Too stressed to de-stress? The experience of work stress and recovery among attorneys during the COVID-19 pandemic

Abstract

It is well-established that legal professionals experience high stress and often high rates of associated health problems. Much less is known about attitudes and behaviours around stress prevention in this occupation. Our study examined views of stress and recovery among 131 U.S. attorneys. In open-ended data, attorneys commonly expressed that their job is very demanding, and it impacts their health. Many respondents felt it was important to manage their stress but had difficulty doing so. Quantitative analyses showed that attitudes about stress (stress-related comparisons, viewing stress as achievement, stress-related impression management, and stigma around stress concerns) demonstrated several significant relationships with perceived stress, recovery experiences, remorse for relaxation, and work–family conflict. Our findings suggest that practical interventions to support the health and well-being of legal professionals may need to target the workload norms, as well as attitudes and beliefs about the normalness of high stress and insufficient recovery.

Attorneys historically report high levels of stress, which has only increased during the pandemic, with 46% of attorneys reporting feeling overwhelmed in a recent survey (American Bar Association [ABA], 2021). The International Bar Association found that 35% of attorneys from around the globe said that their work negatively affected their mental health, with 49% of young attorneys between the ages of 24 and 29 reporting the same (International Bar Association [IBA], 2021). A survey of British attorneys found that risk of burnout was particularly high during the pandemic and that 69% of respondents were reporting poor mental health (LawCare, 2021). The pandemic’s effects on mental health were felt among a variety of nations (Kola et al., 2021) and resulted in higher work and family demands and blurred boundaries for many (Aguiar et al., 2021), which further resulted in many feeling increased stress and burnout (APA, 2020; Kerr et al., 2021). Though resources, such as self-care or recovery, can help protect workers from the negative effects of high levels of stress, often those who need recovery the most do not get it because they face too many demands (i.e. a recovery paradox; Sonnentag et al., 2017). Barriers to resource replenishment need to be better understood, particularly among attorneys and others in high-stress occupations. The goal of the present study was to better understand such barriers by exploring personal attitudes toward stress as correlates of recovery in a mixed-method study of attorneys in the United States during the COVID-19 pandemic.

Theoretical background

Often used as a framework for organizational stress research, Demerouti et al.'s (2001) Job Demands-Resources Model (JD-R) proposes that strain and burnout result from a lack of job resources and too many or too high a level of job demands. Job demands can include any aspects of work that require sustained effort (e.g. role conflict, complex tasks), while job resources are aspects of work that help to achieve work goals (e.g. feedback from superiors and skill variety) and can serve as a buffer for the possible negative effects of job demands (Bakker & Demerouti, 2017). Applying the JD-R Model, many studies have found evidence of psychological and physical health problems associated with high levels of stress that are attributed to high demands and insufficient resources (e.g. Bakker et al., 2003; Melamed et al., 2006; Wang et al., 2012). Such a pattern has emerged with studies of attorneys, where demands, such as workload, correlate with higher burnout, while resources, like decision latitude, correlate with less burnout (e.g. Nickum & Desrumaux, 2022).

Using the JD-R framework, Bakker and Demerouti (2017) proposed a dichotomy between high prestige careers (which includes attorneys as an example) and more mundane jobs. They argued that high-prestige jobs include high demands and high resources and are thus experienced as engaging. However, research and anecdotal evidence suggest that many attorneys do have high job demands and workloads, yet critically lack sufficient resources to sustain their personal well-being. Furthermore, having a high level of investment in their work (i.e. over-engagement) can relate to higher levels of burnout (Nickum & Desrumaux, 2022). It is important to understand how to encourage resource replenishment for attorneys to be able to perform optimally and experience the benefits of engaging work without over-extending oneself.

