Poynton, Suzanne; Chan, Janet; Vogt, Melissa; Grunseit, Anne; Bruce, Jasmine --- "Assessing The Effectiveness Of Wellbeing Initiatives For Lawyers And Support Staff" [2018] UNSWLawJl 20; (2018) 41(2) UNSW Law Journal 584

	Home \| Databases \| WorldLII \| Search \| Feedback University of New South Wales Law Journal

ASSESSING THE EFFECTIVENESS OF WELLBEING INITIATIVES FOR LAWYERS AND SUPPORT STAFF

SUZANNE POYNTON,^[*] JANET CHAN,^[**] MELISSA VOGT,^[***] ANNE GRUNSEIT^[****] AND JASMINE BRUCE^[*****]

This article reports on the findings of a study on the use and effectiveness of wellbeing initiatives implemented by a large public sector legal service organisation in Australia. The study, which was conducted from 2011 to 2015, employed a mixed-methods approach, consisting of three waves of staff surveys, interviews with key informants and staff, and workplace observation. The article considers the drivers of work stress, the impact of the introduction of certain initiatives on the mental health and wellbeing outcomes of participants, and participants’ perceptions of the characteristics necessary for an effective intervention and, more broadly, a supportive work environment. In doing so, the article seeks to inform a more general understanding of the prospects and limitations of wellbeing initiatives for addressing mental health and stress issues in the legal profession.

I INTRODUCTION

In recent years, awareness of mental health and wellbeing issues within the legal profession has increased significantly in Australia and internationally. As the body of empirical research finding high levels of stress, depression and anxiety amongst legal practitioners has grown, professional bodies, legal education institutions and workplaces have responded by introducing a range of initiatives. These initiatives are intended to serve multiple purposes, including raising awareness of mental health and wellbeing issues among management and/or employees, reducing workplace stress, building lawyers’ resilience, and supporting lawyers undergoing challenges. Despite the considerable effort and resources directed towards the implementation of wellbeing initiatives, however, there has been limited systematic evaluation of their impacts and outcomes.

This article seeks to respond to this gap in the empirical literature by reporting the findings of a study on the use and effectiveness of wellbeing initiatives implemented by a large public sector legal service organisation in Australia. The study, which was conducted from 2011 to 2015, employed a mixed-methods approach, consisting of three waves of staff surveys, interviews with key informants and staff, and workplace observation. The article considers the drivers of work stress among employees, the impact of the introduction of certain initiatives on the mental health and wellbeing of participants, and participants’ perceptions of the characteristics necessary for an effective intervention and, more broadly, a supportive work environment. In doing so, the article seeks to inform a more general understanding of the prospects and limitations of wellbeing initiatives for addressing mental health and stress issues in the legal profession.

The article proceeds as follows: Part II provides an overview of previous research on mental health and wellbeing in the Australian legal profession, models for understanding lawyering stress, and interventions in response to these findings. Part III provides details of the study. Part IV identifies the most significant drivers of work stress among lawyers and support staff. Part V presents key findings on the awareness and usage, impact and perceived effectiveness of wellbeing initiatives provided by the subject organisation. Part VI concludes by considering the implementation of these findings for the development of wellbeing initiatives within the legal profession.

II RESEARCH ON LAWYERING STRESS AND WELLBEING INITIATIVES

This Part reviews the research evidence on the prevalence of mental health and wellbeing issues among lawyers in Australia and elsewhere. It also reviews the literature on models for explaining work stress and the interventions directed at reducing work stress.

A Mental Health and Wellbeing in the Legal Profession

The Australian legal profession has been consistently recognised in empirical research as having a particularly high rate of mental health issues when compared with other professions and the general population. In the 2006 Beaton Consulting survey of Australian professions, around 16 per cent of legal professionals reported moderate to severe symptoms of depression, compared with 6 per cent of the general population, 8 per cent of engineers and 10 per cent of accounting professionals.^[1] In the 2009 Brain and Mind Research Institute study of around 2500 lawyers and law students, nearly 60 per cent reported moderate to very high levels of psychological distress.^[2] Using a different set of measures, a 2013 survey of nearly 1000 Australian legal practitioners found that between 26 and 32 per cent of participants reported experiencing moderate to extremely severe symptoms of depression, anxiety and stress,^[3] while 49 per cent of lawyers in a 2014 study reported stress symptoms.^[4] A number of studies of Australian law students have found comparably high levels of stress and mental health issues.^[5] There are, however, methodological problems with some of these studies.^[6] North American research has found conflicting results regarding the extent of stress and unhappiness among lawyers.^[7] Nevertheless, the presence of a sizeable proportion of lawyers suffering from depression or other mental illness is still a problem that cannot be ignored.

Some research has suggested that lawyers who practise in the public or community sector are ‘happier’,^[8] while other studies found no statistically significant differences in depression, anxiety and stress scores between lawyers working in different workplace types.^[9] The limited research that exists, however, suggests these professionals may face unique sources of stress. For example, a recent study of NSW community sector lawyers found that while only four per cent of lawyers work in the public or community sector, they are disproportionately represented in remote and disadvantaged areas where working conditions are more difficult.^[10] Public sector lawyers may also be more likely to be exposed to vulnerable clients and potentially traumatic materials, and therefore are more vulnerable to secondary traumatic stress or vicarious trauma.^[11]

Work stress among paralegals and administrative support staff has been a neglected area of research. In one of the most insightful ethnographic studies of American law firms, Jennifer Pierce describes how paralegals and legal secretaries (mostly female workers), in addition to having to meet often unscheduled and urgent demands for research and preparation work, must also perform ‘emotional labour’ by staying calm and deferential, providing emotional support for lawyers, reassuring witnesses and clients, and managing their own feelings of resentment.^[12] There is research evidence that suggests that emotional labour is ‘at least as important a source of job stress as are job conditions’.^[13]

B Understanding Work Stress

LaMontagne et al report that all models or theories of work stress postulate that stress is a process that begins with exposure to stressors, which can be physical or psychosocial.^[14] Such exposure can lead to perceived distress, which can result in adverse short-term responses or enduring health outcomes.^[15] These responses and outcomes may be psychological, behavioural or physiological.

The process is not linear and is affected by a range of social, psychological, biophysical, behavioural and genetic factors. Three theoretical frameworks are useful for understanding stressors in work: the demand‑–control model (‘DCM’),^[16] the effort–reward imbalance (‘ERI’),^[17] and the organisational justice model.^[18] Briefly, the DCM postulates that job strain arises from situations where job demands are high but job control is low. The ERI model hypothesises that jobs involving a high level of effort but a low degree of reward (eg, financial, status, advancement) are stressful. Finally, the organisational justice model suggests

that negative perceptions of procedural justice (eg, decision-making in an organisation) or relational justice (eg, respect from a supervisor) can be a form of stressor. These models have all found support in empirical research examining various dimensions of work stress with mental health outcomes such as depression, anxiety, suicide, burnout and emotional distress.^[19]

An important issue for understanding work stress is the extent to which the perceptions of working conditions can be equated to objective working conditions. LaMontagne et al report a ‘strong relationship between worker perceptions and objective working conditions’ but acknowledge that this conclusion is based on the ‘limited’ evidence available at the time of writing.^[20] Both Pearlin et al’s stress process model^[21] and Lazarus and Folkman’s relational model^[22] suggest that the same objective working conditions can lead to different responses in different individuals. Although both models emphasise the mediating/moderating effects of coping efforts, the former also highlights the importance of social supports,^[23] while the latter gives prominence to the cognitive appraisal process.^[24]

These models are all useful for understanding lawyering stress, although very few have been tested. An important study is Wallace’s use of Karasek’s demand–control model to predict depression and work-to-family conflict^[25] for married lawyers working full-time in Alberta, Canada. Wallace found that Karasek’s model applied to depression as well as work-to-family conflict, ie, job demands contribute significantly to both depression and work-to-family conflict, while job control reduces them.^[26] She also found that different kinds of social support had different effects on depression and work-to-family conflict. For example, while organisational support reduces depression and work-to-family conflict, co-worker support produces some unexpected effects: it effectively buffers the negative effects of overload and extra professional work activities, but accentuates the negative effects of long work hours. Wallace’s research suggests that ‘not all support strategies are beneficial and instead, some may be detrimental to worker well-being’. The possibility of a ‘stress-transfer effect’ implies that where job demands are excessive, mutual support among workers is likely to amplify the negative effects of these demands rather than reduce them.^[27]

In light of these findings, a number of scholars have sought to explain the high incidence of mental health and wellbeing issues in the legal profession. It has been suggested that this incidence is, at least in part, a result of the significant stress and pressure experienced by many legal professionals in the workplace.^[28] For example, Chan has developed a model of work stress which suggests that difficult objective working conditions contribute to stress, but their effect can be mediated by both organisational culture and individual workers’ characteristics. In particular, where an individual perceives their work demands as outstripping their ability to cope with these demands, this is considered to lead to stress outcomes. Such a perception will be influenced by personal and organisational resources, including social support and coping strategies.^[29] The role of working conditions in contributing to stress is supported by an Australian study of

long hours, management issues, billable hour requirements, and a lack of mentoring and supervision for new lawyers.^[30] Similarly, the 2013 survey found that excessive job demands, minimal control over workload and the spill over of work commitments into personal life were significantly correlated with poorer mental health outcomes.^[31]

C Wellbeing Interventions

In response to direct experience and research findings on mental health and wellbeing issues in the legal profession, professional bodies, law schools, firms and other organisations in Australia have developed a range of initiatives. The emphasis broadly has been on three areas: first, raising awareness of, and reducing stigma related to, mental health and wellbeing issues; second, equipping students and lawyers with the skills and resilience necessary to manage stressors and ensure their wellbeing; and third, identifying and addressing underlying structural or cultural issues contributing to the problem, such as billable target pressures, inflexible working arrangements, inadequate management systems, and a lack of response to bullying and harassment.^[32] For example, five major national law firms and the College of Law have collaborated to establish the Resilience@Law initiative, an educational module which raises awareness and provides strategies and resources for dealing with mental health concerns.^[33] A number of law schools, professional associations and workplaces have also introduced initiatives such as counselling services, mindfulness and exercise classes, and training on mental health issues.^[34] A set of best practice Guidelines for workplace wellbeing was developed by the Tristan Jepson Memorial Foundation to ‘assist legal organisations to create workplaces that fulfil

each of the workplace factors, identified by extensive research as critical to psychological health’;^[35] numerous legal organisations, including courts, law societies and bar associations, legal education institutions, government, community and private legal service organisations have become signatories to the Guidelines. The competency standards for entry-level lawyers have also been amended to include the requirements for practical legal training providers to provide applicants with information regarding mental health issues and access to support for developing resilience and wellbeing.^[36]

