human Research Ethics Commit- tee, a covering letter from SAFF, information sheet and survey were sent to the households of potential participants. Members who received the questionnaire and accompanying materials via fax or

5 While comparable studies that have used email and fax to communicate with farmers are scarce, the response rate from the posted version of the questionnaire was similar to that obtained via post from farmers in the United States (19%) by Olson and Schellenberg (1986) and response rates are generally lower when par- ticipants are contacted via email rather than post (Kongsved et al., 2007). Booth and

K.M. Fennell et al. / Journal of Rural Studies 46 (2016) 102e110104

post were asked to send back their survey via post, reply-paid to the School of Psychology. The members whose preferred mode of communication was email, were sent a generic email from SAFF including a link to the website where participants could access further information and (if eligible, i.e. the participant or their spouse owned or played an active role in the operation of a farming or pastoral enterprise in South Australia and were English- speaking), complete the questionnaire online. Reminders were placed in two subsequent SAFF newsletters and two generic reminder emails (one including a file of the survey that could be printed and completed) were sent to original email recipients.3

Reminders could not be posted due to budget constraints. All cor- respondence to SAFF members was addressed and sent by SAFF staff for privacy reasons.

The online version of the questionnaire was also advertised in the written rural media, on radio and through personal networks to maximise the total number of respondents. Data were collected between April 28 and July 17, 2008. During this period of time, much of rural Australia was experiencing severe drought (Berry et al., 2008). The qualitative data reported in this paper were collected at the same time as quantitative data published in 2012 (Gunn et al., 2012).

2.2. Content of questionnaire

A pilot questionnaire was administered to two males and one female of different ages and regions who fitted the eligibility criteria. Similar concerns were raised by all three pilot participants so additional instructions were added to the questionnaire (mainly on how to return the survey as this appeared to cause some confusion). The questionnaire was formatted using plain text, bold instructions and obvious boxes for responses to be written in, to maximise ease of comprehension and ease of response, while an- onymity was emphasised to reduce social desirability response bias.

2.3. Measures

Demographic information (8 items): Participants were asked about personal (i.e. gender, marital status, age, region, postcode) and work-related (i.e. type of farming enterprise they are involved in) characteristics, SAFF membership and how they obtained the survey. Questions were answered by either selecting one of a number of options (including other with the option to add more detail) or in the case of age and postcode, by completing an open- ended response box.

Stressors (2 items): Participants were given a definition of ‘stressful’ from the Ways of Coping Questionnaire (Folkman and Lazarus, 1988) and asked to list in an open-ended response box, the most stressful situation they had experienced in the past month. They were told the situation may relate to their family, job, friends, or something else important to them and were provided with one farm-related and one non-farm-related example.4 Partici- pants were then asked to list the two most stressful events they had experienced in the past 12 months.

2.4. Participants

The cross-sectional sample consisted of 309 South Australian

3 Reminder emails were sent after it became apparent that the response-rate from emails was lower than from other methods of communication.

4 The farm-related example was ‘Expensive ram dying’ and the non-farm-related example was ‘Mother sick in hospital’.

farmers and their spouses (collectively referred to as farmers here in), recruited through SAFF, the media and personal networks. Participants ranged in age from 23 to 85 years (M ¼ 51.81 years, SD ¼ 11.69). The sample was 34.6% female and 63.4% male. The majority (89.6%) of participants were married or living with a partner, 3.2% were separated or divorced, 3.2% widowed and 2.6% never married, and 81.2% were members of SAFF.

Participants predominantly described their farming enterprises as a grain, sheep and/or cattle farm or a sheep and/or cattle property (see Fig. 1) and were from the Mid North, Eyre Peninsula or the South East regions of South Australia (see Fig. 2).

To test for differences in the sample between recruitment methods, a Mann Whitney U test was employed. This revealed that the only significant difference between online participants who were and were not members of SAFF on key descriptive variables (including age, distress and coping strategies reported elsewhere (Gunn et al., 2012)) was on age 1223.00 (z ¼ �3.03) (p < .05). Those not recruited through SAFF were younger probably because the first author’s personal contacts are mainly young farmers.

