This was an exploratory, mixed-methods study. The research aims were to explore the nature, prevalence, severity and sources of judicial stress in Australia.

Support from Heads of Jurisdiction was essential to the viability of the study. Five courts, from summary to appellate level, were identified as possible participant jurisdictions. Informal consultations with each of the five Heads of Jurisdiction indicated that safeguarding confidentiality and anonymity of both the individual participants and the courts was a key concern. In light of this, formal support was sought and granted from the five Heads of Jurisdiction on the basis that neither the names of the courts nor their geographical locations would be reported. Also in principle and in kind support was extended from the Judicial College of Victoria and the Australasian Institute of Judicial Administration. Subsequently, approval to conduct the research was granted from the Human Research Ethics Committees for the University of Melbourne and the participating courts.

With the support of the Heads of the five participating jurisdictions, the researcher delivered a presentation on Judicial Stress and Wellbeing at each of the courts’ internal judicial conferences. These took place between July 2016 and April 2017. The presentation set out the rationale for empirical research into judicial wellbeing, and explained the aims and methodology of the current research project. During the presentation, printed copies of the survey were distributed, together with stamped envelopes addressed to a secure post office box. To reach judicial officers who were not present at the conferences, and to provide an alternative means of survey completion, a follow-up email was sent the next day to all judicial officers of the relevant court with a link to an electronic version of the survey hosted on Qualtrics — a secure online survey platform. A participant information sheet and consent form accompanied both versions of the survey. Participation was voluntary and anonymous.

The study involved three stages, with participants invited to participate in one, two or all three stages. The first stage was a short survey (Part 1), taking 10–15 minutes to complete, relating to stress symptoms and experiences, and involved the collection of some limited demographic information (see below for details of survey contents). The second stage involved a longer survey (Part 2), taking 15–20 minutes to complete, relating to mental health literacy, burnout, secondary trauma and alcohol use. The third stage comprised an in-depth semi-structured interview, taking 40–60 minutes, relating to particular experiences of work-related stress, the major sources of judicial stress, and ideas for programs and initiatives to reduce judicial stress. At the completion of each stage, participants were invited to opt in to the next stage or to decline further participation in the study. To opt in to a third stage interview, participants provided their contact details by completing a form and returning it in a separate envelope from their completed survey, thereby preserving the anonymity of the survey responses. Participants were offered the opportunity to opt in to receive personalised feedback regarding their responses to the survey measures by providing their email address at the end of the survey.

The rationale for the staged recruitment approach was to balance the competing aims of maximising sample size and optimising data richness. In the knowledge of judicial officers’ general time poverty, it was anticipated that a shorter survey would attract greater voluntary participation than a longer survey. However, in order to collect data on additional relevant measures, the longer survey was retained as an option for those judicial officers prepared to dedicate the additional time.

The rationale for offering personalised feedback was to balance another pair of competing considerations — anonymity of survey completion, and notifying and supporting participants when scores indicated psychological distress. Given the survey included several widely used screening tools for mental health concerns, scores in the very high ranges could warrant further investigation and possibly supportive intervention. However, the provision of such support was dependent upon being able to contact the relevant participants and, therefore, incompatible with participant anonymity. In order to reconcile these competing concerns, participants had the option to voluntarily forsake their anonymity by providing an email address in order to receive a personalised feedback report on their survey scores. Seventy-three participants (48%) took up this option. The report, which was emailed as a password-protected document, set out several options for independent and confidential psychological support, and recommended accessing such support if any scores were in the high or severe range.

The survey was closed in May 2017. Hardcopy and online survey data were entered into the Software Package for the Social Sciences (SPSS) by an independent research assistant and checked by the researcher prior to analysis.

Five Australian courts, from summary to appellate level, participated in the study. From these courts, 152 judicial officers participated in Part 1 of the survey.¹⁵ Of these, 125 went on to complete Part 2 of the survey, and 60 participated in in-depth interviews. Table 1 sets out the sample size according to jurisdiction, gender, age, stage of career and main area of practice. Participants’ average age was 57.8 years (range = 38–71; standard deviation = 6.9) and average length of service was 9.5 years (range = .2–30; standard deviation = 6.6).

Note: *This demographic data was not collected for interviewees.