The effectiveness of these initiatives has rarely been assessed and the few studies that have considered this issue have also generally been unable to establish whether there is a causal link between accessing certain initiatives and improved mental health and wellbeing outcomes. For example, while the Beaton 2011 annual Australian Business and Professions Study found that lawyers ‘reported the highest levels of mental health training’, and were also less likely than in the 2007 survey to have stigmatising views about depression,^[37] it could not be shown that this was necessarily a result of increased participation in training. In fact, the survey found that professionals ‘felt their organisation was not well equipped to manage mental health issues in the workplace effectively’,^[38] and lawyers in particular felt that their organisation was ‘less likely to actively help an individual to seek treatment’.^[39] Another example of a somewhat more comprehensive study on the implementation of mental health and wellbeing initiatives is a 2013 survey of Australian lawyers, in which participants were asked to rate the effectiveness of a series of initiatives for dealing with work-related stress.^[40] A significant percentage of participants rated the following as highly or quite effective: having time off work (76 per cent), sport and exercise classes (74 per cent), redistribution of work to other colleagues (70 per cent), extra time to complete work (70 per cent) and mentoring programs (60 per cent).^[41] Meanwhile, less than half the participants rated as effective time management training (44 per cent), health check-ups (41 per cent) and an Employee Assistance Program (‘EAP’) (30 per cent).^[42]

The findings of this 2013 study align, to some extent, with the broader health literature on the use of interventions to respond to work stress and mental health issues. This literature distinguishes between three levels of intervention. First, primary interventions, which are proactive initiatives intended to eliminate or reduce job stressors in the organisation or work environment (eg, reducing job demands, improving job control and improving social support). Second, secondary interventions, which are ameliorative efforts designed to alter and improve the ways individuals perceive and respond to stressors (eg, screening for stress symptoms, training programs for relaxation and anger management and cognitive behavioural therapy). Third, tertiary interventions, which are reactive efforts intended to treat, compensate and rehabilitate workers with stress-related illness or mental health issues (eg, medical care, counselling and modification of job stressors in return-to-work plans).^[43]

In general, the literature advocates a systems approach that integrates all three levels of intervention.^[44] However, LaMontagne et al have observed that primary interventions are most likely to be effective, while tertiary interventions are least likely to be, because ‘the further upstream one is from an adverse health outcome, the greater the prevention effectiveness’.^[45] This is somewhat consistent with the 2013 study, in which primary interventions such as leave allowances and reduction of workloads and time pressure were rated as most effective, along with some secondary interventions such as sport and exercise classes and mentoring programs, while tertiary interventions such as the EAP were rated as less effective.^[46] In contrast, however, organisations may often prefer to implement secondary or tertiary interventions, due to cost and resourcing considerations. The significance of primary, secondary and tertiary interventions is considered further below by reference to the research findings.

III THE STUDY

This article reports the findings of the first Australian study to rigorously investigate the drivers of work stress amongst public sector legal professionals and support staff, and the conditions under which workplace initiatives can reduce stress and improve wellbeing. Participants – over half of whom were lawyers and one quarter of whom were legal support officers^[47] – were drawn from a large public sector legal organisation. This organisation employs staff across metropolitan and regional offices, and provides a range of criminal, family and civil legal services at each jurisdictional level, with a particular focus on criminal law. The organisation introduced a number of primary, secondary and tertiary interventions intended to support the mental health and wellbeing of its employees prior to, and during the course of, the study.

A Research Design

1 Theoretical Framework

The research design is informed by the general explanatory model of work stress constructed by Chan’s synthesis of the literature on work stress.^[48] A schematic representation of this model is shown in Figure 1. At the left-hand side of the model is a group of variables labeled as ‘objective working conditions’; these are the stressors in the work environment. On the far right-hand side of the model is a group of indicators measuring ‘stress/wellbeing’, which are individual or organisational responses/outcomes or manifestations of stress. At the top of the figure is the factor ‘occupational or organisational culture’, which includes various assumptions, values and beliefs shared by the occupation or workplace. At the bottom of the figure is a group of variables pertaining to individual, psychological, physical and social characteristics. It is hypothesised that difficult objective working conditions will contribute to stress, but the effect of working conditions on stress can be mediated or moderated by both organisational culture and individual workers’ characteristics through cognitive appraisal processes which include both an appraisal of work demands and an appraisal of the person’s ability to cope with the demands. These appraisals will take account of personal and organisational resources such as social support and coping strategies. It is hypothesised that it is the perception that work demands are outstripping the worker’s ability to cope that will lead to stress outcomes.^[49]

When conceptualised this way, primary wellbeing interventions are those directed mainly at changing objective working conditions or aspects of occupational and workplace culture, secondary interventions are those designed for improving the individual’s level of psychological, physical or social resilience, while tertiary interventions are those aimed at directly relieving stress or increasing wellbeing.

2 Research Methods

The study was conducted from 2011 to 2015,^[50] and employed a quasi-experimental design to evaluate the effectiveness of wellbeing initiatives. Central to this design is three waves of staff surveys: the first collected baseline data, prior to the commencement of the project and the introduction of several wellbeing initiatives (‘Baseline’), while the second (‘FU1’) and third (‘FU2’) repeated the questions from the Baseline to assess the impact of these initiatives after 6 and 28 months of implementation. Participants were also encouraged to provide additional comments in free-form text.

To supplement the surveys and contextualise their findings, additional data was also collected in three ways. First, interviews were conducted with nine key management and human resource development personnel on characteristics of the organisation, including its management structure and workload, and the designs and objectives of the workplace initiatives introduced to reduce stress and improve wellbeing. These interviews were used mainly to provide background information (see Part III(B)) and assist with the design of follow-up surveys. Second, semi-structured interviews were conducted in person or by phone with a stratified random sample of lawyers and support staff. A total of 324 invitations were sent to staff; interviews were conducted with 44 lawyers and 25 support staff, including 10 volunteers (an average response rate of 18 per cent). These interviews were also analysed thematically in the same way as comments provided in the surveys. Finally, researchers also conducted fieldwork observing lawyers and support staff in their daily work. A total of 44 observation episodes were conducted, totalling over 250 hours, at various office locations and involving all areas of legal service.

All qualitative data was analysed thematically to provide personal and contextualised accounts of participants’ experiences with, and perceptions of, the wellbeing initiatives (see Part V). The analysis was informed by the theoretical framework and the pre-developed themes related to this framework.^[51]

3 The Survey Instruments

The online questionnaire included questions on participants’ awareness of available health and wellbeing initiatives, and their views on the effectiveness of initiatives they had accessed.^[52] The survey was also designed to measure key psychosocial risk factors for job stress.

To measure participants’ perceptions of work demands relative to their ability to cope with the demands (see Figure 1), the short version of the Effort Reward Imbalance Questionnaire was used.^[53] This questionnaire is based on the Effort Reward Imbalance (‘ERI’) model, a theoretical model of job stress which maintains that people who expend high effort in their work (eg, work demand, workload) but who receive few rewards (eg, lack of promotion prospects, job insecurity) will experience recurrent negative emotions and sustained stress responses and therefore be at greater risk of developing adverse health outcomes, such as anxiety and depression. Furthermore, people who exhibit a specific cognitive and motivational pattern of coping with work demands, known as ‘Overcommitment’, will be more likely to experience poorer health outcomes (even when ERI is absent) because this coping style prevents them from accurately assessing their psychosocial working conditions.^[54] The short-version of the ERI Questionnaire used in the current study asks respondents to indicate on a four-point Likert scale the extent to which they agree or disagree with 16 different statements. The ratio between effort and reward (weighted by item numbers) is calculated to quantify the degree of mismatch or imbalance. A value greater than 1.0 reflects higher risk for stress related responses.

Aspects of organisational culture (see Figure 1) were assessed in the survey using the ‘quality of leadership’, ‘social support from supervisors’ and ‘social support from colleagues’ composite scales drawn from the Copenhagen Psychosocial Questionnaire (COPSOQ).^[55] The Managerial/Collegial Support composite scores include ratings on the extent to which respondents believed that their manager/colleagues listened to them, the extent to which respondents believed their manager/colleagues provided help and support, and the extent to which their manager/colleagues provided feedback on their performance. The leadership composite score includes ratings on the extent to which the respondent’s immediate supervisor provided good development opportunities, gave high priority to job satisfaction, was good at work planning and good at solving conflict. Several of the interventions introduced by the organisation focused on improving managerial capabilities; measurement of this moderating factor was therefore essential.

The final section of the survey contained questions referring to personal mental health and wellbeing (see Figure 1). A standardised measure, known as the DASS-21 (Depression Anxiety Stress Scales), was included to assess mental health.^[56] The DASS-21 is a well-validated and reliable instrument designed to measure dimensional levels on three aspects of psychological distress: depression, anxiety and tension/stress. Wellbeing was assessed through 15 questions asking respondents to rate their level of satisfaction with 15 different aspects of their work role,^[57] and to give an overall satisfaction rating with their decision to work for their organisation. These measures were repeated in the follow-up surveys to identify any organisation-level changes over time.

Figure 1 also documents the importance of individual psychological, physical and social characteristics in explaining work stress. Whilst measuring all relevant individual factors was beyond the scope of the current work, the surveys captured essential demographic information including age, gender, dependent children, living arrangements and Aboriginality, as well as information regarding the respondent’s current work role and history with the organisation. Questions on respondents’ histories of depressive or anxiety-related symptoms and help-seeking behaviours were also included in the questionnaire. These factors were taken into account when determining the contribution of the above moderating or mediating factors to experiences of work stress and when assessing changes in mental health and wellbeing over time.

4 Response Rates and Limitations

Baseline survey in 2011; 374 participated. Following the introduction of new wellbeing initiatives, 205 participants completed FU1 in mid-2013, and 300 participants completed FU2 in early 2015. Although the response rates were quite low relative to the number of staff invited to participate,^[58] they are comparable with other online surveys.^[59] Furthermore, a comparison of the participants’ characteristics with equivalent information on the organisation’s staff suggests that the samples are generally representative of the organisation as a whole. Any systematic differences in the demographic profile of survey respondents across the three waves of the survey were taken into account in the multivariate analysis presented later in this article.

When interpreting the results from this study there are limitations that should be considered. The quasi-experimental design enables us to examine organisation-level changes in mental health and wellbeing indicators soon after the introduction of the new wellbeing initiatives. However, like other quasi-experimental designs implemented in complex field settings,^[60] this study was unable to control extraneous variables that had the potential to influence our outcome measures and which threaten the validity of causal inferences. Of particular relevance to the current study is the introduction of new legislation and regulations during the follow-up period which reformed employment conditions and management structure throughout the public sector. While this limits our ability to make definitive causal statements about program effects, this research, being the first of its kind in Australia, still provides valuable insights into drivers of work stress amongst public sector lawyers and support staff and the necessary conditions for effective intervention.