Demands and resources in high-stress work

Even prior to the pandemic, levels of stress and occupational pressures had long been acknowledged among attorneys across many countries and contexts (e.g. Australia, Bergin & Jimmieson, 2014; Canda/United States, Koltai et al., 2018; Estonia, Teichmann et al., 2015). The high levels of stress experienced by attorneys often correlate with unhealthy coping strategies and negative outcomes for attorneys—physically, mentally, emotionally, and interpersonally. For instance, Bergin and Jimmieson (2014) found that 37% of their sample of attorneys had depressive symptoms or anxiety and that 35% of the attorneys exhibited unhealthy levels of alcohol consumption. Similar high rates of mental health symptoms and addictions have been found in several other studies (e.g. Anker & Krill, 2021; Krill et al., 2016). It is unsurprising that attorneys report high levels of health concerns, as job stress is known to affect psychological health, physical health (e.g. Kivimäki et al., 2006; Nixon et al., 2011), and health behaviours (e.g. physical activity; Bennett et al., 2006; Kouvonen et al., 2013). Even though some types of stressors can be perceived as a challenge and thus experienced as motivating (Crawford et al., 2010), both challenge and hindrance stressors can negatively impact health if resources are insufficient (Mazzola & Disselhorst, 2019). It is likely the case that attorneys in high-prestige work are frequently motivated by the challenges of their job, and as a result can be very committed and engaged, yet when investment in work outweighs personal resources, they may still be subject to ill effects on their well-being (Nickum & Desrumaux, 2022). Therefore, it is important to not only understand the impacts of stressors on attorneys, but also the barriers to getting adequate resource replenishment.

Compared to the studies documenting the high stress and poor health among attorneys, there is far less research on stress prevention and recovery. There are programs and resources to support attorney mental health (e.g. Lawyer Assistance Programs through the ABA) and tests of stress management interventions in law school (e.g. stress inoculation training, Sheehy & Horan, 2004; cognitive behavioural therapy training, Lawrence et al., 1983), but there are relatively few investigations focused on truly understanding attitudes and experiences of recovery for this occupational group. To the authors’ knowledge, the only other study that has focused on attorney’s strategy to manage stress found that techniques like cognitive restructuring can improve work-related attitudes, but still may not reduce psychological strain (Brough & Boase, 2019). These authors further found that relaxation methods (e.g. deep breathing, meditation) did not significantly correlate with work attitudes and correlated with higher reported psychological strain. Therefore, the factors that affect quality recovery need further attention.

Sonnentag and Fritz (2007) more specifically describe four aspects that characterize restorative recovery experiences: relaxation, control of leisure time, mastery experiences (i.e. activities that provide a sense of personal challenge or growth), and psychological detachment (i.e. mentally separating from work). They found that all four recovery experiences were positively related to well-being, and additional studies find these experiences in the evening to relate to more positive affect the next day (Sonnentag et al., 2008). Psychological detachment in particular has exhibited some of the strongest relationships with well-being (Sonnentag & Fritz, 2007) and may help to offset potential negative effects of a high amount of job demands (Sonnentag et al., 2010). Alternatively, not recovering well, or feeling remorse for taking time to relax, can be related to worse recovery experiences and worse health (Black & Britt, 2022). Knowing the benefits of recovery, but also that many who experience high demands have difficulty recovering (Sonnentag et al., 2017), our study attempted to understand more specific attitudes and beliefs that may both increase stress and inhibit recovery.

Recent studies have brought attention to the impact of how workers themselves view stress. First, stress badge perceptions (Black & Britt, 2022), or views that a high workload is a part of being successful and something that should be displayed to others, have been found to negatively correlate with employee recovery and psychological health. The stress badge perceptions measure captures views of a high workload as part of being successful (stress as achievement), tendencies to compare one’s workload to others (stress-related comparison), and efforts to appear to have a heavy workload to others (stress-related impression management). The measure also included relaxation remorse, or guilt for taking time to recover, as an additional component of viewing stress as a badge of honour. Second, stress-reporting stigma (Klinefelter et al., 2021), or views that being unable to cope with stress is undesirable by others and oneself, has been found to relate to higher burnout. Initial investigations of these constructs were conducted with general employee samples, but these perceptions may be particularly relevant for attorneys and others in high-prestige jobs.