2.5. Response rate

The response rate for SAFF members varied depending on the mode of communication by which farmers were contacted. Of those who were posted a printed version of the questionnaire, 14.7% returned the questionnaire completed. 10.7% of SAFF members who were faxed the questionnaire returned them. Only 2.9% of members who were sent generic emails (on three separate occasions) and asked to ‘click’ on a link and complete the online version of the questionnaire or print and complete a paper copy, did so. Possible explanations for this response rate include: in- vitations to participate and reminders not being personally addressed, the questionnaire being quite demanding (30e60 min and 145 questions long-many of which have been reported upon elsewhere (Gunn et al., 2012)) and on a sensitive topic, ques- tionnaires not printed in coloured ink, printed second copies of the questionnaire were not sent to non-respondents, no incentive to participate was offered (Edwards et al., 2002) and invitations were sent at a particularly stressful and busy time (i.e. drought). The particularly low response rate from emails may at least in part, be attributed to widespread problems with internet access in rural South Australia at the time data were collected.5 For obvious reasons the response rate of participants who heard about the study via the media or through a friend or acquaintance could not be calculated.6

To examine potential bias from the relatively low response rate, we compared the characteristics of our sample with representative population-based data collected by the Australian Bureau of Sta- tistics (ABS) as well as data provided by SAFF about their members. Notably, the proportion of surveys returned from each region was generally representative of the number of SAFF members from each region and the ABS' figures on the total number employed in agri- culture in each region (Crabb, 2008). The only overrepresentation

Lloyd (2000) received a 30% response rate in their study of members of the National Farmers Union in the United Kingdom; however, they sent a specific follow-up reminder, which has been shown to markedly improve response rates (Kongsved et al., 2007; Mavis and Brocato, 1998).

6 However, they comprised approximately 14% and 5.5% of the total number of surveys returned, respectively.

Fig. 1. The percentage of participants from each farm type.

K.M. Fennell et al. / Journal of Rural Studies 46 (2016) 102e110 105

of a region in the sample was of farmers from Eyre Peninsula, which is likely to be the result of widespread media coverage and the first author’s contacts in that area. A smaller proportion of people who participated in this survey were separated/divorced (3.2%) than in the broader Australian farming community (8%) (Australian Bureau of Statistics, 2012). However, participants in our study were found to be representative of the Australian farming population in terms of age (this study’s median ¼ 51, farming population median ¼ 53) and reasonably consistent (given the varying definitions of ‘farmer’) with the Australian farming population in terms of gender (our study included ‘spouses of farmers’ and 63% were male, while ABS data does not include those who identify as ‘farmers’ spouses’, so they have an understandably higher rate of males; 72% male) (Australian Bureau of Statistics, 2012). In 2012, the gross value of South Australian farm production at the farm gate was $839 million for wheat, $480 million for vegetables and $448 million for sheep and lambs (Schirmer et al., 2015), suggesting under representation of horticulture in our sample compared to outputs from South Australian farms. This is simply because very few horticulturalists were members of SAFF.

7 For example a busy harvest might have been experienced in the last year, but not in the previous month.

8 These non-farm-related themes of stress are therefore included in the model derived from the analysis as isolated stressors with the ability to cause stress in their own right, or compound the effects of other problems.

2.6. Analysis

A thematic analysis was conducted to identify dominant types or themes of stressors from participants’ reports of the stressful situations they had encountered both during the past month and past year. A theme was considered “… something important about the data in relation to the research question, and represent[ed] some level of patterned response or meaning within the data set” (Braun and Clarke, 2006, p. 82).

An essentialist/realist methodology was adopted because we wanted to report on the experiences, meanings and reality of par- ticipants, with the view that these descriptions provided a direct insight into their experiences of stress, rather than examining ways in which their events, realities, meanings and/or experiences of stress were socially-constructed through language (Braun and Clarke, 2006). Further, an inductive (or bottom-up) approach to the analyses meant the identified themes were closely linked with the data themselves, and thematic meaning was often situated at the semantic or surface level of the data (Braun and Clarke, 2006).