As explained above, approval for the research project was obtained on the basis that neither the names nor geographical locations of the participating courts would be reported. It is not possible to report specific response rates according to these demographic variables without risking identification of the courts involved; however, two important points can be made. First, within each of the five participating courts the response rates were very strong, varying between 51% and 85%, and averaging 67%. That is, of the judicial officers to whom the survey was distributed, 67% responded by completing at least Part 1. Second, the sample was, with one exception, broadly representative across the five demographic variables measured. The exception was that female judicial officers were significantly over-represented among those who chose to participate in an interview: women comprised 43.4% of survey respondents, but 60% of interviewees. The interview data therefore will need to be interpreted with this in mind.

In constructing the survey, much thought was given to the choice of both psychological constructs to be measured and psychometric instruments by which to measure them. The term psychological construct arises from the field of empirical psychology and refers to psychological phenomena and experiences one may want to measure and analyse, but are not directly observable — for example, intelligence, happiness and stress. Psychometric instruments are the questionnaires, scales, tests and tasks that have been scientifically developed to validly and reliably measure different psychological constructs. While stress and wellbeing are psychological constructs in their own right, their conceptual breadth is such that they could more accurately be described as psychological categories comprising a broad range of qualitatively different psychological experiences, each of which would be defined and measured differently.

In the current study, the following principles guided the choice of psychological constructs and psychometric instruments:

The application of these principles, and the piloting of a draft survey in May 2016 with a group of 14 retired judicial offices, lead to the inclusion of the constructs and instruments set out in Table 2. These are briefly described in detail in Part II below.

Notes:

^A B Ilardi et al, “Employee and supervisor ratings of motivation: main effects and discrepancies associated with job satisfaction and adjustment in a factory setting” (1993) 23(21) Journal of Applied Social Psychology 1789; T Kasser, J Davey and R Ryan, “Motivation and employee-supervisor discrepancies in psychiatric vocational rehabilitation setting” (1992) 37(3) Rehabilitation Psychology 175; E Deci et al, “Need satisfaction, motivation, and well-being in the work organisations of a former Eastern Bloc country: a cross-cultural study of self-determination” (2001) 27(8) Personality and Social Psychology Bulletin 930.

^B R Kessler et al, “Short screening scales to monitor population prevalences and trends in non-specific psychological distress” (2002) 32 Psychological Medicine 959.

^C P Lovibond and S Lovibond, “The structure of negative emotional states: comparison of the depression anxiety stress scales (DASS) with the Beck Depression Inventories” (1995) 33(3) Behaviour Research and Therapy 335.

^D N Highet, I Hickie and T Davenport, “Monitoring awareness of and attitudes to depression in Australia” (2002) 176 Medical Journal of Australia S63.

^E W Schaufeli et al, “MBI general survey”, 1996, available for purchase from www.mindgarden.com; W Schaufeli, M Leiter and R Kalimo, “The Maslach Burnout Inventory — general survey: a self-report questionnaire to assess burnout in the workplace” in M Leiter (ed), “Extending the burnout construct: reflecting changing career paths”, APA/NIOSH Conference, Work, Stress and Health, Creating Healthier Workplaces, 14–16 September 1995, Washington, DC, 1995. Copyright © 1996 W Scjaufeli, M Leiter, C Maslach and S Jackson, Licence to reproduce in hardcopy and online format granted on 23 June 2016.

^FB Bride et al, “Development and validation of the secondary traumatic stress scale” (2004) 14(1) Research on Social Work Practice 27.

^GJ Saunders et al, “Development of the alcohol use disorders identification test (AUDIT): WHO collaborative project on early detection of persons with harmful alcohol consumption II” (1993) 88 Addiction 791.

Judicial officers who completed both parts of the survey were invited to participate in an in-depth semi-structured interview. Sixty judicial officers accepted this invitation by returning a form with their contact details and signing a consent form at the start of their interview. The interviews were conducted by the lead researcher in the judicial officers’ chambers between November 2016 and July 2017, each lasting approximately one hour. They were audio-recorded and independently transcribed. All identifying information was removed prior to analysis. The interview questions were developed following a review of the literature relating to judicial stress and discussions with research colleagues and several judicial officers regarding the areas of greatest relevance to understanding and responding to judicial occupational stress. The first draft of the questions was developed by the lead researcher for an earlier internal project within one of the participating jurisdictions, and further refined following feedback from that pilot. The final list of nine stem questions, each with several supplementary questions, was framed around four board topics: experiences of judicial stress; perceived sources of judicial stress and satisfaction; strategies for managing judicial stress; and ideas of initiatives to support judicial wellbeing.