Given the high stress nature of the work done by attorneys, it is likely that norms of stress being perceived as to some degree normative and honourable and stigma around not being able to handle stress may be prevalent. Indeed, concerns about norms around stress are already evident in IBA (2021) surveys indicating that 41% of surveyed attorneys would not discuss mental well-being concerns with their employer, and 32% feared being treated negatively if they were to share a mental well-being concern. Therefore, in this study we explored how stress badge perceptions and stress-related stigma relate to perceived stress, recovery and interference between work and non-work domains in this occupation. Importantly, we considered whether these effects were significant, even when controlling for reports of workload demands, given the strong and consistent relationships that have been found between a high workload and outcomes such as work–family conflict (WFC) and worker health (Nixon et al., 2011; Schulz et al., 1998; Shirom et al., 1997; Wallace, 2002). In addition to quantitative analyses, we examined these questions using open-ended data about how attorneys feel stress impacts them and their motives for managing stress. In sum, our hypotheses and research questions were:

Research question 1: How do attorneys feel stress affects them and how do they feel about their ability/need to manage stress?

Hypothesis 1: Stress badge perceptions will correlate with fewer recovery experiences, more relaxation remorse, more WFC, and more perceived stress, even when controlling for workload.

Hypothesis 2: Stress-reporting stigma will correlate with fewer recovery experiences, more relaxation remorse, more WFC, and more perceived stress, even when controlling for workload.

Individual differences

We also explored quantitative differences in attitudes related to stress and recovery based on participants’ gender and tenure with their organization. Discrepancies exist between stress levels for men and women attorneys beginning in law school (McIntosh et al., 1994) and continuing into their career. Recent surveys indicated 49% of men attorneys report high levels of stress, while 67% of women report the same (ABA, 2021). Rates of mental health concerns tend to be higher among women attorneys compared to men (Anker & Krill, 2021). During the COVID-19 pandemic, a larger percentage of women attorneys (60%) and of attorneys of colour (54%) reported feeling overwhelmed than their white male counterparts (38%; ABA, 2021). Women attorneys in the British legal system also reported higher rates of burnout and lower levels of autonomy at their work compared to men during the pandemic (LawCare, 2021). Some of these differences may be rooted in structural disparities between men and women within the legal field, particularly in stress, responsibilities, pay, and partnerships (e.g. Dinovitzer et al., 2009). Family obligations can play a large role in burnout, as women attorneys with high WFC were 4.5 times more likely to leave or seriously contemplate leaving the legal profession (Anker & Krill, 2021). Given differences in actual stress and health concerns, as well as differences in stress coping strategies (Flynn et al., 2009), it is likely that men and women also differ in their attitudes toward stress and recovery experiences. Given the elevated stress and challenges for women advancing in the field, we hypothesize the following:

Hypothesis 3:

Women will report higher workload stressors, perceived stress, stress badge perceptions, stress-reporting stigma, WFC and relaxation remorse, compared to men and fewer recovery experiences compared to men.

An attorney’s time spent with an organization is also likely to impact stress attitudes and recovery because of the hierarchical nature of many law firms, in which new attorneys are often required to work their way up to higher statuses. Anecdotal evidence suggests that the types and levels of stress experienced can differ for younger versus more established attorneys, where younger attorneys may face particularly high stress levels. Studies have found that less tenure with an organization correlates with more working hours for a sample of attorneys (Wallace, 2002). Other studies found that younger attorneys reported more stress-related health concerns and addiction (Krill et al., 2016). Some differences may also vary by context, as Koltai et al. (2018) found that higher-status lawyers in large firms reported poorer health compared to lower status lawyers in an American sample, but there were no differences in low- versus high-status lawyers among those surveyed in Canada. Despite some inconsistencies in prior findings, we hypothesize that a longer time in the profession will generally relate to more adaptive attitudes and better recovery.

Hypothesis 4:

Tenure will be negatively correlated with workload stressors, perceived stress, stress badge perceptions, stress-reporting stigma, WFC, and relaxation remorse, and positively correlated with recovery experiences.

Method

Participants and procedure

The study was approved by the author’s institutional review board prior to data collection (IRB #20-112). Attorneys from the American state of Tennessee were invited to complete the survey study through several local and regional organizations for legal professionals in the Spring of 2021. Interested attorneys provided informed consent and then responded to questions regarding basic demographics, open-ended questions about work and stress, and self-report scales assessing constructs related to stress, work experiences, health, and well-being. The survey took approximately 20–30 min to complete. Upon completion, participants could provide their contact information to be entered into a raffle for one of 20 $20 gift cards.