Using Braun and Clarke’s (2006) method, the author became familiar with the data and then assigned grouping codes to each datum. Following consultation with the relevant literature available

at the time (e.g. Carruth and Logan, 2002; Deary et al., 1997; Fraser et al., 2005; Gray and Lawrence, 1996; Sartore et al., 2007), broader themes were identified from the codes, and the definitions and relationships between the resulting themes were considered. With the interrelationships in mind, a hierarchy of themes emerged. Following further review of codes and definitions, thematic network diagrams or webs were drawn (based on Attride-Stirling’s work (2001)) to assist in confirming and reporting themes, in- terrelationships and hierarchies.

3. Results

Descriptions of stressors in the past month were coded into three categories: farm-related, non-farm-related and combination (farm and other) (see Table 1). These data were coded on two separate occasions by the first author and by an acquaintance with an interest in the study and an understanding of farming issues. Following discussion, the coding of the 24 data (approximately 8%) that were initially classed inconsistently, were agreed upon.

Of the 297 stressors experienced in the past month that were analysed, more than twice as many were considered farm-related, than non-farm-related or combination (farm and other) (Table 1).

Using thematic analysis, one key stressor, two salient sub- stressors, a number of specific drought-related stressors, contex- tual stressors and non drought-related farm stressors, were then identified from approximately 900 descriptions of situations that caused participants stress (both in the past month and the past year). There was generally little difference in the types of stress reported in the two time frames, with the exception of a few seasonally-based factors.7

The subsequent analysis focuses on the dominant causes of stress for participants in this study, which as previously outlined, were overwhelmingly farm-related. However, illness and death were common non-farm-related themes of stress.8 Stressors were grouped into different levels to account for the fact that some re- sponses were general, and some more specific (but still obviously related to a broader issue).

Fig. 2. The percentage of participants from each region of South Australia.

K.M. Fennell et al. / Journal of Rural Studies 46 (2016) 102e110106

3.1. Key stressor: drought

Drought (or lack of rain) was the most commonly reported and most salient theme of stress. Drought was not only identified as being a stressor within itself as Example 1 demonstrates, but also on a deeper level, as the origin of other stressful situations (see Example 2).

6 droughts in 9 years since I took over the farm.

Example 1. Male, mid 40s, Eyre Peninsula.

Little income due to drought. Dealing with the bank- make you feel like you are begging to survive.

Example 2. Male, early 50s, Mid North.

3.2. Key sub-stressors

The next two most salient, broad themes of stress were:

� financial pressure and � uncertainty about the future.

Like drought, these two themes were sometimes identified directly by participants, and were found to be additionally capable of encapsulating broader sources of stress, below the surface of the data. Using financial pressure as an example, the following extracts demonstrate the two levels at which these stressors were identi- fied. In the first (Example 3), financial pressure is evident at the surface level of the data, and in the second (Example 4), it is located less explicitly below the surface.

Overdraft running out.

Example 3. Male, late 40s, Eyre Peninsula.

Having to talk to my in-laws about the strong possibility of stop- ping the business and selling family farm after 6 generations.

Example 4. Female, mid 30s, Mid North.

Financial pressure and uncertainty about the future were considered sub-stressors of drought because as Example 2 dem- onstrates, there was considerable evidence to suggest drought

contributed to their emergence. Example 5 shows how all three of the previously mentioned

stressors (drought, financial pressure, uncertainty about the future), were found to be interrelated.

Finances, lack of rain, doubts about farming in the future because of the lack of price for our commodities, the price of inputs.

Example 5. Male, mid 40s, South East.

3.3. Specific drought-related stressors

A number of causes of stress of a more specific nature also became evident. These were often closely associated with the stressors mentioned previously and included:

� poor crop yields, � unsatisfactory conditions for livestock, � overwhelming workloads (e.g. carting water, feeding sheep, preparing financial reports),

� problems with labour (e.g. having to reduce number of employees),

� strain on relationships/family conflict, � role conflict (e.g. managing a family and having to work off-farm) and

� stress associated with the consideration or execution of selling or leasing a property.

The following example demonstrates how specific and broader stressors were found to commonly interact, in this case over- whelming workload, drought and financial pressure.

The last of our reliable dams dried up - the first time since con- structed in 1953. Our days now have added workload of carting water as well as the growing expense of buying in water supplies.

Example 6. Female, mid 60s, Mid North.

Further, Example 7 demonstrates the association between strain on relationships, drought and financial pressure.