B Health and Wellbeing Initiatives

The subject organisation introduced a range of initiatives targeting the mental health and wellbeing of its staff, both prior to and over the course of the study. Many of these initiatives were developed by employee committees created by the organisation, and following reviews intended to assist in identifying the causes of workplace stress and mental health issues. According to key informants, the main objectives in introducing a range of initiatives were to build awareness of mental health and wellbeing, reduce stress, facilitate access to resources and support, promote self-care and build resilience.

The initiatives provided by the organisation consisted of a broad range of primary, secondary and tertiary interventions. The primary interventions

included provisions for flexible working practices (eg, starting work late or leaving work early, and working from home); family and community services leave; and the rotation of staff working in ‘high risk’ areas.^[61] In respect of secondary interventions, health check-up sessions were introduced following the 2011 Baseline survey. These sessions enabled staff to meet with a psychologist to discuss their mental health, and gain insight into and assistance in managing the compounding factors that contribute to stress or psychological injury. They are targeted particularly at those who work in ‘high risk’ areas. Key informants described the main objectives of the sessions as being to improve the self-awareness of staff of their own mental health and build resilience.

Other secondary interventions introduced following the 2011 Baseline survey included a series of workshops on issues such as building resilience, managing stress and psychological injury, improving wellbeing and conflict resolution. These workshops were aimed at increasing awareness and resilience and capacity for self-care. Workshops were also conducted for managers on a range of management issues, including when to organise debriefing sessions for staff members or refer them to psychological services. Key informants explained that the purposes of this training were to address interpersonal conflict, clarify the role of managers, support managers in having difficult conversations with their staff, build a respectful workplace and send a message that bullying would not be tolerated. The management workshops operated alongside a support network and helpline for managers.

The organisation provided for several other secondary interventions, including peer support programs, requirements for regular staff meetings, promotion of healthy eating and quit-smoking programs, discounted gym memberships, flu vaccination programs and mental health- and wellbeing-dedicated pages on the intranet. In respect of tertiary interventions, the organisation provided for an EAP, which enables staff to access professional counselling services over the phone or in person. The organisation also provided debriefing sessions aimed at staff who regularly viewed distressing material at work and following critical incidents.

IV DRIVERS OF WORK STRESS AMONGST LAWYERS AND LEGAL SUPPORT STAFF

To understand the working conditions within this organisation and to explore the drivers of stress, anxiety and depression amongst lawyers and legal support staff, a series of nested linear regressions were performed. These regression models were based on responses provided in the Baseline survey only and included as the dependent variable the overall DASS score (modelled as a continuous variable), which combined scores from the three DASS subscales to produce a composite measure of negative state.^[62],[63] These analyses were designed to test the extent to which various factors (including manager and colleague support) independently explain variation in overall DASS scores. Variables were added sequentially to the regression model in blocks as follows:

The p-value associated with the addition of each block is reported along with an estimate of the change in proportion of variation explained by the block (R-squared change). Categorical covariates were operationalised as:

A Overall Drivers of Work Stress

Table 1 summarises the results of the nested linear regressions modelling the square root of overall DASS scores of respondents at Baseline. As seen here, various work-related factors are significant independent predictors of negative state. Not surprisingly, the results from this model show that one of the biggest drivers of overall scores on the DASS is previous mental health history. Respondents who reported previously experiencing depression and anxiety-related symptoms scored much higher on the DASS than those with no prior history of these symptoms.

Variable	Beta(95%CI)*	p-value*	p-Block	Change in R-squared
DEMOGRAPHICS			0.107	3.6%
Gender (male)	-0.46 (-0.94-0.02)	0.062
Age (<30 years)
Age 30–9	0.54 (-0.10, 1.19)	0.099
Age 40–9	-0.40 (-1.09, 0.29)	0.252
Age 50–9	-0.17 (-0.87, 0.53)	0.636
Age 60+	-0.37 (-1.52, 0.79)	0.535
Has children (No)	0.05 (-0.41, 0.50)	0.838
Full-time employment (Part-time)	-0.49 (-1.06, 0.07)	0.084
Lawyer role (support staff)	-0.35 (-0.75, 0.04)	0.079
Previous depression/anxiety (No)			<0.001	13.9%
Previous depression	0.78 (0.09, 1.47)	0.027
Previous anxiety	0.60 (0.01, 1.18)	0.045
Previous depression & anxiety	1.47 (1.00, 1.95)	<0.001
Effort Reward Imbalance ratio (<1.0)			<0.001	9.1%
Ratio 1.0 to 1.10	0.47 (-0.16, 1.10)	0.146
Ratio >=1.11	0.63 (0.15, 1.11)	0.010
Overcommitment scale	0.30 (0.24, 0.36)	<0.001	<0.001	16.6%
Managerial Support	-0.01 (-0.02, -0.01)	0.047	0.007	1.2%
Collegial Support	-0.01 (-0.01, 0.01)	0.564	0.564	<0.1%

Block 1: Gender, age, children, employment status, role; Block 2 added Experience of anxiety and/or depression; Block 3 added Effort/Reward ratio categories; Block 4 added Overcommitment scale; Block 5 added Managerial Support scale; Block 6 added Collegial Support scale.

However, even after taking into account mental health history, demographic information and the type of work being undertaken, we still find a significant association between DASS scores and three important work-related variables: ERI ratio, Overcommitment and perceived levels of Managerial Support. In terms of ERI ratio, respondents who reported expending a very high level of effort in their work compared to the rewards they receive reported more negative symptomatology than other respondents. Respondents who reported high levels of Overcommitment also recorded higher DASS scores. In fact, scores on the Overcommitment scale were able to explain more of the variation in DASS scores than any other variable explored here, including prior history of depression and anxiety. Finally, respondents who reported a lower level of support from their immediate manager scored higher on the DASS than those who reported a high level of Managerial Support. Collegial Support, on the other hand, was not significantly associated with respondents’ overall scores on the DASS after taking into account the contribution of other relevant variables.

B Comparison between Lawyers and Support Staff

Table 2 and Table 3 present results from the nested linear regressions performed separately for the lawyer and support staff subsamples. Stratifying the survey sample in this way reveals some interesting differences between these two subgroups in terms of work-related factors that drive psychological distress.

Factors related to workload and rewards appear to be more important predictors of mental health amongst lawyers than amongst legal support staff. Overcommitment is a significant independent predictor of overall DASS scores for both subgroups, but the ERI ratio was a significant independent predictor only in the lawyer model. Further, scores on the ERI scale and Overcommitment scale accounted for a greater amount of variation in lawyer DASS scores compared with support staff DASS scores. Meanwhile, a lack of Managerial Support appears to be an important predictor of negative symptoms only in the case of legal support staff. For lawyers, once other demographic and work-related factors are taken into account, levels of Managerial and Collegial Support were not significantly associated with overall DASS scores.^[64]

Variable	Beta(95%CI)*	p-value*	p-Block	Change in R-squared
DEMOGRAPHICS			0.02	9.3%
Gender (male)	-0.21 (-0.80, 0.38)	0.474
Age (<30 years)
Age 30–9	0.65 (-0.12, 1.41)	0.097
Age 40–9	-0.15 (-1.02, 0.71)	0.726
Age 50–9	-0.18 (-1.05, 0.69)	0.682
Age 60+	-0.81 (-2.34, 0.71)	0.295
Has children (No)	0.32 (-0.25, 0.90)	0.267
Full-time employment (Part-time)	-0.49 (-1.30, 0.31)	0.229
Previous depression/anxiety (No)			<0.001	16.9%
Previous depression	0.80 (-0.01, 1.60)	0.051
Previous anxiety	0.92 (0.14, 1.69)	0.020
Previous depression & anxiety	1.60 (1.02, 2.18)	<0.001
Effort/reward ratio (Ratio <1.0)			<0.001	11.1%
Ratio 1.0 to 1.10	0.71 (-0.12, 1.54)	0.094
Ratio >=1.11	0.78 (0.14, 1.43)	0.018
Overcommitment scale	0.30 (0.22, 0.38)	<0.001	<0.001	16.2%
Manager support	-0.01 (-0.02, 0.01)	0.497	0.237	<0.1%
Colleague support	-0.01 (-0.02, 0.01)	0.479	0.479	<0.1%

Block 1: Gender, age, children, employment status, role; Block 2 added Experience of anxiety and/or depression; Block 3 added Effort/Reward ratio categories; Block 4 added Overcommitment scale; Block 5 added Manager Support scale; Block 6 added Colleague Support scale.

Table 3: Drivers of overall DASS scores at Baseline – SUPPORT STAFF ONLY (n=183)

Variable	Beta(95%CI)*	p-value*	p-Block	Change in R-squared
DEMOGRAPHICS			0.327	4.4%
Gender (male)	-0.82 (-1.64, -0.01)	0.049
Age (<30 years)
Age 30–9	0.30 (-0.81, 1.40)	0.598
Age 40–9	-0.69 (-1.82, 0.43)	0.225
Age 50–9	-0.11 (-1.29, 1.06)	0.850
Age 60+	-0.16 (-1.95, 1.63)	0.856
Has children (No)	-0.23 (-0.97, 0.52)	0.547
Full-time employment (Part-time)	-0.50 (-1.31, 0.31)	0.227
Previous depression/anxiety (No)			<0.001	11.9%
Previous depression	0.98 (-0.27, 2.24)	0.124
Previous anxiety	0.41 (-0.49, 1.30)	0.369
Previous depression & anxiety	1.32 (0.55, 2.09)	0.001
Effort/reward ratio (Ratio <1.0)			<0.001	7.0%
Ratio 1.0 to 1.10	0.27 (-0.73, 1.27)	0.595
Ratio >=1.11	0.50 (-0.23, 1.23)	0.175
Overcommitment scale	0.28 (0.18, 0.38)	<0.001	<0.001	13.7%
Manager support	-0.01 (-0.03, -0.01)	0.043	0.011	2.4%
Colleague support	-0.01 (-0.02, 0.01)	0.788	0.788	<0.1%

Block 1: Gender, age, children, employment status, role; Block 2 added Experience of anxiety and/or depression; Block 3 added Effort/Reward ratio categories; Block 4 added Overcommitment scale; Block 5 added Manager Support scale; Block 6 added Colleague Support scale.

These analyses confirm that the relationships between workplace/individual variables and mental health are complex: workload and workplace factors turned out to be very significant over and above personal characteristics and previous mental health history. Furthermore, these relationships were not exactly the same for all staff: for lawyers, the most significant drivers of overall DASS scores were Overcommitment and ERI; these factors were also important for support staff, but (the lack of) Managerial Support was an additional factor significant for this group.

Importantly, these results identify numerous avenues through which legal organisations could have a positive impact on the mental health of their employees while highlighting the importance of tailoring options to the particular work in which an individual is engaged. Primary interventions aimed at reducing workload and increasing perceived work benefits would appear most promising for effecting change across a legal organisation, but may be particularly beneficial to lawyers. On the other hand, significant changes to organisational culture through increased managerial support may serve to improve outcomes for legal support staff but may be less relevant for lawyers. Secondary interventions designed to increase individual resilience to stress and an individual’s perceived ability to cope with work demands would appear from our work to have universal applicability across an organisation.