The sample consisted of 131 attorneys, practicing various forms of law (e.g. litigation, corporate, family). The average age of participants was 49.43 (SD = 14.54). Individuals had been practicing law for 21.79 years on average (SD = 15.50) and had been with their current employer an average of 13.03 years (SD = 11.98). The sample was fairly evenly split among men (53%) and women (47%). Reported race was predominantly White (92%), with the remaining being Black or African American (2%), Hispanic/Latinx (2%), Asian (2%), and less than 1% being American Indian/Alaskan Native or Middle Eastern/North African.

Measures

Demographics

Demographic variables collected included gender, age, race, marital status, area of practice, typical work hours, income, and tenure with one’s organization and in the field in general. For focal analyses, gender and tenure were used as control variables, given the likely relationships between these variables and stress and recovery.

Quantitative work overload

Spector and Jex's (1998) four item measure of quantitative work overload was used to index workload stressors, which was used as a control variable in our hypothesis tests. A sample item was “How often does your job require you to work very fast?” Items were rated on a 1 (less than once per month or never) to 5 (several times a day) scale. The scale demonstrated good internal consistency (α = .88).

Stress badge perceptions and relaxation remorse

Attitudes toward stress and recovery were assessed using Black and Britt (2022) four subscales, three of which were considered predictors and one that assessed a theoretical outcome (i.e. relaxation remorse). All items were rated on a 1 (strongly disagree) to 7 (strongly agree) scale. Three subscales focused on views of having high workload demands: stress as achievement was measured with seven items (e.g. Having a high workload makes it apparent that I’m successful; α = .92), stress-related social comparisons were assessed with three items (e.g. I look to my co-workers to determine if my workload is high enough; α = .89), and stress-related impression management was assessed with five items (e.g. I list off all my tasks to my co-workers so they know my workload is high; α = .89).

Relaxation remorse was assessed with six items (e.g. When I try to relax, I feel like I should be doing work instead; α = .95).

Stress-stigma

Stigma related to stress was measured using the self-stigma and career stigma subscales from Klinefelter et al. (2021), along with a modified version of the views of others at work subscale (rather than differentiating supervisors and co-workers, we used a single set of items referencing “people at work”). The career and others at work subscales were highly correlated (r = .86) and were thus combined into one average score, resulting in one stigma from others score and one self-stigma score. Sample items included, “People at work would make negative comments if I talked about a problem with handling the amount of stress I am under” (stigma from others) and “If I had a problem with excessive stress, I would be less satisfied with myself” (self-stigma). All items were rated on a 1 (strongly disagree) to 5 (strongly agree) scale. Cronbach’s alpha was .96 for the stigma from others scale and .94 for self-stigma.

Recovery experiences

Recovery experiences were assessed with Sonnentag and Fritz's (2007) 16-item recovery experiences questionnaire. Four items assessed each type of recovery experience (i.e. psychological detachment from work, relaxation outside work, mastery experiences that promote personal growth, control over how non-work time is spent). Participants rated their agreement for how often they experienced each item during their free time outside work on a 1 (strongly disagree) to 5 (strongly agree) scale. Sample items and internal consistency values were as follows: Psychological detachment (e.g. I forget about work; α = .85); relaxation (e.g. I kick back and relax; α = .84); mastery (e.g. I learn new things; α = .90); and, control (e.g. I feel like I can decide for myself what to do; α = .86).

Work–family conflict

We assessed WFC with six items (Carlson et al., 2000) focused on strain-based (e.g. When I get home from work I am often too frazzled to participate in family activities/responsibilities) and time-based WFC (e.g. My work keeps me from my family activities more than I would like). Items were rated on a 1 (strongly disagree) to 5 (strongly agree) scale. When combined into one WFC score, alpha was .89.

Perceived stress

Perceived stress was assessed with a short, four-item version of the Perceived Stress Scale (Cohen et al., 1983). Items assessed how often in the past month participants felt their life was overwhelming versus controllable (e.g. How often have you felt that you were unable to control the important things in your life?). Items were rated on a 1 (never) to 5 (very often) scale. Cronbach’s alpha was acceptable, at .72.