Wife said she couldn’t stand the stress, drought, poverty, and my sourness any longer. Packed up, took the car and left. After 38 years. Probably 35 reasonably happy.

Example 7. Male, early 60s, South East.

Table 1 Summary of the types of ‘most stressful situations’ participants reported experiencing in the month prior to the study.

Type of stressor n % Definition and examples

Farm-related 154 49.8 Stressors solely relating to one’s farming enterprise. � Lack of rain � Doing the farm budget � Workload building up � Lost over $500,000 on grain futures

Non-farm-related 71 23.0 Stressors solely relating to issues removed from one’s farming enterprise. � Death of a friend � Aging parents � Advised of long term medical problem � Daughter upset at boarding school

Combination (farm & other) 72 23.3 Stressors involving a combination of farm and other aspects of one’s life. � Retiring from farm � Arguing with son about farm-related issues � Son’s marriage breaking up and him deciding to leave the farm � Partner upset about lack of rain � A family member and co-farmer suffering severe depression � Water for stock and Mum’s health

Missing 12 3.9 I.e. no stressor stated

K.M. Fennell et al. / Journal of Rural Studies 46 (2016) 102e110 107

See Fig. 3 for further illustrations of how these stressors interrelate.

3.4. Contextual stressors

Additionally, a number of contextual stressors (i.e. factors or pressures other than drought, largely outside of the direct control of farmers and their families that are capable of compounding farm-related stress) were identified. These contextual stressors included:

� rising input costs (e.g. fuel, fertiliser and chemicals), � isolation, � pressure to take part in community work, � family involvement in the farm business, � lack of understanding of rural issues from ‘outsiders’, � and regulation of and interference with agricultural practices by both governments and other organisations (e.g. People for the Ethical Treatment of Animals (PETA) with the mulesing debate and governments with changes to grain-marketing systems).

As with the other stressors, but to a lesser degree, sometimes these factors were identified as direct sources of stress and located at the surface level of the data by participants, while in other cases they were less obvious and identified as underlying themes below the surface of the data.

Example 8 demonstrates how accounts of these contextual types of stressors were also integrated with other types of stressors. In this example contextual stressors, including pressure to take part in community work (“help neighbours”) and isolation (“trouble sourcing products”, “freight getting lost”) are apparent among more specific stressors including overwhelming workloads, role conflict (“little one sick”, “need to help family”) and strain on relationships.

New machinery, not working as we planned, trouble sourcing products, breakdowns and neighbours asking for help at same time, little one sick, need to help family with building new home, freight getting lost, and don’t seem to have enough time to do any one job properly- exhausted and still want to be their (sic) for the parents when they need a hand. Communication [with] spouse fades on any day like this.

Example 8. Female, early 50s, South East.

Further, the following extract is an example of a contextual stressor (regulation and interference with agricultural practices) be- ing integrated with the broader themes of drought and financial pressure and arguably, specific stressors; isolation and uncertainty about the future.

In the office doing the monthly bookwork. Dealing with incoming mail, outgoing expenses, and rules and regulations introduced to make our lives more difficult- making it harder to get income- and expenditure increasing from every angle. All about what we have got instead of what we (living in isolated area) have not got. Watching the bank balance build up then blow away … and the rain clouds do the same.

Example 9. Female, late 40s, Far North.

3.5. Non-drought-related farm stressors

A relatively small proportion of farm-related stressors were identified that could not be integrated into the main model of stressors as they were not found to be closely related to the drought. Examples of such stressors include; machinery breakdowns (see Example 8), disease outbreaks and farm accidents. However, these stressors were included in the model to illustrate the addi- tional type of contextual or compounding stressors faced by farmers. These stressors are positioned in the separate box entitled ‘Compounded by … ’, in Fig. 3.

4. Discussion

This study aimed to identify the specific events or factors causing South Australian farmers (and their spouses) stress during times of drought. A larger proportion of stressful events or episodes were related to farm-life for those in the present study than for the farmers who participated in Walker and Walker’s (1987) Canadian research. This may be a result of the additional farm-related pres- sure placed on farmers by the drought, given that drought was found to be a dominant source of stress for those in the present sample and Walker and Walker’s (1987) research was not con- ducted during a drought. Financial pressure and uncertainty about the future were the next two most salient themes of stressors identified by our sample who were predominantly sheep, cattle and/or grain farmers. There was evidence to suggest that in many cases, both of these stressors were also closely tied to drought. A

Note. Approximately 900 descriptions of farmers’ greatest stressors informed this analysis (participants described three stressors each).