V EFFECTIVENESS OF INITIATIVES

This Part discusses the research findings on the effectiveness of the mental health and wellbeing initiatives provided by the subject organisation. It examines participants’ awareness and usage of the initiatives, and the quantitative evidence for the impact of the introduction of new initiatives on the mental health of participants. It also considers participants’ views on the effectiveness of the initiatives and organisational factors necessary to support their mental health and wellbeing. In summary, primary interventions were positively received by participants, especially where they were directed at reducing workloads and other pressures, or improving peer/managerial support. Perceptions of secondary and tertiary interventions were more mixed. Further, concerns were expressed by many participants about their ability to access a range of initiatives due to organisational and cultural factors, including their heavy workloads and a risk of stigmatisation.

A Awareness and Usage of Initiatives

The survey results suggested there was a strong awareness amongst participants of many of the mental health and wellbeing initiatives available through the organisation. Participants reported a high level of awareness of a number of primary initiatives, including provisions for flexible working practices (94–6 per cent)^[65] and family and community services leave (95 per cent in Baseline). A significant percentage of participants also reported awareness of a number of mental health and wellbeing workshops: for example, 87 per cent of participants in the Baseline survey were aware of a workshop on demystifying mental illness, and 88–91 per cent of participants (in FU1–2) were aware of a building resilience workshop. Most participants were also aware of the EAP (89–95 per cent).

Despite these high levels of awareness, a much smaller proportion of survey participants reported accessing the initiatives offered through the organisation. Three primary initiatives were the most accessed by participants: family and community services leave (69 per cent in Baseline), flexible working practices (73–6 per cent), along with regular staff meetings (83 per cent in FU2). A relatively high percentage of participants also reported accessing workshops on building resilience (63–8 per cent in FU1–2) and demystifying mental illness (46 per cent in Baseline). In contrast, only a minority of participants had accessed the EAP (25–33 per cent) and the health and wellbeing pages on the intranet (34–7 per cent in FU1–2). An even smaller percentage of participants reported accessing the manager helpline (3–6 per cent), the health check-up sessions (11–21 per cent in FU1–2) and discounted gym memberships (1–2 per cent in FU1–2).

Participants in the survey and interviews provided a range of reasons for not accessing initiatives. In particular, several participants felt unable to utilise the initiatives offered due to their heavy workload and a lack of available time. A key concern was the time taken to access some initiatives, such as family and community services leave, flexible working practices, or workshop sessions. This resulted in a need to catch up on work that had not been completed during the time out, which could counterproductively increase participants’ stress levels:

We just don't have the time to do all those things. I don't have time to take flex days, I don't have time to come in late or go home early ... I know there was a wellbeing training we were supposed to do but I didn't ever have time to do it, because one of the other problems for us is that because we're in court so much [when] a lot of these things are on (INT, Metro, Lawyer).^[66]

The on-going difficulty in accessing any programs or initiatives is finding time to participate. The productivity expectations placed on solicitors are high and do not build in time for training or engagement in initiatives. Participation in such initiatives can mean that time spent in the activity or initiative needs to be made up and can increase the stress burden (FU2, Regional, Lawyer).

For some participants, the lack of available time to access initiatives was compounded by the times at which some initiatives, such as yoga and mental health and wellbeing workshops, were run. This issue was particularly acute for participants who regularly attend court:

For the people who are in the coalface at the court from 8:30 in the morning till 4:30 in the afternoon they can't come to yoga. They haven't got the time to take time out of the court regime to get all these resources that are available to them (INT, Metro, Lawyer).

Challenges in accessing initiatives were also noted especially by participants in regional offices, due both to the dearth of initiatives available in regional offices, and the difficulties involved in travelling to access initiatives available only in the city offices:

[The initiatives] are simply not available in a lot of regional offices and our work commitments mean they are just impossible to access. There is a huge discrepancy between what is available in head office, and what is available to the rest of us (FU1, Regional, Lawyer).

Working in a smaller regional office is always a challenge to attending any courses in HO, as it means that the office is staff down which impacts on other staff, workloads and efficiency of office (FU2, Regional, Support).

Outside of practical access issues, some participants cited the risk of stigmatisation as a deterrent to using the available mental health and wellbeing initiatives. This risk was attributed by some to a workplace culture that prioritises productivity, which means that an employee’s career may be disadvantaged if it becomes known they need mental health support:

There is a culture that exists that if you complain or indicate that you have been suffering from stress or depression then [your] career is over. This might not be the actual case but it is the perception that prevails (FU1, Regional, Lawyer).

I was going to [access one of the initiatives] at my office. However, a phone session was deemed by both myself and my manager to not be the best idea. This was because walls have ears. Also, if someone saw me in an office on the phone for some time, people would talk (FU2, Regional, Support).

These findings suggest that while legal professionals may be aware of wellbeing initiatives, and identify a personal need to engage with them, they may experience a range of both organisational and cultural barriers in doing so. This indicates the potential importance of organisations’ ensuring that structural factors – such as workloads, staff numbers and the location of initiatives – support the use of initiatives, while also seeking to positively change workplace culture and reduce stigmatisation associated with mental health issues and accessing wellbeing initiatives.

B The Impact of New Wellbeing Programs

Exploratory analyses^[67] were conducted to assess the impact of the organisation’s new wellbeing initiatives following the Baseline survey. These analyses were directed towards two questions: first, whether participation in the new initiatives improved mental health and workplace satisfaction; and second, whether there was an overall improvement in mental health and workplace satisfaction amongst staff after the new initiatives were launched. Note that there was no capacity to link participants’ pre- and post- measures of psychological distress. The study was designed in this way to ensure anonymity of participants’ responses. While this means that definitive causal attributions cannot be made from these analyses, the analysis still provides valuable insights into the overall impact of the interventions at the organisational level.

To answer our first question, differences in mental health and satisfaction by rates of program participation were examined. This investigates whether employees who participated in the new initiatives reported improved outcomes. One limitation of this approach is that employees voluntarily participated in these programs and thus any observed differences may be due to systematic differences between the groups. Therefore the impact of the new wellbeing programs was also examined using an ecological approach. Here we examine whether the introduction of multiple programs and other systemic changes has had an impact on the ‘average’ psychological and workplace culture outcomes.

1 Differences by Rates of Program Participation

Data from the three surveys were combined into one dataset and a variable was generated which indicated, by year, whether an individual had participated in at least one of three new initiatives: health check-up, workshop on building resilience and other health and wellbeing workshops (eg, on managing stress, improving wellbeing, conflict resolution and managing psychological injury). This variable separates the participants into five groups:

Linear regressions were run to compare these groups on overall DASS scores, scores on the DASS subscales (ie, stress, depression and anxiety)^[68] and overall levels of satisfaction with the organisation. Higher scores on the DASS measures indicate greater symptomatology and lower scores on the satisfaction measure indicate greater satisfaction. A series of planned contrasts^[69] was then used to separate out each group effect and joint effects of each year.

Table 4 summarises the results from the linear regression analyses exploring whether participation in the new initiatives led to an improvement in mental health and workplace satisfaction. Marginal means and significance values for the contrasts are reported here (after adjusting for gender, age, whether the person has children, whether they are full- or part-time, whether they are support or legal staff, years of service worked and whether the person had participated in the EAP). Contrasts that were statistically significant at the 0.1 level are marked by asterisks.^[70]

There was a significant joint effect (ie, all post-implementation groups scores pooled) on the DASS stress subscale and overall DASS scores. Respondents in the 2013 and 2015 survey waves had lower overall DASS scores, on average, than respondents in the 2011 Baseline survey. Furthermore, survey respondents in the latter two survey waves reported significantly fewer symptoms of stress compared with respondents in the Baseline sample. There were no significant differences between the post-implementation samples and the Baseline in terms of overall levels of satisfaction with the organisation.

When we examine individual survey year effects, the biggest difference between the post-implementation and Baseline samples with regard to mental health outcomes was in 2015. The contrasts shown suggest that 2015 survey participants who reported participating in a health and wellbeing program had significantly lower scores on all three DASS subscales, as well as significantly lower DASS scores overall, compared with the Baseline sample. 2015 survey participants who had not participated in a health and wellbeing program also had lower scores on the stress and anxiety subscales than the baseline sample, but these differences were not statistically significant at the 0.1 level.

Table 4: Significance and adjusted parameter estimates of joint effects and contrasts for program participation and year of survey for measures of mental health and satisfaction with organisation

Effect tested	Specific effect	Stress	Depression	Anxiety	Overall DASS	Satisfaction
		n=819	n=819	n=819	n=819	n=804
Pre-programs vs post¹	Joint effect 2013+2015 (vs 2011)	9.54 vs 10.97	6.19 vs 7.33	4.38 vs 5.35	20.1 vs 23.65	2.43 vs 2.32
Pre-programs vs post¹	Joint effect 2013+2015 (vs 2011)	p=0.030**	p=0.181	p=0.350	p=0.091*	p=0.127
	2013 No program vs 2011	7.58 vs 10.97	5.63 vs 7.33	4.19 vs 5.35	17.41 vs 23.65	2.21 vs 2.32
	2013 No program vs 2011	p=0.013**	p=0.223	p=0.291	p=0.07*	p=0.653
	2013 Program vs 2011	10.56 vs 10.97	6.66 vs 7.33	4.61 vs 5.35	21.81 vs 23.65	2.66 vs 2.32
	2013 Program vs 2011	p=0.61	p=0.424	p=0.257	p=0.375	p=0.023**
	2015 No program vs 2011	9.7 vs 10.97	7.23 vs 7.33	4.32 vs 5.35	21.24 vs 23.65	2.48 vs 2.32
	2015 No program vs 2011	p=0.23	p=0.926	p=0.230	p=0.375	p=0.411
	2015 Program vs 2011	9.15 vs 10.97	5.59 vs 7.33	4.28 vs 5.35	19.02 vs 23.65	2.29 vs 2.32
	2015 Program vs 2011	p=0.014**	p=0.021**	p=0.070*	p=0.013**	p=0.814

1 Adjusted for: male (reference)/female; age category <30 years (reference), 30–9, 40–9, 50–9, 60 plus; no children (reference)/has children; part-time work (reference)/ full-time work; support (reference)/lawyer; years of service five years or less (reference), 6–10 years, 11–20 years, more than 20 years; not participated in EAP (reference)/participated in EAP.

The comparison of outcomes for the 2013 sample with the Baseline was less conclusive. The results suggest that 2013 survey respondents who did not participate in a program had lower overall DASS scores and scores on the DASS stress subscale than the Baseline. This result may reflect selection bias (eg, respondents less concerned with their mental health are less likely to participate in a program). A more rigorous repeated measures study design would be needed to disentangle this effect. There was also evidence that 2013 survey respondents who did participate in a program were less satisfied with the organisation than the Baseline sample, an unexpected finding which is discussed in more detail below.