Over the course of conducting the 60 interviews, the researcher noted themes that emerged as prominent and recurring, refining these throughout the interview process. This preliminary observational process yielded six key themes, which provided the initial basis of the formal thematic analysis of interview transcripts. As a further step, the researcher read the interview transcripts, totalling 1,386 pages and over 400,000 words, in full three times. In the first reading, the researcher noted all statements that supported or contradicted the six key themes, also recording other recurring themes. From these, codes were created for each theme. In the second reading, the codes were applied to the relevant passages and colour-coded on the transcript. In the third reading, the colour-coded passages were reviewed for patterns and themes, and names applied to each of the themes. A final review was undertaken, using the Microsoft Word search function, to extract the passages that best illustrated the identified themes.¹⁶

Three questions provided an initial measure of judicial officers’ perceived stress and wellbeing. The first two related to the frequency (1 = none of the time; 5 = almost all the time) that judicial officers had experienced stress and personal wellbeing and satisfaction related to their work since their appointment. As shown in Fig 1, perceived stress was negatively skewed and perceived wellbeing was positively skewed. Approximately one-eighth (12.7%) of respondents reported experiencing stress most or almost all of the time, while more than three-quarters (76.0%) reporting experiencing personal wellbeing and satisfaction most or almost all of the time. An analysis of individual responses revealed that subjective experiences of stress and wellbeing were not mutually exclusive — for example, of those who reported experiencing stress most of the time, the majority (60.0%) also reported experiencing personal wellbeing and satisfaction most of the time.

A third question was directed to perceptions of the relative stress (1 = much less stressful; 5 = much more stressful) of judicial office compared with respondents’ career prior to appointment. The results are show in Fig 2. Sixty-two per cent of judicial officers reported that judicial office was a little or much less stressful than their pre-appointment careers, 20.3% reported that it was a little or much more stressful, with the remainder (17.7%) reporting that it was equally as stressful.

The descriptive statistics for judicial officers’ K10 scores were as follows: mean = 16.64; standard deviation = 4.87; possible range = 10–50; observed range = 10–31. The K10 asks about the frequency of experiences over the past four weeks. The highest scoring items were about how often did you feel tired out for no good reason? and about how often did you feel nervous?, with 14.2% and 4.5% of judicial officers respectively reporting they felt this way most of the time or all of the time over the past four weeks. As noted above, K10 scores can be categorised into one of four severity ranges, from low or no distress through to severe distress. The frequencies of K10 scores within each severity range for judicial officers in the present study, and those reported in previous studies of the Australian legal profession⁷⁸ and general population,⁷⁹ are set out in Table 3 and expressed graphically in Fig 3. The majority (52.9%) of judicial officers’ scores indicated some level of non-specific psychological distress, with 14.8% of scores indicating high or very high distress.

Sources: Data for barristers, solicitors and law students is as reported in the landmark study by N Kelk et al, Courting the blues: attitudes towards depression in Australian law students and legal practitioners, Monograph 2009-1, Brain and Mind Research Institute, 2009; Brain and Mind Research Institute, January 2009, 9; data for the general population are as reported in Australian Bureau of Statistics, Use of the Kessler Psychological Distress Scale in ABS Health Surveys, Australia, 2007–2008, National Health Survey 2007–2008 (Information Paper, 4817.0.55.001, 4 April 2012).

Sources: Data for barristers, solicitors, and law students is as reported in the landmark study by N Kelk et al, Courting the blues: attitudes towards depression in Australian law students and legal practitioners, Monograph 2009-1, Brain and Mind Research Institute, 2009 9; data for the general population are as reported in Australian Bureau of Statistics, Use of the Kessler Psychological Distress Scale in ABS Health Surveys, Australia, 2007–2008, National Health Survey 2007–2008 (Information Paper, 4817.0.55.001, 4 April 2012).

These results indicate that Australian judicial officers experience elevated rates of non-specific psychological distress, as measured by the K10, in the moderate to high ranges, compared to the general population and the barrister arm of the legal profession. However, when we consider distress within the very high range, judicial officers’ rates are considerably lower than all levels of the profession and the general population.

The descriptive statistics for judicial officers’ scores on the DASS-21, across the three sub-scales of depression, anxiety and stress, are set out in Table 4. The frequencies of judicial officers’ scores within each of the five normative severity ranges are set out in Table 5.

Note: Score cut-offs for the severity ranges are derived from normative percentiles as set out in the DASS official manual.