Open-ended questions

A series of open-ended questions encouraged more detailed remarks about attorneys’ stress and attitudes toward stress management. Those specific questions coded in the present study were: (1) How has your work life been impacted by COVID-19? Please give two or three examples, if applicable. (2) What would you say are the main sources of stress for you at work? Please list and explain. (3) In what ways does your work stress affect your life (a) at work and (b) outside work? (4) What things, if any, do you do to handle your work-related stress?, and (5) How important is it for you to manage your stress? Please briefly explain.

Data analysis

Quantitative hypothesis tests were conducted using SPSS v. 28. First, bivariate correlations were calculated among all study variables. Then a series of analyses were conducted to determine how stress badge perceptions and stress stigma correlated with recovery experiences, relaxation remorse, WFC and perceived stress after accounting for control variables. First, we used partial correlations to examine bivariate relationships, controlling for work overload demands, gender and tenure. Then, as an additional exploratory analysis, to determine which stress badge perceptions or stress stigma subscales may exert unique influence on outcomes, we used a series of hierarchical regressions with the control variables (work overload, gender, tenure) entered in Step 1 and the predictors of interest in Step 2. Finally, gender differences were examined in more detail using a series of independent-samples t-tests.

Open-ended questions were coded by two independent raters. An initial coding guide was developed for each question, which included themes expected from the research literature as well as novel ideas that were extracted by the principal investigator upon initial review of all remarks. Using the coding guides, all responses were reviewed by two independent coders. Each respondent’s answer could receive multiple codes based on the presence of one or more themes. Therefore, in reviewing the two sets of codes, there was either full agreement on all codes, partial agreement on some codes, or no agreement (see Table 1). Disagreements between the two coders were resolved by a third coder to determine a final code for each response.

Table 1.

Rater agreement for coding open ended questions.

Question	Full agreement N (%)	Partial agreement N (%)	No agreement N (%)	Total responses
What would you say are the main sources of stress for you at work? Please list and explain.	51 (46)	39 (35)	22 (20)	112
How has your work life been impacted by COVID-19? Please give 2 or 3 examples, if applicable.	59 (46)	49 (39)	19 (15)	127
In what ways does your work stress affect your life (a) at work and (b) outside of work?	37 (33)	53 (47)	22 (20)	112
What things, if any, do you do to handle your work-related stress?	58 (51)	43 (38)	12 (11)	113
How important is it for you to manage your stress? Please briefly explain.	102 (92)	0 (0)	9 (8)	111

Note. Any single response could be assigned one or more codes based on the number of themes present. Partial agreement represents agreement on at least one code, full agreement represents agreement on all assignment codes, and no agreement represents agreement on no codes.

Results

Our first research question concerned how attorneys felt stress affects them and about their ability and need to manage stress. This research question was primarily assessed with responses to open-ended questions. To understand views of stress, we first wanted to understand what typical stressors were (Table 2) and how the attorneys’ work life had changed since the COVID-19 pandemic (Table 3). By far the most common stressors mentioned in the open-ended question were the workload and deadlines (42%). Other relatively common stressors were billable hour requirements (13%), inadequate work–life balance (10%), unrealistic expectations (19%), conflict involved with work (12%), and general management challenges (13%). As a result of the pandemic, almost half (45%) transitioned into working from home at some point during the pandemic. Similarly, 35% used technology more and had more virtual meetings. Several stressors were noted as heightened in the pandemic, including reduced workload (14%) and reduced income (9%), challenges related to childcare (13%), missing workplace social connections (13%), and working odd hours or longer hours (7%).

Table 2.

Common sources of stress reported by attorneys.