2 Organisation-Wide Changes Post-implementation

The second set of analyses were ‘ecological’ analyses seeing whether prevalence of distress and workplace culture improved once the initiatives were in place. An ecological analysis is one in which the relationship between a risk factor and an outcome is examined at the population level rather than the level of individual change; in this case the analysis examined changes in the provision of programs, changes in workplace practice and training of managers.

Table 5 summarises the results from the linear regression analyses exploring whether there was an overall improvement in mental health and workplace satisfaction of employees after the new health and wellbeing initiatives were launched during the 2011/12 and 2012/13 financial years. Presented in this table are the regression results for five separate models in which overall DASS scores, the three DASS subscales and overall levels of satisfaction with the organisation are defined as the dependent variables. Marginal means and significance values are shown for the planned contrasts and indicated by an asterisk if statistically significant at the 0.1 level. Consistent with the analyses described previously, there appears to be improvements in the mental health of participants, particularly in the last wave of the survey (FU2). The Baseline survey respondents reported higher overall DASS scores and higher stress scores than respondents in the 2015 survey wave. These differences in reported stress levels and negative symptomatology across survey waves were, however, relatively small.

There was also evidence of a difference between the 2013 and 2011 survey samples with regard to overall levels of satisfaction with the organisation. Again, the difference was small, but the higher value on this outcome suggests that the 2013 survey respondents reported being less satisfied than the 2011 survey respondents. This result is inconsistent with what was found for the DASS measures but may reflect general uncertainties about organisational change and job security during the 2013 wave resulting from broader public sector wide changes to employment conditions and management structures which were implemented around this time.^[71]

Table 5: Significance and adjusted parameter estimates of joint effects and contrasts for year of survey for measures of mental health and satisfaction with organisation

Effect tested	Specific effect	Stress	Depression	Anxiety	Overall DASS	Satisfaction
		n=819	n=819	n=819	n=819	n=804
Culture change¹	Joint effect 2013+2015 vs 2011	9.75 vs 10.79	6.34 vs 7.16	4.57 vs 5.32	20.56 vs 23.28	2.44 vs 2.29
Culture change¹	Joint effect 2013+2015 vs 2011	p=0.171	p= 0.349	p=0.288	p=0.193	p=0.237
	2013 vs 2011	10.08 vs 10.79	6.56 vs 7.16	4.63 vs 5.32	21.27 vs 23.28	2.54 vs 2.29
	2013 vs 2011	p=0.344	p=0.422	p=0.257	p=0.29	p=0.069*
	2015 vs 2011	9.53 vs 10.79	6.19 vs 7.16	4.52 vs 5.32	20.24 vs 23.28	2.37 vs 2.29
	2015 vs 2011	p=0.062*	p=0.153	p=0.145	p=0.076*	p=0.538

1 Adjusted for: male (reference)/female; age category <30 years (reference), 30–9, 40–9, 50–9, 60 plus; no children (reference)/has children; part-time work (reference)/full-time work; support (reference)/lawyer; years of service five years or less (reference), 6–10 years, 11–20 years, more than 20 years; no term for program participation.

Table 6 summarises the results from the linear regression exploring changes in interpersonal relationships and leadership as measured by the three COPSOQ composite scales – quality of leadership, Managerial Support and Collegial Support.^[72] Note that high scores correspond to higher frequency of the respective behaviours. The results in Table 6 show significant improvements in two of these measures over time. The biggest effects were found for the quality of leadership and Collegial Support outcomes, and the largest difference in these outcome measures was observed in the last wave of the survey. For the measures of Managerial Support, however, there was no evidence of significant differences over time.

Table 6: Significance and adjusted parameter estimates of joint effects and contrasts for year of survey for COPSOQ leadership, manager and colleague support scales

Effect tested	Specific effect	COPSOQ Leadership	COPSOQ Manager	COPSOQ Colleagues
		n=896	n=909	n=910
Culture change*	Joint effect 2013+2015 vs 2011	55.33 vs 50.06	62.41 vs 60.55	67.42 vs 64.55
Culture change*	Joint effect 2013+2015 vs 2011	p=0.016**	p= 0.368	p=0.018**
	2013 vs 2011	53.30 vs 50.06	61.0 vs 60.55	64.93 vs 64.55
	2013 vs 2011	p=0.208	p=0.846	p=0.838
	2015 vs 2011	56.70 vs 50.06	63.37 vs 60.55	69.12 vs 64.55
	2015 vs 2011	p=0.004**	p=0.174	p=0.008**

1 Adjusted for: male (reference)/female; age category <30 years (reference), 30–9, 40–9, 50–9, 60 plus; no children (reference)/has children; part-time work (reference)/full-time work; support (reference)/lawyer; years of service five years or less (reference), 6–10 years, 11–20 years, more than 20 years; no term for program participation.

In summary, the multivariate analyses suggest some improvements in mental health and stress amongst employees over the three waves of survey. These improvements were generally confined to 2015 survey respondents and were particularly evident for those who had participated in a health and wellbeing program in that survey year. The effects were, however, relatively small, and in most cases only significant at the 0.1 level. The analyses exploring potential mechanisms by which these changes occurred suggest that program participation may go some way towards explaining the observed effects but the inconsistent results across waves raise some doubt about this potential explanation. Where there were clear changes over time was with regard to employee perceptions of managerial leadership and collegial support. Survey participants reported higher leadership quality and greater collegial support in the 2015 survey wave compared with the 2011 survey wave. There were, however, no changes in perceptions of managerial support over time, a factor that was found in Part IV to be a significant driver of stress, at least for legal support staff.

While these analyses revealed relatively small intervention effects, it is possible that particular employees experienced much larger improvements in their health and wellbeing but by analysing outcomes at the aggregate-level, treatment effects were diluted. Linkage of individual records across the three surveys would have permitted further exploration of the characteristics of employees for whom the initiatives were most effective. It is also possible that the DASS is not sensitive to subtle short-term improvements in mental health and wellbeing. Ideally these results would be replicated with an additional follow-up survey to observe whether any longer-term gains were achieved post reforms. Finally, the minimal evidence for improved managerial support, along with respondents’ reports of limited opportunities to access some of the initiatives, may indicate broader implementation issues which impeded the success of the reforms. These implementation issues are explored in greater detail in the following section.^[73]

C Users’ Perceptions of Program Effectiveness

This section provides an overview of survey and interview participants’ perceptions of the effectiveness of wellbeing initiatives. Consistent with evidence that law firms, schools and organisations are making increased efforts to better respond to mental health and wellbeing issues, many participants observed that the organisation had, in recent years, considerably improved in this area.

Comments from survey participants and interviewees raised a range of additional issues.^[74] While the organisation was perceived as trying to improve support for staff, it was still considered to have ‘a long way to go’. Critics saw the organisation as providing only surface-level support, motivated by an interest in improving its own image rather than a meaningful commitment to the mental health and wellbeing of its staff. The initiatives were described as ‘minor’, ‘superficial’ and ‘tokenistic’, and seen as doing little to make a difference in participants’ lives. Even where participants approved of the initiatives as introduced, they suggested that effective implementation had been hindered by poor decision-making and a lack of support on the part of Human Resources (‘HR’) and managers. This was connected with a view that the initiatives had not been fully integrated into the day-to-day activities of staff, potentially due to an absence of structural changes to the organisation, and thus had to be actively sought out by staff.

1 Primary Interventions

A large majority of survey participants identified as effective the two primary interventions assessed: flexible working practices (91 per cent at Baseline) and family and community services leave (94 per cent at Baseline). These findings were supported by the qualitative responses of survey and interview participants, who identified the use of flexible working conditions as important to maintaining a work/life balance, thereby supporting their wellbeing. This was particularly the case for participants with family and childcare responsibilities. Flexible working conditions discussed with approval included the ability to work from home, take flex leave, and start or finish work at a more convenient time (eg, leaving at 3.00pm to pick up children):

Flexible work support remains the most important contribution to my health and wellbeing as a working mother (FU2, Metro, Lawyer).

The flexible work practices in particular assist with managing stress and ensuring a good work life balance. The use of flex time allows you to take an effective ‘mental health’ day if you need it ... Also the ability to commence earlier to finish earlier or vice versa also really helps you feel like you have control over your work hours which coming from the private sector is a completely different and empowering feeling (FU2, Metro, Lawyer).

While participants perceived flexible working practices as supporting their mental health and wellbeing, there were challenges in engaging in these practices. For example, there was a perception that management were critical of staff taking time off. Similarly, it was considered too difficult to engage in flexible working practices because of their high workload, understaffing issues, or a need to be consistently present in the office or court:

[It’s] difficult to take flex leave rostered without penalty. Management forgets that this is time which is already worked, but the priorities of workload and/or activities prevent staff from taking this leave even when entitled to do so. In effect the outcome is lots of unpaid overtime is worked (FU2, Regional, Support).

I mean it's so much effort just to go on leave. Even just to take one day out of the office. Sometimes it feels like what’s the point. I have to work so hard to go on leave and then I know that as soon as I get back it's just going to be all of this stuff; that happened while I was away (INT, Regional, Lawyer).

2 Secondary Interventions

Participants’ perceptions of the effectiveness of the secondary interventions offered by the organisation were more mixed. This reflects in part the number and variety of secondary interventions, which range from mental health and wellbeing workshops to discounted gym memberships. In the survey, several mental health and wellbeing workshops were rated as effective by a significant percentage of participants, including training on demystifying mental illness (85 per cent in Baseline) and building resilience (73–5 per cent in FU1 and FU2).

Further insights into these statistics are provided by the qualitative responses in the survey and interviews. Participants’ perceptions appeared to vary largely depending on the individual workshop they had attended. Participants were highly positive about training on mental illness where they were provided with real, practical strategies they could implement to help them manage stress on a day-to-day basis:

[The speaker] talked to us once about mental illness and it was great ... She was very down to earth and pragmatic and she had real responses to issues ... (INT, Metro, Lawyer)

The building resilience workshop helped me to understand more about stress and vicarious trauma and gave me some strategies to help cope when work is difficult or upsetting (FU1, Regional, Lawyer).

In contrast, participants reported negative experiences with workshops where they felt the information and advice given was impractical, obvious or overly simplistic. Participants also criticised training that was too general or insufficiently targeted to their organisation, practice group or role. Further, there was a view that the organisation itself did not support the implementation of some of the strategies promoted in the workshops. This lack of structural support negated the benefits of the workshops for these participants:

I thought I benefited from the building resilience conflict resolution workshops but apparently [the organisation] does not abide by what we learnt in those courses. They simply make us attend these courses so they look good but completely dismiss their contents (FU2, Regional, Support).