The three most strongly endorsed DASS-21 items were all from the stress sub-scale: I found it hard to wind down; I found it difficult to relax; and I found myself getting agitated. The percentage of respondents who indicated that these experiences applied to them a good part or most of the time over the past week was 25.0%, 15.3% and 8.6%, respectively.

A number of lawyer and law student wellbeing studies have focused on the rates of scores in the moderate to extremely severe ranges — sometimes reporting rates three times higher than the general populations — as an indication of the scale of the mental health problem afflicting the legal profession.⁸⁰ Figure 4 shows the rates of judicial officers’ scores in the moderate to extremely severe ranges, compared with Australian lawyers and the general population — according to results reported in a large-scale study of the Australian legal profession (n = 965)⁸¹ and the population normative percentiles set out in the DASS-21 official manual.⁸² The results show that judicial officers’ rates of moderate to extremely severe depression, anxiety and stress symptoms (9.3%, 7.2% and 7.8%, respectively) were dramatically lower than those reported for Australian lawyers, and lower also than those suggested for the Australian general population.

Sources: J Chan, S Poynton and J Bruce, “Lawyering stress and work culture: an Australian study” (2014) 37(3) UNSW Law Journal 1062 at 1081; P Lovibond and S Lovibond, Manual for the Depression Anxiety Stress Scales (DASS), Psychology Foundation Monograph, 1993 (DASS Manual).

The descriptive statistics for judicial officers’ scores on the MBI-GS, across the three burnout sub-scales of exhaustion, cynicism and professional efficacy, are set out in Table 6. Scores on the professional efficacy sub-scale are interpreted in reverse to the exhaustion and cynicism sub-scales — that is burnout is associated with higher exhaustion and cynicism scores, but lower professional efficacy scores. The frequencies of judicial officers’ scores within these normative severity ranges of low, moderate and high are set out in Table 7. On each of the exhaustion and cynicism sub-scales, a little over one quarter (26.6% and 29.6%, respectively) of judicial officers scored in the high range. Approximately one in 10 (10.4%) scored in the low range on the professional efficacy sub-scale. Four per cent (4.0%) scored high on both exhaustion and cynicism, and low on professional efficacy, indicating high risk of burnout, whereas 24.8% scored low on both exhaustion and cynicism, and high on professional efficacy, indicating low risk of burnout.

Note: One incomplete response on the exhaustion sub-scale lead to an overall sample size of n = 124 for that sub-scale.

Note: One incomplete response on the “exhaustion” sub-scale lead to an overall sample size of n = 124 for that sub-scale.

The most strongly endorsed items were all on the professional efficacy sub-scale, for which high scores indicate lower burnout. Among the items for which high scores indicate higher burnout (ie items on the exhaustion and cynicism sub-scales), the three most strongly endorsed were: I feel used up at the end of the workday (exhaustion); I just want to do my work and not be bothered (cynicism); and I feel emotionally drained by my work (exhaustion). The percentages of respondents who indicated that they experience these feelings at least once a week were 40.4%, 38.4%, and 30.4%, respectively. Almost 8% (7.8%) of respondents indicated that they feel used up at the end of the workday every day.

To the knowledge of the author, the MBI-GS has not been used in other studies involving judicial officers or lawyers. It has, however, been used in a number of large-scale North American and European studies across a range of occupational groups (including military, nurses, civil servants and psychiatric workers).⁸³ Figure 5 shows judicial officers’ mean scores across each of the sub-scales, alongside those reported for different occupational groups in previous studies. The results reveal that, on average, judicial officers in this sample experienced moderate levels of exhaustion and cynicism, but very high levels of professional efficacy, compared with the available data for other occupational groups.

Source: C Maslach, S Jackson and M Leiter, Maslach Burnout Inventory Manual, Mind Garden, 3 edn, 2013, (MBI Manual) p 24.