Theme	N	%	Sample quote
Workload and deadlines	47	42	“Too many emails. Too many interruptions. Too much work.” / “Deadlines—litigators are always up against deadlines, and anticipating the time needed to complete a quality work product at the time of filing can be challenging.”
Unrealistic expectations (from others and from self)	21	19	“Dealing with clients who have unrealistic expectations. A close second is dealing with other attorneys who have unrealistic expectations.” / “Self-imposed expectations; client expectations.”
General organization and management	15	13	“File maintenance and management—making sure nothing falls through the cracks.”
Billable hour requirements	14	13	“Expectations to work billable hours plus engage in significant business development and marketing activities” / “Billable hours. I have always primarily worked in firms where I had to record time and bill by the hour.… Even today, stress is present due to the absence of billable work caused in part by slowing down due to age and also Covid-19”
Conflict involved in work	13	12	“Practicing law is stressful. Trial law is even more so, as it is adversarial (conflict-ridden) by its very nature. Family law might be the most stressful practice, as one deals with family conflict on a daily basis.”
Work–life balance / Need to be always available	11	10	“Trying to be here but also be a good mom.” / “Juggling family life, being on constant call for client emergencies…”
Lack of support from others	7	6	“My management does not seem to fully know the sources of the stress across the office, and sometimes acts almost as if it discounts it.”
Emotional difficulty/Trauma/Stress of others	7	6	“Many cases involve tragic circumstances (injuries, death, abuse) and have serious consequences to my clients. Stressful sometimes to put work out of mind during the rest of my life.”
High stakes work	6	5	“Responsibility for multi-million dollar problems—weighs heavily on you until and unless you get a favorable outcome”
Managing others	6	5	“Managing and training young attorneys to handle cases; allowing them room to learn and grow without messing up a client[’]s case.”
Feeling the need to prove oneself	3	3	“The largest source of stress at work is due to being a young female attorney in a primarily male-dominated field of law. A lot of clients automatically look down on me or are convinced I am not as effective as my male partners in the firm because of my age and gender. Clients second guess what I say only to be told the exact same thing by the male partners. This makes it more difficult to gain the trust of clients… and affects self-confidence.”
Lack of recognition/ affirmation	1	1	“…there is very little affirmation in this job.… Losing in litigation is difficult. Clients often do not acknowledge our effort on their behalf, and sometimes firm management does not seem to be appreciative.”
Other	24	21	“Commute to/from office” / “Non-responsive opposing counsel.… Attorneys not responding causes me to have to file unnecessary motions and be a person I don’t like to be.”

Note. N = 114.

Table 3.

Changes in work related to COVID-19 reported by attorneys.

Theme	N	%	Sample quote
Work from home	57	45	“More working from home, even though I don[’]t have a dedicated home office. My office is the playroom I share with our two toddlers.”
Increase use of technology and virtual meetings	45	35	“I have tried to meet with clients over zoom whenever possible. This is not always the most effective way to communicate with family law clients.”
Reduction in workload	18	14	“Courts were closed, so less work. More leisure as a result but then more stress due to decreased production and then too much work when things began opening back up.”
Missing social connection at work	17	13	“Lack of motivation, too isolated, lack of meaningful inter-action with other attorneys”
Childcare demands changed	16	13	“I work from home mostly now. I am balancing taking care of a toddler with client calls and emails.”
Reduction in income, insecurity related to income and workload	11	9	“Income has decreased from lack of ability to have trials. Without the threat of trial, not able to settle cases for as much money as before, or with the same frequency as before.”
Having to work odd hours & many working longer hours	9	7	“The biggest impact I noticed was the disruption caused by other attorney[s] transitioning to work from home, specifically the increase in after hours communications. Attorneys I deal with seemed far more likely to send an email or telephone (and expect a response) after normal business hours during the pandemic vs prior to the pandemic.”
Health concerns related to COVID-19	7	6	“I caught COVID, most likely from other firm employees who were not diligent. I was the careful one.”
No substantial impacts	8	6	“I came into the workplace once COVID had started, so this is all I have known.”
Other	27	21	“The move to many virtual things have made court appearance much more efficient generally. Instead of spending hours in court there are now assigned times which has been nice.”

Note. N = 127.

Next, we more specifically asked how attorneys’ felt that stress impacts them in their work and non-work roles (Table 4). Concerning work, 12% of attorneys in the study noticed negative cognitive effects of stress, and 9% directly stated that stress affects their work output. Outside work, 23% noted that they had difficulty disconnecting from work, and 32% felt fatigued or suffered from a lack of sleep. On an interpersonal level, 22% of participants expressed a strain on personal relationships, most importantly their partner and children.

Table 4.

Effects of stress reported by attorneys.