Non-drought-related farm stressors Other non-farm stressors

Drought/lack of rain

Rising input costs

Isolation

Unwanted regulation of and interference with agricultural practices by governments and

other organisations

Pressure to take part in community work

Unsatisfactory conditions for livestock

Overwhelming workloads

Problems with labour

Strain on relationships/family

conflict Stress associated with the consideration or execution of selling or leasing properties

Role conflict

Lack of understanding of rural issues from

‘outsiders’

Poor crop yields

Family involvement in the farm businesses

Compounded by...

Non drought-related farm stressors: machinery break-downs, disease outbreaks (in crops or livestock), farm accidents

Other non-farm stressors: illness, death, family responsibilities

Key stressor

Key sub-stressor

Contextual stressors

Specific drought- related stressors

Financial pressure

Uncertainty about future

Fig. 3. Sources of stress reported by South Australian farmers in a time of drought.

K.M. Fennell et al. / Journal of Rural Studies 46 (2016) 102e110108

number of previous studies have described financial pressure as a major source of stress for farmers in other settings during drought (Deary et al., 1997; Gray and Lawrence, 1996; Hossain et al., 2008; Lunner Kolstrup et al., 2013; Olson and Schellenberg, 1986; Walker, 2012; Staniford et al., 2009). While uncertainty about the future has been flagged as one of a number of farm-stressors by previous research (Walker and Walker, 1987), it has not previously been identified as such a dominant stressor, other than in the recent qualitative research on the effects of drought on rural communities (Sartore et al., 2008; Rickards, 2012).

This research also highlights a range of more specific drought- related issues that contribute to farmers’ levels of stress during drought. These include poor crop yields, unsatisfactory conditions for

livestock, overwhelming workloads, problems with labour, strain on relationships/family conflict, role conflict and stress associated with the consideration or execution of selling or leasing a property. Many of the stressors at this level are consistent with those reported by citrus growers in rural South Australia (Staniford et al., 2009) at the same time the present work was being carried out with sheep, cattle and/or grain farmers. The issues of role conflict, relationship/ family conflict have been highlighted by case study research that showed the significant impact of drought on Australian women (Alston, 2006). Contextual stressors also appear to cause stress or add to it; namely rising input costs, isolation, pressure to take part in community work, family involvement in the farming business, the lack of understanding of rural issues from ‘outsiders’ and regulation of, and

K.M. Fennell et al. / Journal of Rural Studies 46 (2016) 102e110 109

interference with, agricultural practices by governments and other organisations. With the exceptions of poor crop yields, unsatisfactory conditions for livestock and stress associated with the possible ne- cessity of selling or leasing a property (all things closely tied to the drought), interestingly, these stressors have all been identified consistently by previous research as causing farmers stress gener- ally (not specifically during droughts) (Fraser et al., 2005; Gray and Lawrence, 1996; Hossain et al., 2008; Lunner Kolstrup et al., 2013; Walker, 2012; Walker and Walker, 1987).

Sources of stress less obviously linked with drought were also revealed, for example machinery breakdowns, disease outbreaks and farm accidents, which are consistent with previous research (Lunner Kolstrup et al., 2013). However, it should be noted that it is possible that even those stressors to some degree, may be attrib- uted to drought; for example if stress and fatigue levels are high, accidents and less attention being paid to maintenance leading to breakdowns, may be more likely at these times. Illness and death were dominant non-farm-related themes of stress, but reported less frequently. It is unlikely that farmers experience fewer non- farm-related stressors than other populations, but instead, they may perceive these events to be less stressful than farm-related occurrences. Weigel (1981) found for example, that farmers generally rate machinery breakdown and disease outbreak as more stressful than divorce and major illness.