In contrast to the overall positive perception of the wellbeing workshops, only 49 per cent of participants in FU2 who had accessed the health check-up rated the session as effective. This is supported by the qualitative findings: views were divided between those who considered it a positive opportunity to speak about their experiences in their role, and those who criticised the one-off nature of the initiative. It was suggested that, rather than being therapeutic, the health check-ups were a ‘tick the box’ exercise for management to be seen to be doing something.

Initiatives targeted at managers were also perceived as effective by a majority of participants accessing them. In FU2, the manager helpline was rated as effective by 65 per cent of participants (n=17), and the management development program by 61 per cent of participants (n=62).

3 Tertiary Interventions

The main tertiary intervention provided by the organisation was access to the EAP. This service was viewed positively by around two-thirds of participants who had accessed it (61–4 per cent), while 75 per cent of participants in the Baseline rated as effective debriefing sessions sourced through the EAP.^[75] Participants reported accessing the EAP in relation to a broad range of issues, including workplace bullying, problems with management and aggressive clients.

A range of views were expressed. Those who reported that accessing the EAP had a positive impact on their mental health and wellbeing appreciated having someone to talk to about stressful incidents and pressures experienced in the workplace and received useful strategies to help them manage their stress levels. The EAP was said to be particularly useful when unexpected, individual incidents arose that generated high stress for participants, especially as the service can be accessed by phone at any location. The EAP was also described as a useful stepping stone towards seeking further help from a private practitioner:

I did use them once ... The team leader was driving me nuts ... Yeah it did [help], not that [it] solved anything but it was just good to talk to somebody about it and she gave me a couple of strategies about how to approach the boss about it which I used (INT, Metro, Support).

Those who were more critical of their interactions with the EAP reported a range of experiences from feeling they did not ‘get a lot out’ of the counselling sessions to feeling worse after accessing the service. Criticisms included that counsellors were not able to give constructive advice or support tailored to individuals in the legal profession, and appeared unsympathetic towards participants’ issues:

I went to EAP yeah. I didn't get a lot out of that session, I think I had two sessions, and other than - yeah, I didn't get a lot out of it ... Yeah, well given that it was a specific work issue and that's our work, I didn't feel the person I saw had any particularly affinity with what legal work might be like (INT, Metro, Lawyer).

Specific reasons were provided by participants on why they, or their colleagues, had not accessed the EAP. Concerns were raised about the stigma surrounding mental health issues and the perceived lack of privacy. While participating in the EAP was meant to be confidential, there was a view that employees needed to inform their bosses when they had booked an appointment, and a perception that information told to counsellors could be communicated to management or HR. There was also a concern that phone conversations with counsellors could be easily overheard or interrupted in the office.

In summary, the research results thus suggest that primary initiatives, such as flexible working practices and family and community services leave, were considered by participants to be most effective in reducing their stress levels and supporting their mental health and wellbeing. Perceptions of secondary and tertiary interventions were more mixed, and often depended on whether the participant felt the information or advice they received was practical and could be implemented in the workplace.

D Recommendations for Improvements

Survey and interview participants were invited to provide recommendations for improving existing interventions, and introducing new interventions, to support their mental health and wellbeing. These comments are not meant to be representative of all participants but provide some useful additional insights on wellbeing initiatives. It should be mentioned that rather than improving on individual initiatives, the need for more fundamental, structural changes to the organisation and its culture was also advocated. These changes were seen as important to facilitate greater access to interventions, especially primary interventions, reduce stigma surrounding mental health and wellbeing issues and improve peer/managerial relationships.

1 Primary Interventions

The reduction of workload was seen as an important way to improve employees’ mental health and wellbeing. However, this was considered unlikely to occur because of funding constraints:

I think really what an employer can do to reduce stress is, I suppose, look at our workloads and see what they can do in terms of people that do have extreme workloads and allowing those people to have assistance with managing those workloads. But then, in the funding situation we’re in and they're telling us, sorry, you can't have [back-up staff] to assist with court and you should be trying to save money ... how do you do that? (INT, Metro, Lawyer)

While flexible working practices were generally praised, there was a strong view that organisational and cultural changes were necessary to increase the capacity of staff to engage in such practices. Some managers were reported as highly reluctant to enable staff to alter their working hours, take flex leave, or participate in work-share arrangements. In addition, the organisational culture was considered to be inconsistent with flexible working practices, for example scheduling meetings before 9.30am or after 3.30pm. There was therefore a need for more support and encouragement of flexible working practices amongst management and the organisation as a whole, especially for staff with parenting responsibilities. It was suggested that responsibility for approving flexible working practices should be transferred from individual managers to the organisation, to ensure greater consistency and prevent favouritism:

flex is great but managers and corporate culture doesn't reflect flexible working arrangements – our core hours are 9.30–3.30 but often people schedule meetings before/after then (FU2, Metro, Support).

Flexible work practices work better in some offices than in others. Attitude of local management makes a difference in encouraging or discouraging staff from utilising this initiative. I have worked in offices where I felt discouraged from applying for flex leave even though I had accumulated well in excess of contract hours (FU2, Regional, Lawyer).

Another key recommendation concerned improving interactions and creating more supportive relationships between peers, and between staff and management. These relationships were viewed as integral in assisting them in responding to stress and pressure and improving their mental health and wellbeing:

The biggest single thing that influences my feeling of well-being at work is having a supportive Manager/or even other colleague or mentor that I can talk to/get advice/support/brainstorm solutions to problems. The times where I have not had this have been the most stressful in my career (FU2, Regional, Lawyer).

Consistent with this emphasis on building relationships, participants also recommended the use of team building exercises to encourage the development of a more supportive workplace. It was considered important that these exercises involve both lawyers and support staff to assist the two groups in working better together. It was suggested that the organisation could better support such exercises by allowing offices to close for staff training or meetings or by facilitating more social gatherings in the workplace or with colleagues. Social gatherings were perceived as central to creating a supportive, positive working environment and allowing employees to de-stress.

2 Secondary Interventions

Participants from regional offices identified a need to offer more initiatives in, or easily accessible from, their offices. Examples included having training occur in regional offices or via audiovisual link, rather than flying employees to training in metro offices, and introducing yoga or meditation in the office. Alongside existing workshops and training, one recommendation was to have more training for management and supervisors on how to respond to work stress, anxiety and depression amongst their staff. Such training was considered important to reforming the organisational culture and leading to a more supportive environment for professionals who are experiencing mental health and wellbeing issues:

I don't believe there is a culture at [the organisation] that truly understands the consequences of stress and the resultant anxiety. When I have suffered anxiety (and [it was] diagnosed) the response I got from a number of managers was ‘come on buck up/pull yourself together’ – a bit like telling someone with depression to be happy. Managers need to have training to have a better understanding of depression and anxiety (FU2, Regional, Lawyer).

[M]anagers/supervisors need more training in bullying, mental health issues caused by work stress and how to deal better with staff and these issues (FU1, Metro, Support).

There was also a need for more training for all staff on addressing negative inter-staff relations, including bullying and discrimination, and the stigmatisation of mental illness. It was suggested that the training should be mandatory to ensure that staff who are not personally experiencing mental health or wellbeing issues are still equipped to support others:

Health and wellbeing of employees will improve if there were more workplace anti-bullying and anti-discrimination workshops (FU2, Regional, Lawyer).

Staff need to be properly trained not to stigmatise those who suffer from stress related ailments. The more people put things off so they don't get embarrassed by what their co-workers think, the worse they can get (BL, Metro, Support).

One suggestion was the introduction of a mentoring or buddy system, wherein employees are paired with more experienced employees who supervise their work, discuss their experiences and challenges, and provide them with feedback and advice. This was seen as a way to reduce stress, especially that experienced by newer lawyers and managers, and ameliorate the feeling that these individuals are currently being ‘thrown into the deep end’ with very little guidance, which generates high levels of stress and pressure. It was considered that the mentor would need to be someone other than the individual’s immediate supervisor to prevent the mentee from feeling concerned about receiving a negative performance review or losing their position. Participants perceived the formalisation of such a policy as important to ensure that employees knew they had someone to speak to who would have time for them:

I really do wish our organisation had a buddy system where for your first three months you're not left on your own ever. You're paired up with someone and you walk beside them like an apprenticeship or like a junior samurai, I don't know, but you need to have a bit of handholding in the early days. I remember when I joined I was given my files and I was told you know which way the court's in, off you go (INT, Metro, Lawyer).

I would like to have a mentor. As a manager I often need to talk things over with someone – who may not necessarily be my supervisor. I have found someone who I talk to informally, but perhaps a more formal arrangement could be good (FU2, Regional, Lawyer).

Participants also recommended a range of secondary interventions largely aimed at improving their day-to-day working life. Many of these interventions were focused on creating opportunities for exercise and meditation, including by providing facilities and space for staff to use when engaging in these activities.

3 Tertiary Interventions

Debriefing was seen as a potential way to reduce stress by allowing participants to express their feelings and frustrations in a supportive environment. One suggestion was to have debriefing with an independent professional, such as a staff social worker or counsellor. The implementation of formal debriefing was seen as important, despite the informal debriefing reported by some as already occurring, because it could be made compulsory, and thus capture staff who do not necessarily recognise that they are experiencing a mental health or wellness problem, or are concerned about stigma associated with seeking help:

What would work really well is if all staff working in the front line, particularly the lawyers, had to do regular debriefing sessions with a psychologist (FU2, Regional, Lawyer).

Structured debriefing sessions built into the work schedule, where staff can offload, debrief and get support as we deal with a range of clients with challenging behaviours (FU1, Regional, Support).

Another suggestion was that instead of in-house counsellors, the organisation could provide subsidies for external counselling sessions. This was in part due to concerns about the privacy of in-house counselling, and also the need for counselling more extensive than that provided by the EAP:

It would be more beneficial if staff could anonymously access subsidies for financial support to see their own psychologist/psychiatrist as this is a significant expense. I pay $185 for a one hour visit and it provides important health benefits beyond what I feel I could attain through EAP (FU2, Metro, Support).

In summary, while the recommendations made by participants varied considerably, there was a clear focus on the need for broader organisational changes to create greater opportunities for staff to access wellbeing initiatives, for example by reducing workloads and ensuring greater managerial support. The concern in particular was about increasing access to existing primary interventions, including flexible working practices, and creating programs, such as mentoring, training and team-building exercises, to improve staff/manager relationships. Other recommendations were directed at improving day-to-day office life, for example through exercise and meditation initiatives, and facilitating regular debriefing.

VI CONCLUSION

Recent years have seen a growth in the body of research demonstrating high levels of stress, anxiety and depression in the Australian legal profession, and a corresponding increase in initiatives introduced by professional organisations and workplaces intended to improve lawyers’ mental health and wellbeing. Despite this, relatively little research has been conducted to evaluate the outcome or effectiveness of these initiatives.