The descriptive statistics for judicial officers’ raw and average scores on the STSS, for total STSS and across the three sub-scales of intrusion, avoidance and arousal, are set out in Table 8. An STS symptom was considered to be endorsed if the respondent indicated that it was experienced occasionally, often or very often in the preceding seven days. The great majority (83.6%) of judicial officers endorsed at least one STS symptom, with the arousal sub-scale (mean = 2.04; standard deviation = .70) more strongly endorsed than the intrusion (mean = 1.83; standard deviation = .62) and the avoidance (mean = 1.79; standard deviation = .63) sub-scales. The most strongly endorsed items were: I had trouble sleeping (arousal; endorsed by 48.7%); I thought about my court work when I didn’t intend to (intrusion; endorsed by 46.7%); and I was easily annoyed (arousal; endorsed by 37.5%). Responses to a number of other items are also notable: 22.4% of judicial officers endorsed I wanted to avoid working on certain types of cases or cases involving certain people (avoidance); 18.4% endorsed it seemed as if I was reliving the traumas experienced by a person/some people who came before me (intrusion); and 10.5% endorsed I had disturbing dreams about the people that came before me (intrusion). Figure 6 shows judicial officers’ mean scores for the three STSS sub-scales, as compared with those reported for three occupational groups in the United States: immigration judges (n = 96);⁸⁴ social workers (n = 287);⁸⁵ and clinical social workers (n = 256).⁸⁶

Note: Several incomplete responses lead to slightly different sample sizes across the three sub-scales: intrusion (n = 123); avoidance (n = 124); arousal (n = 125); and total STSS (n = 122).

Sources: Immigration judges (S Lustig et al, “Burnout and stress among United States Immigration Judges” (2008) 13 Bender’s Immigration Bulletin 22; social workers (B Bride et al, “Development and validation of the secondary traumatic stress scale” (2004) 14(1) Research on Social Work Practice 27); clinical social workers (J Caringi et al, “Secondary traumatic stress and licensed clinical social workers” (2016) 23(2) Traumatology 186 at 190).

The frequencies of judicial officers’ total STSS raw scores within each severity range are set out in Table 9. The results indicate that 62.3% of judicial officers were experiencing mild or higher levels of STS, with 13.2% experiencing high or severe STS. As noted above, STSS items mirror the diagnostic criteria for PTSD⁸⁷ and its lead author has suggested it could be used as a proxy measure of PTSD symptomatology.⁸⁸ Total STSS raw scores of 38 or higher (ie scores in the moderate to severe ranges) are considered pertinent to the question of whether formal assessment for PTSD might be indicated.⁸⁹ In the current study, 30.4% (n = 37) of judicial officers scored at or above 38. Figure 7 shows judicial officers’ rates of scores at or over 38, alongside those previously reported for two samples of US social workers.

Note: Severity ranges and score cut-offs were suggested by Bride based on percentile ranks of total STSS raw scores found in his study of 282 US social workers; see B Bride, “Prevalence of secondary traumatic stress among social workers” (2007) 52(1) Social Work 63.

Sources: B Bride, “Prevalence of secondary traumatic stress among social workers” (2007) 52(1) Social Work 63 at 67; J Caringi et al, “Secondary traumatic stress and licensed clinical social workers” (2016) 23(2) Traumatology 186 at 190.

The descriptive statistics for judicial officers’ AUDIT scores were as follows: mean = 6.12; standard deviation = 4.11; possible range = 0–30; observed range = 0–23. The three most strongly endorsed items all related to frequency and quantity of drinking, as opposed to dependent or harmful drinking. The highest scoring item was How often do you have a drink containing alcohol, with 76% of judicial officers indicating that they drink at least twice weekly. Almost half (48.4%) reported that they have at least 3–4 standard drinks on a typical day when they are drinking. Fourteen per cent (14.6%) reported that they consume six or more standard drinks on one occasion at least weekly.

The frequencies of judicial officers’ AUDIT scores within each normative level-of-risk range, alongside those reported in previous studies of the Australian legal profession⁹⁰ and general population,⁹¹ are set out in Table 10 and expressed graphically in Fig 8. Just over 30% of judicial officers scored in the moderate- to high-risk or problematic use levels, as compared with 32% of lawyers and 18.8% of the general population.

Sources: J Chan, S Poynton and J Bruce, “Lawyering stress and work culture: an Australian study” (2014) 37(3) UNSW Law Journal 1062 at 1087; C Moller, R Tait, and D Byrne, “Self-harm, substance use and psychological distress in the Australian general population” (2013) 108(1) Addiction 211 at 215.

Sources: J Chan, S Poynton and J Bruce, “Lawyering stress and work culture: an Australian study” (2014) 37(3) UNSW Law Journal 1062 at 1087; C Moller, R Tait, and D Byrne, “Self-harm, substance use and psychological distress in the Australian general population” (2013) 108(1) Addiction 211 at 215.