While previous research has already produced exhaustive lists of farm stressors (e.g. Gray and Lawrence, 1996; Olson and Schellenberg, 1986; Walker, 2012), the present study uniquely highlights those most prevalent during drought, and relevant to the South Australian sheep, cattle and grain farming population. Therefore, media campaigns about such stressors may facilitate greater understanding of farmers in the wider community and help reduce some of the frustration (identified in the present study) that farmers endure as a result of poor understanding among ‘outsiders’ of their occupation and way of life, in future periods of drought. Improved public awareness and more widespread acknowledge- ment of the stressors farming communities face may also reduce stigma, empower individuals to seek help, and be something that organisations that promote good mental health and early mental- health help-seeking (e.g. beyondblue in Australia) may be able to assist with. Although it is acknowledged that some rural mental health initiatives have been established since these data were collected, a recent representative survey of regional Australia found that 49% of farmers’ were likely to have a mild or worse mental disorder, compared with 26% of the general rural population (Schirmer et al., 2015), suggesting that additional interventions to improve farmers’ mental well-being are still required.

The identification of dominant stressors is the first step rec- ommended in Kenkel’s stress-coping-support model (Kenkel,1986) which provides a framework for undertaking in the development of primary prevention mental-health programs in rural areas. Iden- tifying stressors is also commended for such purposes by Walker and Walker (1987). Greater awareness of the stressors peculiar to farming may also foster “trust and engagement within clinical relationship[s]” (Booth and Lloyd, 2000: p. 991) and assist in overcoming suspicion many rural people have of mental health services (Fuller et al., 2000). These findings could also inform the development of strategies beyond mental health, to reduce the burden of drought experienced by many farmers. For example, interventions aimed at easing financial pressure (e.g. via loans) and providing some certainty about the future (e.g. via assistance with succession planning) may be of benefit. More broadly, they also highlight the important social value of investing in the develop- ment of drought tolerant crops and strategies such as innovative methods of weather forecasting and feeding stock, in preparation for future periods of drought.

5. Conclusion

This research highlights that in farming communities, financial pressure and uncertainty about the future are major stressors during times of drought. Contextual factors such as rising input costs and isolation also contribute to farmers’ high levels of psychological-distress at these times. It also confirms the complexity of factors that affect farmers’ concerns about the future, during these difficult periods (Rickards, 2012).

The major challenge that lies ahead in the development of mental health intervention strategies for drought-affected farmers, is in striking the delicate balance between raising their awareness of high levels of psychological distress and encouraging them to cope more effectively or seek help, without creating alarm and adding to the burden of stress. Although there are no quick fixes to solve the stress associated with life on the land, this study has made a contribution by drawing attention to a needy sector of the pop- ulation and in laying foundations for tailored and appropriate intervention strategies, to help farmers cope more effectively in future times of drought.

Conflict of interest

No conflicts of interest have been identified.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Acknowledgements

The authors wish to acknowledge the farmers who participated in this research, The University of Adelaide and the South Austra- lian Farmers Federation (now Primary Producers SA), in particular, Deanne Crabb, for assistance with recruitment.

References

Alston, M., 2004. Who is down on the farm? Social aspects of Australian agriculture in the 21st century. Agric. Hum. Values 21, 37e46.

Alston, M., 2006. ‘I’d like to just walk out of here’: Australian Women’s experience of drought. Sociol. Rural 46, 154e170.

Andersen, K., Hawgood, J., Klieve, H., et al., 2010. Suicide in selected occupations in Queensland: evidence from the State suicide register. Aust. N. Z. J. Psychiatry 44, 243e249.

Attride-Stirling, J., 2001. Thematic networks: an Analytic tool for qualitative research. Qual. Res. 1, 385e405.

Australian Bureau of Statistics, 2006. National Health Survey: Summary of Results, 2004-05. Australian Bureau of Statistics, Canberra, ACT.

Australian Bureau of Statistics, 2012. Australian Farming and Farmers, 4102.0-. Australian Social Trends.

Australian Bureau of Statistics, 2014. Value of Agricultural Commodities Produced, Australia 2012-2013. ABS, Canberra.

Barclay, E., Reeve, I., Foskey, R., 2011. Australian Farmers’ Attitudes toward Succes- sion and Inheritance. Keeping it in the Family, International Perspectives on Succession and Retirement on Family Farms. Ashgate, London.

Berry, H.L., Bowen, K., Kjellstrom, T., 2010. Climate change and mental health: a causal pathways framework. Int. J. Public Health 55, 123e132.