This article responds to this gap in the literature by reporting the results of a quasi-experimental study on the effectiveness of wellbeing initiatives provided by a large public sector legal organisation in Australia. An initial examination of drivers of work stress at Baseline confirmed that although the relationship between the work environment and mental health is complex, there are various avenues for legal organisations to help improve employee wellbeing; initiatives focusing on workload and benefits, organisational culture and individual coping styles were identified as the most promising. Exploratory analyses of the longitudinal survey data indicated that the mental health of staff of the subject organisation did, to some extent, improve following the introduction of a number of these types of initiatives. However, improvements were small and would need to be confirmed with a longer follow-up period. The limited ability to access some of the initiatives due to organisational and cultural factors (including heavy workload and risk of stigmatisation) and little evidence for any change in managerial support weakened, and potentially reduced the organisation-level impact of, the reforms. The public sector wide changes to employment conditions and management structures which coincided with the launch of the interventions may also have diminished the effect on wellbeing outcomes.

The qualitative statements of participants provide further insight into these results. Consistent with the findings of the 2013 study,^[76] they suggest that primary interventions, such as flexible working practices and leave provisions, are more likely to be perceived as effective in supporting the mental health and wellbeing of staff. Also viewed as effective were initiatives intended to assist in the development of supportive relationships between staff, and between staff and management. Participants reported more mixed experiences with secondary and tertiary interventions, with interventions which provided practical information and resources to assist staff cope with stress and pressure on a daily basis viewed most positively.

These findings provide substantial insights which can assist in the future development of initiatives for legal organisations. They suggest the importance of primary interventions and providing means for staff to reduce their experience of stress, rather than simply responding to stressors. The study also, however, indicates that, where funding and resourcing considerations restrict the capacity of workplaces to reduce lawyers’ workload or facilitate extensive flexible working practices, initiatives which support improved peer and staff/managerial relationships may play a significant role.

Note, however, that it is not always within the power of organisations to change working conditions or workplace culture from the top down. Some working conditions are heavily affected by external factors and workplace culture is influenced by the wider professional or even national culture.^[77] While primary interventions to some extent place the onus of change on organisations and their leadership, secondary interventions tend to place emphasis on individuals being responsible for building their own resilience in managing work demands. There is a danger that such an emphasis on ‘individual troubles’ can ‘divert attention away from fundamental social, economic and political issues in the way law is practised and taught that are known to create or sustain depression and anxiety’.^[78]

VII ACKNOWLEDGEMENT

This project was funded by an Australian Research Council Linkage Project Grant 2012–2015 (LP120200328). The authors would like to thank the management and staff of the Partner Organisation for their support and cooperation throughout the project. We also thank Sally Doran for her valuable contributions, especially in conducting interviews and observation, and the analysis of survey data. In addition, we gratefully acknowledge the assistance of Roland Alexander with field research, Holly Blackmore with project management and data analysis, and Jenny Bargen with administrative support in 2013.

[****] Senior Research Fellow, Sydney School of Public Health, University of Sydney.

[1] Beyondblue, ‘Annual Professions Survey’ (Research Summary, April 2007) 2.

[2] Norm Kelk et al, ‘Courting the Blues: Attitudes towards Depression in Australian Law Students and Legal Practitioners’ (Report, Brain & Mind Research Institute, University of Sydney, 2009) 12. The percentages of respondents reporting moderate to very high levels of psychological distress were: 69 per cent (law students), 63 per cent (solicitors) and 44 per cent (barristers), compared with 37 per cent of the general population aged greater than 17 years.

[4] Adele J Bergin and Nerina L Jimmieson, ‘Australian Lawyer Well-Being: Workplace Demands, Resources and the Impact of Time-Billing Targets’ (2014) 21 Psychiatry, Psychology and Law 427, 434.

[6] See the general discussion of the reliability and validity of the Australian evidence in ibid 1110–23. Parker identifies three ways in which ambiguity can be introduced into the statistics presented: (1) the ‘way the results from the psychological distress scales used in these surveys tend to be cited to imply categorical clinical conclusions’: at 1113, 1114–15; (2) the surveys’ reliance on ‘voluntary, self-selected convenience samples’ suggests that the survey samples are not necessarily representative of the population: at 1116, 1117–20, and (3) the impression that there is a causal link between law and mental health problems: at 1120–3. Also, for a note on the representativeness of samples, see Janet Chan, ‘Conceptualising Legal Culture and Lawyering Stress’ (2014) 21 International Journal of the Legal Profession 213, 230.

[7] For evidence that American lawyers were more likely than the general population to be depressed, suffer from anxiety or other mental disorders, abuse alcohol and drugs and commit suicide, see Susan Daicoff, ‘Lawyer, Know Thyself: A Review of Empirical Research on Attorney Attributes Bearing on Professionalism’ (1997) 46 American University Law Review 1337; Patrick J Schiltz, ‘On Being a Happy, Healthy, and Ethical Member of an Unhappy, Unhealthy, and Unethical Profession’ (1999) 52 Vanderbilt Law Review 871. In contrast to this, a comprehensive review of the North American literature on lawyers over 25 years has concluded that a high percentage of lawyers (about 8 in 10) are satisfied with their jobs or careers, see Jerome M Organ, ‘What Do We Know about the Satisfaction/Dissatisfaction of Lawyers? A Meta-Analysis of Research on Lawyer Satisfaction and Well-Being’ (2011) 8 University of St Thomas Law Journal 225. For a discussion of this paradox, see Chan, above n 6.

[8] Kennon M Sheldon and Lawrence S Krieger, ‘Service Job Lawyers Are Happier than Money Job Lawyers, Despite Their Lower Income’ (2014) 9 Journal of Positive Psychology 219. The researchers found that ‘[s]ervice lawyers reported higher aggregate well-being and less negative affect’ even though these lawyers earned much less income: at 223. See also Lawrence S Krieger and Kennon M Sheldon, ‘What Makes Lawyers Happy?: A Data-Driven Prescription to Redefine Professional Success’ (2015) 83 George Washington Law Review 554.

[10] See Suzie Forell, Michael Cain and Abigail Gray, ‘Recruitment and Retention of Lawyers in Regional, Rural and Remote New South Wales’ (Report, Law and Justice Foundation of New South Wales, September 2010); Michael Cain and Suzie Forell, ‘Recruitment and Retention of Community Sector Lawyers: Regional Differences within New South Wales’ (2011) 16 Deakin Law Review 265.

[11] Charles R Figley, ‘Compassion Fatigue as Secondary Traumatic Stress Disorder: An Overview’ in Charles R Figley (ed), Compassion Fatigue: Coping with Secondary Traumatic Stress Disorder in Those Who Treat the Traumatized (Brunner/Mazel Publishers, 1995) 1; Lila Petar Vrklevski and John Franklin, ‘Vicarious Trauma: The Impact on Solicitors of Exposure to Traumatic Material’ (2008) 14 Traumatology 106. See generally on secondary traumatic stress: Joy D Osofsky, Frank W Putnam and Cindy S Lederman, ‘How to Maintain Emotional Health When Working with Trauma’ (2008) 59 Juvenile and Family Court Journal 91.

[12] Jennifer Pierce, Gender Trials: Emotional Lives in Contemporary Law Firms (University of California Press, 1995) ch 4 (Mothering Paralegals: Emotional Labor in a Feminized Occupation) 83, 86.

[13] Karen Pugliesi, ‘The Consequences of Emotional Labor: Effects on Work Stress, Job Satisfaction, and Well-Being’ (1999) 23 Motivation and Emotion 125, 146.

[14] Anthony D LaMontagne et al, ‘Job Stress as a Preventable Upstream Determinant of Common Mental Disorders: A Review for Practitioners and Policy-Makers’ (2010) 9 Advances in Mental Health 17, 19.

[16] Robert Karasek and Töres Theorell, Healthy Work: Stress, Productivity, and the Reconstruction of Working Life (Basic Books, 1990) ch 2.

[17] Johannes Siegrist, ‘Adverse Health Effects of High-Effort/Low-Reward Conditions’ (1996) 1 Journal of Occupational Health Psychology 27, 29 ff.

[18] M Kivimäki et al, ‘Organisational Justice and Health of Employees: Prospective Cohort Study’ (2003) 60 Journal of Occupational and Environmental Medicine 27.

[19] For a summary of these studies, see generally LaMontagne et al, ‘Job Stress as a Preventable Upstream Determinant’, above n 14.

[21] See Leonard I Pearlin et al, ‘The Stress Process’ (1981) 22 Journal of Health and Social Behavior 337.

[22] See Richard S Lazarus and Susan Folkman, Stress, Appraisal, and Coping (Springer Publishing, 1984).

[25] Schieman, Whitestone and Van Gundy’s research in Toronto, Canada found that workers in higher-status occupations report higher levels of work-to-home conflict than those in lower-status jobs: Scott Schieman, Yuko Kurashina Whitestone and Karen Van Gundy, ‘The Nature of Work and the Stress of Higher Status’ (2006) 47 Journal of Health and Social Behavior 242.

[26] See Jean E Wallace, ‘Job Stress, Depression and Work-to-Family Conflict: A Test of the Strain and Buffer Hypotheses’ (2005) 60 Relations Industrielles/Industrial Relations 510.

[28] See, eg, John Briton, ‘Lawyers, Emotional Distress and Regulation’ (Speech delivered at the Bar Association of Queensland 2009 Annual Conference, Gold Coast, 8 March 2009); Christopher Kendall, ‘Report on Psychological Distress and Depression in the Legal Profession’ (Council of the Law Society of Western Australia, March 2011) 12.

[32] See, eg, Rachael Field, James Duffy and Colin James (eds), Promoting Law Student and Lawyer Well-Being in Australia and Beyond (Routledge, 2016). See also Briton, above n 28, 8; Gary Davis and Susanne Owen, ‘Learning and Teaching in the Discipline of Law: Achieving and Sustaining Excellence in a Changed and Changing Environment’ (Report, Australian Learning and Teaching Council, 2009) ch 8.

[33] See The College of Law, Resilience <https://www.collaw.edu.au/learn-with-us/our-programs/practical-legal-training-programs/coursework/resilience>.

[34] See Parker, above n 5, app 2 for a list of nearly 30 national or state-based programs including the work of law schools, the Law Institute of Victoria, the Tristan Jepson Memorial Foundation and the Wellness Network for Law.

[35] Tristan Jepson Memorial Foundation, ‘TJMF Workplace Wellbeing: Best Practice Guidelines for the Legal Profession’ (Guidelines, May 2015) 5 <http://www.tjmf.org.au/wp/wp-content/uploads/TJMF-Legal-Workplace-Guidelines.pdf> . The Guidelines provide a framework for legal organisations to implement cultural change and change attitudes towards psychological health and safety.