A number of important findings have arisen from this research. First, it has revealed that the pattern of stress and psychological ill-health among judicial officers in the study is different from that which has been reported for the Australian legal profession. This finding draws upon the results from the K-10, which measures psychological distress, and the DASS-21, which measures depressive and anxious symptoms. For the former, compared with K-10 data published for Australian barristers⁹² and the general population,⁹³ judicial officers reported elevated rates of non-specific psychological distress in the moderate to high ranges. However, when considering distress in the very high range, judicial officers’ rates were considerably lower than all levels of the profession and the general population. Alongside the K-10 results, the DASS-21 results show that judicial officers reported symptoms of depression and anxiety at rates similar to (and, in the moderate to extremely severe ranges, lower than) those suggested for the general population,⁹⁴ which is dramatically lower than those previously found for the broader legal profession.⁹⁵ In fact, in the moderate to extremely severe ranges, lawyers’ reported rates of depression, anxiety and stress symptoms were more than three times those found for judicial officers in this study. Taken together these results suggest that while there is not a widespread mental health problem among the Australian judiciary, there is a stress problem. This is consistent with the limited empirical research on judicial stress from the United States, which has collectively suggested that judicial officers experience elevated occupational stress, but their rates of stress and mental ill-health are generally lower than practising lawyers.⁹⁶ Given that judicial officers are, almost without exception, drawn from the population of practising lawyers, this difference in rates of mental ill-health between the two groups is interesting. It may be due to a number of factors. It may indicate that judicial work is quantitatively less demanding than legal practice, such that a key driver of mental ill-health within the legal profession is reduced among judges. Alternatively, it could suggest that those appointed to judicial office are generally well-adapted to legal work and well suited to the judicial role — in other words, that the appointment process is generally sound. Finally, it may be a reflection of the well-documented observation that late middle life (the age at which most judicial officers serve their time in office) is a period of relative mental stability. Indeed, it may indicate a combination of all three. Of course, the impact of these factors may vary between different levels of the court hierarchy and different practice areas; these issues will be explored in forthcoming articles.

Second, this research found that symptoms of burnout and secondary trauma were features of the occupational stress experience for many judicial officers. Considering first the MBI-GS burnout results, while only 4.0% of participants’ scores fell within the highest risk profile (ie high exhaustion, high cynicism and low professional efficacy), still only 24.8% fell within the lowest risk profile (ie low exhaustion, low cynicism and high professional efficacy), indicating that three-quarters (75.2%) of judicial officers in this study reported scores on at least one sub-scale that was consistent with some level of burnout risk. The burnout symptoms reported by judicial officers’ were chiefly along the axes of exhaustion and cynicism; while reported experiences of professional efficacy were generally very strong. The mean scores for both exhaustion and cynicism were in the moderate range, whereas the mean score for professional efficacy was in the high range — and higher, in fact, than the professional efficacy means reported for other occupational groups, such as psychiatric workers, civil servants and the military.⁹⁷ These results suggest that although judicial officers’ are vulnerable to burnout, their experiences of burnout symptoms are likely to be characterised by feelings of emotional depletion and loss of meaning, rather than feelings of incompetence and ineffectiveness.

Turning to the STSS results, the overwhelming majority of judicial officers in this study (83.6%) endorsed at least one STS symptom, with almost a third (30.4%) obtaining total STSS scores in the moderate to severe ranges. Given that these statistics relate to symptoms experienced occasionally, often or very often in the one week prior to survey completion, these results suggest that, at any given point in time, a considerable proportion of judicial officers are burdened with one or more symptoms of STS. Promoting the STSS as a proxy measure of PTSD, Bride, the lead author of the STSS, has recommended that a total STSS score of 38 or higher be interpreted as indicating PTSD due to STS⁹⁸ — in other words, all people scoring in the moderate to severe ranges can be considered, according to Bride, likely to meet diagnostic criteria for PTSD.⁹⁹ By this analysis, 30.4% of judicial officers in the present study should be considered to be experiencing symptoms similar to, and likely qualifying for, PTSD. Although Bride reported good sensitivity and specificity for this cut-off value, this methodology has been followed in only one other study¹⁰⁰ and hence has not been externally validated. Further, the STS does not address PTSD diagnostic criteria relating to duration of symptoms and functional impairment.¹⁰¹ It therefore provides, at best, a very crude basis upon which to infer post-traumatic psychopathology. Nonetheless, participants’ mean scores fell somewhere between those previously reported for other high stress professional groups, US social workers¹⁰² and US immigration judges,¹⁰³ suggesting that STS is a prominent feature of the occupational stress experienced by Australian judicial officers. This is consistent with several American studies which have explored secondary trauma (and the related constructs of vicarious trauma and compassion fatigue) among judges.¹⁰⁴ Third, the study found that judicial officers’ alcohol use was comparable to that reported for the Australian legal profession, which is considerably higher than the documented level of use within the Australian general population. A little over 30% of judicial officers’ AUDIT scores were in the medium- to high-risk levels, indicating problematic alcohol use; previous studies with Australian lawyers¹⁰⁵ and the general population¹⁰⁶ reported rates of medium- to high-risk scores as 32.0% and 18.8%, respectively. To the knowledge of the researcher, this is the first study internationally to measure judicial officers’ alcohol use. A number of Australian¹⁰⁷ and American¹⁰⁸ studies have investigated alcohol use among practising lawyers, frequently reporting that lawyers’ levels of use are higher than those of other occupational groups, especially as a way of managing stress and depression. The Law Institute of Victoria’s 2012 report, Mental Health and the Legal Profession: A Preventative Strategy, suggested that there may be a culture within the profession of self-medicating, noting alcohol abuse (and, to a lesser extent, substance abuse) is prevalent in the legal profession as a maladaptive strategy for dealing with stress.¹⁰⁹ Although the present study did not investigate judicial officers’ functional goals for alcohol use, the results suggest that the pattern of alcohol consumption within the broader Australian legal profession extends to judicial officers.