Berry, H.L., Hogan, A., Owen, J., et al., 2011. Climate change and farmers’ mental health: risks and responses. Asia Pac. J. Public Health 23, 119Se132S.

Berry, H.L.K., Brian, J., Hanigan, Ivan C., Coates, Jamie H., McMichael, Anthony J., Welsh, Jennifer A., Kjellstrom, Tord, 2008. Garnaut Climate Change Review: Rural Mental Health Impacts of Climate Change.

Booth, N.J., Lloyd, K., 2000. Stress in farmers. Int. J. Soc. Psychiatry 46, 67e73. Braun, V., Clarke, V., 2006. Using thematic analysis in psychology. Qual. Res. Psychol.

3, 77e101. Brumby, S., Kennedy, A., Chandrasekara, A., 2013. Alcohol consumption, obesity, and

psychological distress in farming communitiesdAn Australian study. J. Rural Health 29, 311e319.

Caldwell, K., Boyd, C.P., 2009. Coping and resilience in farming families affected by drought. Rural Remote Health 9, 1088.

Carruth, A.K., Logan, C.A., 2002. Depressive symptoms in farm women: effects of

K.M. Fennell et al. / Journal of Rural Studies 46 (2016) 102e110110

health status and farming lifestyle characteristics, behaviors, and beliefs. J. Community Health 27, 213e228.

Cole, S., Smith, A., Wilson, C., et al., 2007. An advance notification letter increases participation in colorectal cancer screening. J. Med. Screen. 14, 73e75.

Costello, A., Abbas, M., Allen, A., et al., 2009. Managing the health effects of climate change. In: The Lancet, 373. Lancet and University College London Institute for Global Health Commission, pp. 1693e1733.

Crabb, D., 2008. ABS, 2001, ’employees in farming, Aquaculture and fishing by SA NRM regions’. adapted and added to by. In: Crabb, D. (Ed.).

Deary, I.J., Willock, J., McGregor, M., 1997. Stress in farming. Stress Med. 13, 131e136. Dijk, A.I., Beck, H.E., Crosbie, R.S., et al., 2013. The Millennium Drought in southeast

Australia (2001e2009): natural and human causes and implications for water resources, ecosystems, economy, and society. Water Resour. Res. 49, 1040e1057.

Edwards, B., Gray, M., Hunter, B., 2009. A sunburnt country: the economic and financial impact of drought on rural and regional families in Australia in an era of climate change. Aust. J. Labour Econ. 12, 109e131.

Edwards, P., Roberts, I., Clarke, M., et al., 2002. Increasing Response Rates to Postal Questionnaires: Systematic Review.

Folkman, S., Lazarus, R.S., 1988. Ways of Coping Questionnaire Permissions Set, Manual, Test Booklet, Scoring Key. Consulting Psychologists Press, California.

Franklin, R.C., Mitchell, R.J., Driscoll, T.R., et al., 2001. Agricultural work-related fa- talities in Australia, 1989e1992. J. Agric. Saf. Health 7, 213e227.

Fraser, C.E., Smith, K.B., Judd, F., et al., 2005. Farming and mental health problems and mental illness. Int. J. Soc. Psychiatry 51, 340e349.

Fuller, J., Edwards, J., Procter, N., et al., 2000. How definition of mental health problems can influence help seeking in rural and remote communities. Aust. J. Rural Health 8, 148e153.

Gerrard, C.E., 1998. Farmers’ occupational health: cause for concern, cause for ac- tion. J. Adv. Nurs. 28, 155e163.

Gray, I., Lawrence, G., 1996. Predictors of stress among Australian farmers. Aust. J. Soc. Issues 31, 173e189.

Greenhill, J., King, D., Lane, A., et al., 2009. Understanding resilience in South Australian farm families. Rural Soc. 19, 318e325.