[36] See Law Admissions Consultative Committee, ‘Practical Legal Training: Competency Standards for Entry-Level Lawyers’ (Standards, 1 January 2015) 5.

[37] Beyondblue and Beaton Research + Consulting, ‘2011 Annual Business and Professions Study’ (Research Summary, May 2011) 3.

[42] Ibid 1089–90. EAPs are confidential intervention programs usually paid for by employers to assist employees in resolving personal or workplace problems.

[43] See a review of 90 reports of systematic evaluations of job-stress interventions by Anthony D LaMontagne et al, ‘A Systematic Review of the Job-Stress Intervention Evaluation Literature, 1990–2005’ (2007) 13 International Journal of Occupational and Environmental Health 268. See also Anthony D LaMontagne, Tessa Keegel and Deborah Vallance, ‘Protecting and Promoting Mental Health in the Workplace: Developing a Systems Approach to Job Stress’ (2007) 18 Health Promotion Journal of Australia 221, 223–4.

[44] See the comprehensive review of the literature in LaMontagne et al, ‘Job-Stress Intervention Evaluation’, above n 43.

[47] The remaining employees worked in administrative or business support roles.

[50] Ethics approval for this project was granted by the University of New South Wales Human Ethics Research Committee (HC12439).

[51] This article is primarily based on the survey data; only limited use is made of the interview data and no observational data has been cited.

[52] Since there were some changes to the range of initiatives provided by the organisation over time, survey participants were only asked to assess initiatives that were operating at the time of the survey.

[54] A large body of research provides evidence to support Siegrist’s model, with measures of ERI shown to be linked at the organisational level to absenteeism, job satisfaction, employee turnover, health care expenditures and workers’ compensation claims. Research has also shown overcommitment to be associated with increased risks of poor physical health such as cardiovascular disease. For a review of this literature, see Nanna H Eller et al, ‘Work-Related Psychosocial Factors and the Development of Ischemic Heart Disease: A Systematic Review’ (2009) 17 Cardiology Review 83; Michael Marmot, Johannes Siegrist and Tores Theorell, ‘Health and the Psychosocial Environment at Work’ in Michael Marmot and Richard G Wilkinson (eds), Social Determinants of Health (Oxford University Press, 2^nd ed, 2006) 97; Akizumi Tsutsumi and Norito Kawakami, ‘A Review of Empirical Studies on the Model of Effort-Reward Imbalance at Work: Reducing Occupational Stress by Implementing a New Theory’ (2004) 59 Social Science & Medicine 2335; Natasja van Vegchel et al, ‘Reviewing the Effort-Reward Imbalance Model: Drawing Up the Balance of 45 Empirical Studies’ (2005) 60 Social Science & Medicine 1117.

[55] Tage S Kristensen et al, ‘The Copenhagen Psychosocial Questionnaire – A Tool for the Assessment and Improvement of the Psychosocial Work Environment’ (2005) 31 Scandinavian Journal of Work, Environment & Health 438.

[56] The DASS-21 contains seven items for each of the three DASS scales (21 items in total) and respondents are asked to indicate on a four-point severity/frequency scale the extent to which they have experienced each state over the past week: P F Lovibond and S H Lovibond, ‘The Structure of Negative Emotional States: Comparison of the Depression Anxiety Stress Scales (DASS) with the Beck Depression and Anxiety Inventories’ (1995) 33 Behaviour Research and Therapy 335.

[57] See generally Ronit Dinovitzer et al, ‘After the JD: First Results of a National Study of Legal Careers’ (Report, American Bar Foundation and NALP Foundation, 2004).

[58] To protect the identity of the organisation, its size is not mentioned in this article. The response rates are also not specified for the same reason; they varied between waves and were less than 50 per cent.

[59] See Colleen Cook, Fred Heath and Russel L Thompson, ‘A Meta-analysis of Response Rates in Web- or Internet-Based Surveys’ (2000) 60 Educational and Psychological Measurement 821; Michael D Kaplowitz, Timothy D Hadlock and Ralph Levine, ‘A Comparison of Web and Mail Survey Response Rates’ (2004) 68 Public Opinion Quarterly 94.

[60] See Thomas D Cook and Donald T Campbell, Quasi-experimentation: Design and Analysis Issues for Field Settings (Houghton Mifflin, 1979).

[61] ‘High risk’ areas are those where there may be an increased risk of psychological injury or accumulative stress among staff (eg, exposure to potentially offensive or traumatic materials).

[62] Preliminary analyses revealed that the overall DASS scores would need to be transformed (by taking the square root) to achieve best model fit. Beta coefficients therefore show the change in the square root of the overall DASS score for the relevant category compared with the referent category. Scores on the Overcommitment, Managerial Support and Collegial Support scales were entered as continuous variables. Beta coefficients for the continuous variables show the change in the square root of the overall DASS scores for a one unit increase in the independent variable.

[63] The residuals for most of the linear models had a similar pattern whereby they seemed to become more negative the higher the predicted score (suggesting that the model overestimates for larger values of the outcome). Sensitivity checks were therefore undertaken using non-linear methods (Poisson regression with robust VCE [Huber/White/Sandwich estimator]) which do not require that the errors are independently and identically distributed, or normally distributed. The Poisson results overall support the direction and significance of the linear models.

^[64] After other relevant variables are taken into account, it is male legal support officers who are reporting significantly more negative symptoms than their female counterparts. However, the number of legal support officers who were male is very small.

[65] Not all initiatives were available at each of the surveys. Unless otherwise indicated, the percentage figures indicate the range across all three surveys.

[66] The following notations are used to categorise sources of quotes: INT for interviews; Baseline, FU1 and FU2 for comments included in the respective surveys.

[67] The analysis is exploratory in the sense that the implementation of such programs and their impact is new and therefore we do not have an expectation of an effect size. Hence, a liberal definition of significance is used in order to be able to detect where there may be some effect, albeit at the counterbalancing risk of Type 1 error.

[68] The DASS is specifically designed to differentiate between the states of depression, anxiety and stress/tension. The three scales are described as follows: ‘the Depression scale is characterised principally by a loss of self-esteem and incentive, and is associated with a low perceived probability of attaining life goals of significance for the individual as a person ... The DASS Anxiety scale emphasises the links between the relatively enduring state of anxiety and the acute response of fear ... the DASS Anxiety scale additionally addresses situational anxiety. The content of the Stress scale suggests that it is measuring a state of persistent arousal and tension with a low threshold for becoming upset or frustrated’: Lovibond and Lovibond, above n 56, 342. The short version of DASS (DASS-21) was used here and includes 21 questions asking respondents to rate on a four-point severity scale whether certain statements (eg, ‘I felt I was close to panic’) applied to them at all over the past week. It is important to note that these questions are not designed to diagnose clinical psychological disorders.

[69] Comparisons can still be planned even though exploratory – the term refers to whether the intention is to be definitive about the impact of the exposure variable and that is not the case here. See Ralf Bender and Stefan Lange, ‘Adjusting for Multiple Testing – When and How?’ (2001) 54 Journal of Clinical Epidemiology 343.

[70] Given these analyses are exploratory, a less stringent threshold for statistical significance is appropriate. See ibid.

[71] This is supported by responses provided to the more detailed 15-item job satisfaction measure adopted in Dinovitzer et al, above n 57, 49. This indicated that of the 15 different aspects of their work role, 2013 survey respondents were most dissatisfied with job security relative to their 2011 counterparts. Similar differences on this satisfaction item were not apparent for the 2015 survey sample when compared with the Baseline.

[72] These dimensions were based on responses to a series of questions from the medium version of the Copenhagen Psychosocial Questionnaire COPSOQ, a valid and reliable tool for assessing job stress intervention needs at the organisational level: see Tage S Kristensen and Vilhelm Borg, ‘Copenhagen Psychosocial Questionnaire (COPSOQ): A Questionnaire on Psychosocial Working Conditions, Health and Well-Being in Three Versions’ (2003); see also Kristensen et al, above n 55.

[73] Other broader contextual issues, such as global changes in the public sector, may also be a factor, but such an analysis is beyond the scope of this research.

[74] Note that the three surveys provide both quantitative and qualitative data. While the quantitative data is generally representative of the organisation as a whole, the qualitative data (in the form of comments or answers to open-ended questions) consists of inputs from respondents who are self-selected and therefore not necessarily representative of the staff population. In view of this, the qualitative responses are only used to raise issues. Similarly, because of the low response rates and the self-selected nature of the interview sample, the data does not provide a balanced representation of the views of the population of staff; this data is used only for raising issues. We also focus on issues and avoid using numeric or quasi-numeric statements about the frequency with which comments were made, which could lead to misleading inferences regarding the prevalence of views: see Jane Ritchie et al (eds), Qualitative Research Practice: A Guide for Social Science Students and Researchers (Sage Publications, 2^nd ed, 2014).

[77] For example, there have been significant changes in the field of legal practice (eg, changes in market structure, size, ownership, globalisation and management practices). See Chan, above n 6, 226–8. For global changes that affect private legal firms, see Iain Campbell and Sara Charlesworth, ‘Salaried Lawyers and Billable Hours: A New Perspective from the Sociology of Work’ (2012) 19 International Journal of the Legal Profession 89. For similarly global changes in legal aid practices in the UK, see Hilary Sommerlad, ‘The Implementation of Quality Initiatives and the New Public Management in the Legal Aid Sector in England and Wales: Bureaucratisation, Stratification and Surveillance’ (1999) 6 International Journal of the Legal Profession 311.

University of New South Wales Law Journal

Poynton, Suzanne; Chan, Janet; Vogt, Melissa; Grunseit, Anne; Bruce, Jasmine --- "Assessing The Effectiveness Of Wellbeing Initiatives For Lawyers And Support Staff" [2018] UNSWLawJl 20; (2018) 41(2) UNSW Law Journal 584

I INTRODUCTION

II RESEARCH ON LAWYERING STRESS AND WELLBEING INITIATIVES

A Mental Health and Wellbeing in the Legal Profession

B Understanding Work Stress

C Wellbeing Interventions

III THE STUDY

A Research Design

1 Theoretical Framework

2 Research Methods

3 The Survey Instruments

4 Response Rates and Limitations

B Health and Wellbeing Initiatives

IV DRIVERS OF WORK STRESS AMONGST LAWYERS AND LEGAL SUPPORT STAFF

A Overall Drivers of Work Stress

B Comparison between Lawyers and Support Staff

V EFFECTIVENESS OF INITIATIVES

A Awareness and Usage of Initiatives

B The Impact of New Wellbeing Programs

1 Differences by Rates of Program Participation

2 Organisation-Wide Changes Post-implementation

C Users’ Perceptions of Program Effectiveness

1 Primary Interventions

2 Secondary Interventions

3 Tertiary Interventions

D Recommendations for Improvements

1 Primary Interventions

2 Secondary Interventions

3 Tertiary Interventions

VI CONCLUSION

VII ACKNOWLEDGEMENT