The present study makes a significant contribution to the conversation on judicial wellbeing taking place in Australia and around the world. As the first research to empirically measure levels of stress and psychological ill-health among Australian judicial officers, the present study is groundbreaking nationally, and — given the limited number of judicial stress studies undertaken using empirically validated measurement instruments¹¹⁰ — it is pioneering internationally. In addition, whereas previous studies of lawyer and judicial wellbeing have generally confined the analysis to one or two stress constructs, the present study operationalised judicial stress and wellbeing across six qualitatively different stress constructs and used empirically validated psychometric instruments to measure each. This expansive and rigorous measurement approach has afforded a rich and reliable picture of the varied ways in which the unique pressures of judicial office are experienced by judges and magistrates, and how judicial stress compares to the stress experienced by lawyers, other professionals and the general population. The clinical and social significance of this study is considerable. It directly targeted the suffering of senior public officials, judges and magistrates, which is not readily acknowledged and was historically taboo.¹¹¹ Given the pivotal role that judicial officers play in Australia's democratic system and the daily impact of their decisions on people's lives, their psychological health and wellbeing is a vital community concern. As more and more judicial officers speak out about their experiences of stress and mental ill-health on the bench, and as courts in Australia and around the world begin to grapple with the question of how best to support judicial officers in their complex and critical work, the results of this study provide an initial basis for appropriate intervention and support.

There were of course some limitations with the present study. First, while this was the first empirical research into judicial wellbeing in Australia, it was not a fully national study. There are, at the time of writing, 38 courts in Australia;¹¹² five were invited to participate in the study. These five spanned the court hierarchy from summary to appellate level and, given the study’s strong (67%) response rate and broad representativeness across the demographic variables, the survey data can be considered to reliably reflect the prevalence and severity of judicial stress within the five participating courts. However, the question of whether these findings can be generalised to the Australian judiciary as a whole is more nuanced. The five participating courts represent just over one-eighth of the courts currently operating within the Australian federal, State and Territory systems, and the 152 judicial participants represent approximately 14.4% of the Australian judicial population, which — at the time of writing — is estimated to number 1,052 judges and magistrates.¹¹³ It should be remembered in interpreting the results that the context and circumstances of judicial work varies considerably from jurisdiction to jurisdiction, and therefore the prevalence and severity findings generated by this study may not be representative of every Australian court.