Gregoire, A., 2002. The mental health of farmers. Occup. Med. 52, 471e476. Gunn, K.M., Kettler, L.J., Skaczkowski, G., et al., 2012. Farmers’ stress and coping in a

time of drought. Rural Remote Health 12, 2071. Heffernan, B., Keogh, E., McKew, S., 2008. National farmers’ federation Annual re-

view, 2007-08. In: Federation, N.F. (Ed.). ACT, Canberra. Hossain, D., Eley, R., Coutts, J., et al., 2008. Mental health of farmers in Southern

Queensland: issues and support. Aust. J. Rural Health 16, 343e348. Kenkel, M.B., 1986. Stress-coping-support in rural communities: a model for pri-

mary prevention. Am. J. Community Psychol. 14, 457e478. Kennedy, A., Maple, M.J., McKay, K., et al., 2014. Suicide and accidental death in

Australia’s rural farming communities: a review of the literature. Rural Remote Health 2517.

Kongsved, S.M., Basnov, M., Holm-Christensen, K., et al., 2007. Response rate and completeness of questionnaires: a randomized study of internet versus paper-

and-pencil versions. J. Med. Internet Res. 9, e. 25. Lunner Kolstrup, C., Kallioniemi, M., Lundqvist, P., et al., 2013. International per-

spectives on psychosocial working conditions, mental health, and stress of dairy farm operators. J. Agromedicine 18, 244e255.

Mavis, B.E., Brocato, J.J., 1998. Postal surveys versus electronic mail surveys: the tortoise and the hare revisited. Eval. Health Prof. 21, 395e408.

National Farmers Federation, 2008. Farm Facts. Available at: http://www.nff.org.au/ farm-facts.html.

Olson, K.R., Schellenberg, R.P., 1986. Farm stressors. Am. J. Community Psychol. 14, 555e569.

Pachauri, R.K., Allen, M., Barros, V., et al., 2014. Climate Change 2014: Synthesis Report. In: Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change.

Page, A.N., Fragar, L.J., 2002. Suicide in Australian farming, 1988e1997. Aust. N. Z. J. Psychiatry 36, 81e85.

Polain, J.D., Berry, H.L., Hoskin, J.O., 2011. Rapid change, climate adversity and the next ‘Big Dry’: older farmers’ mental health. Aust. J. Rural Health 19, 239e243.

Rickards, L., 2012. Critical Breaking Point: the Effects of Climate Variability, Change and Other Pressures on Farm Households. In: Report for the Birchip Cropping Group and the Sustainable Agriculture Initiative Platform Australia. www.bcg. org.au/cb_pages/publications.php.

Roufeil, L., Lipzker, A., 2007. Psychology services in rural and remote Australia. InPsych 29, 8e11.

Sartore, G.M., Kelly, B., Stain, H., et al., 2008. Control, uncertainty, and expectations for the future: a qualitative study of the impact of drought on a rural Australian community. Rural Remote Health 8, 950.

Sartore, G.M., Kelly, B., Stain, H.J., 2007. Drought and its effect on mental health - how GPs can help. Aust. Fam. Phys. 36, 990e993.

Schirmer, J., Peel, D., Mylek, M., 2015. The 2014 Regional Wellbeing Survey: Farmers and Agriculture. University of Canberra, Canberra.

Staniford, A.K., Dollard, M.F., Guerin, B., 2009. Stress and help-seeking for drought- stricken citrus growers in the Riverland of South Australia. Aust. J. Rural Health 17, 147e154.

Stanke, C., Kerac, M., Prudhomme, C., et al., 2013. Health effects of drought: a sys- tematic review of the evidence. PLoS Curr. Disasters 1.

Walker, J.F., 2012. Mental Health in the Rural Sector: A Review. Walker, L.S., Walker, J.L., 1987. Stressors and symptoms predictive of distress in

farmers. Fam. Relat. 36, 374e378. Weigel, R., 1981. Stress on the farmean overview. PM-Iowa State Univ. Coop. Ext.

Serv. Weigel, R.R., Weigel, D.J., Blundall, J., 1987. Stress, coping, and satisfaction: gener-

ational differences in farm families. Fam. Relat. 36, 45e48. Wheeler, S., Bjornlund, H., Zuo, A., et al., 2012. Handing down the farm? the

increasing uncertainty of irrigated farm succession in Australia. J. Rural Stud. 28, 266e275.

Whittaker, J., Handmer, J., Mercer, D., 2012. Vulnerability to bushfires in rural Australia: a case study from East Gippsland, Victoria. J. Rural Stud. 28, 161e173.