Second, several aspects of the study design and recruitment methodology may have influenced the sample and responses. First, regarding the use of a survey — self-report bias and social-desirability bias are well-documented phenomena,¹¹⁴ and are limitations intrinsic to all survey-based psychological research. In this regard, it is possible that participants’ responses reflected their aspirational, rather than their actual, experiences. For example, in responding to the AUDIT, participants may have indicated their intended frequency and quantity of alcohol use, which may well be considerably lower than their actual level of use.¹¹⁵ Even if participants were trying to answer honestly, some may have lacked the introspective ability to provide accurate responses. Given that most of the measures assessed the frequency and intensity of negative stress symptoms, a lack of insight or introspection would likely lead to an under-reporting of symptoms. Although validated measurement instruments, like those used in this study, are scientifically developed to minimise the impact of self-report bias, it is not possible to completely eliminate under- or over-reporting of experience. This, of course, is equally a limitation for the survey-based studies cited throughout this article relating to the legal profession and the general population. However, given the reputed stress denying culture¹¹⁶ of the Australian judiciary, it is possible that self-report and social-desirability biases are more pronounced among judicial officers. Second, the recruitment methodology (distributing the survey during internal courts conferences) may not have reached judicial officers’ at their most stressed, given that conferences usually provide a short reprieve from the many stressors faced by judicial officers in their daily work. It was also unlikely to have reached the most acutely unwell judicial officers, who may have been on extended sick or stress leave or who may not have been prepared to dedicate the time to survey completion. While these limitations cannot be overlooked, they are arguably most likely to have contributed to an under-representation of the true prevalence and severity of judicial stress, such that is it possible the real extent of the stress experienced by judicial officers is higher than the findings suggest.

The results of the present study raise a number of questions that should be addressed in future research. Perhaps most importantly: are the findings indicative of the state of stress and wellbeing among judicial officers throughout Australia, or are they the idiosyncratic experience of the five participating courts? The Australian judicial system is diverse and dynamic, and while there are undoubtedly many commonalities in judicial work across the Australian courts, it is equally apparent that courts vary dramatically in terms of workload, work type, and work culture, any or all of which may impact the pressures confronted by judicial officers and the stress they experience. The present study highlights the need for a national study, involving all Australian courts, to determine the overall nature, prevalence and severity of judicial stress in Australia and to identify jurisdictions, practice areas and demographic groups that may be particularly affected. If such a study were to occur, it would be a world first. National judicial studies have been undertaken in Australia¹¹⁷ and the United Kingdom¹¹⁸ with great success and practical application; however, to date these have focused on workload, job satisfaction and judicial attitudes, and have not included validated measures of stress and wellbeing. As already noted, the small number of previous studies measuring judicial stress were all relatively small scale and jurisdiction specific. No country has yet embarked upon a full-scale national study of judicial wellbeing.

A related question is to what extent the findings of the current study reflect the general experience of judicial stress around the world. The current research adds to a very small body of empirical research into the psychological impact of judicial work, most of which has been conducted in North America. Judges around the world encounter many of the same occupational stressors (heavy workloads, dealing with distressing subject matter, daily immersion in conflict and disagreement, responsibility for decisions that significantly impact people’s lives); however, there are obviously also enormous differences in terms of history and culture, political and social context, terms and conditions, and practical support, to name but a few. Future studies into judicial stress and wellbeing in other countries should attempt to use the same validated measurement instruments as used in the current study, or one of the previous North American studies, to enable meaningful transnational comparisons.

A further area that needs to be explored is the positive side of judicial wellbeing. The present study has focused on the negative side of judicial wellbeing — that is, judicial stress. However, the results from the three questions measuring perceived stress and wellbeing provide an initial indication that personal wellbeing and satisfaction were prominent feelings alongside the stress of the role. It is possible that the sources of fulfilment, accomplishment and purpose within judicial work compensate or offset for the sources of stress, providing for a demanding but meaningful professional life. As research within the relatively recent but now very well-established field of Positive Psychology has demonstrated, meaning may be more integral to sustained wellbeing than ease or pleasure.¹¹⁹ Numerous validated measures of positive wellbeing now exist. Future studies could look into the nature and parameters of judicial wellbeing and satisfaction and explore the relationships between wellbeing variables and stress variables.

Finally, future research could attempt to determine the workplace factors that contribute greater judicial stress or wellbeing. There is much to guide such an inquiry. Extensive research in the fields of occupational stress and organisational psychology has generated a large evidence base for a number of models of workplace wellbeing, including the Mental Health Commission of Canada’s 13 factors for psychological health and safety in the workplace,¹²⁰ which were adopted by the Tristan Jepson Memorial Foundation (now Minds Count, the peak body for lawyer wellbeing in Australia) in its best practice guidelines for psychologically healthy legal workplaces.¹²¹ The 13 factors include organisational culture, civility and respect, psychological demands, and workload management, and provide a framework both for assessing the extent to which a workplace is supportive of psychological health and wellbeing, and for developing initiatives to improve wellbeing. Future research could be directed to exploring the presence and impact of these factors within the court environment, and their relationship to experiences of judicial stress.