Theme	N	%	Sample quote
Poor or insufficient sleep/fatigue	36	32	“Wake at night thinking about cases; hard to shut down at night because start thinking about cases.”
Ability to relax/recover in non-work time impacted	26	23	“I find myself very irritable at home, and struggle to relax on vacation, mostly due to the continued ability to be connected to the office through devices that give access to emails and other apps/programs related to work.…”
Strain in relationships (e.g. irritable, conflict)	25	22	“I often feel short-tempered and irritable or when I get home.”
Emotional health	22	20	“I do not feel as happy and satisfied with work or home life.”
Little to no effects/stress is manageable	16	14	“My work stress does not have much of an a[e]ffect on my life at work because everyone else in our office is dealing with the same high volume workload. I just show up every day ready to work hard and do the best that I can do.”
Cognitive functioning	13	12	“When I get overwhelmed at work, it[’]s hard to really focus in on what tasks I need to do first and instead I waste a lot of time sitting there stressing, which only makes it worse!”
Time for desirable activities outside of work impacted	12	11	“Makes me tired, keeps me from taking time off and being able to enjoy weekends and weeknights, makes me care less about things”
Physical health concerns	11	10	“At work, frequently on edge. Has caused extreme hypertension where I[’]m now on 3 medications and carry a 4th as an emergency pill if blood pressure rises above 165.”
Work performance negatively impacted	10	9	“At work, if I[’]m stressed I sometimes make mistakes that I would not otherwise because my mind is on my stress. And then it becomes a vicious cycle stress → mistake → more stress → more mistakes.”
Unhealthy behaviours	6	5	“Also in these weeks, I will sometimes skip planned workouts and/or sleep to meet a deadline.” / “I drink more when I[’]m stressed.”
Other	21	19	“Stay organized” / “Keep perspectives and priorities in the right spot.”

Note. N = 113.

Digging into stress management and recovery, we asked attorneys about their typical stress management strategies (Table 5). Participants reported a variety of strategies. Using exercise to help manage stress emerged as a clear favourite strategy (58%), with other common strategies being hobbies and/or personal projects (30%), intentional self-care and/or health-protecting practices (19%), time with friends and family (15%) and religious and/or spiritual activities (14%). As noted before, unhealthy levels of drinking are problematic within the legal field; in the present study, 14% of participants listed alcohol use as a recovery activity.

Table 5.

Stress management activities and habits reported by attorneys.

Theme	N	%	Sample quote
Exercise	65	58	“I swim, bike, and run regularly and have for years, and a workout always helps with the stress. I also try to break up the work day with little one-minute sets working with free weights or stretching.” / “I enjoy walking and exercising to relieve stress. However, due to the COVID-19 pandemic, I have not felt comfortable with working out in a gym.”
Hobbies/personal projects	34	30	“I try to maintain my hobbies—although, given time commitments, that is often very hard to do.” / “Read quality literature, primarily biographical books and novels. Periodically will go for long walk. Recently began to enjoy a lake house retreat.…”
Intentional self-care/ health-protecting practices	22	19	“Stop and take deep breathes and exhale slowly from time-to-time.”
Time with friends or family outside of work	17	15	“I leave work at a regular time and spend time with my wife and non-work colleagues.” / “I play games at home with my kids. I read books to my kids before bedtime.”
Religious involvement/ spiritual practices	16	14	“Praying and medi[t]ating really help, too.”
Alcohol use	16	14	“Too many cocktails.…” / “unwinding at night with a glass of wine (or two!)”
Time and/or conversations with co-workers/ colleagues	8	7	“I vent to colleagues and my husband, as talking about it makes me feel better. Social events, like happy hours, dinner parties, drinks out, are my main way of destressing.”
Removing technology/disconnecting	6	5	“I[’]ve recently experimented with snoozing email notifications on the weekends and evenings… not having the message preview pop up on my screen has been nice.” / “I try to do outdoor activities, like golf or hiking, with my phone off, and not with me if possible.”
Don’t do anything	5	4	“Nothing really.”
Therapy/counselling	5	4	“I see my psychiatrist at least once per month, I take Zoloft.”
Vacations/travel	4	4	“Vacations with wife and travel to leave my professional life behind.”
Watch TV/movies	4	4	“Watch sports, other things that can turn mind off some.” / “Take too many TV breaks.”
Work more	3	3	“I wind up working more hours to try to reduce stress.”
Other	22	19	“Each week try to devote time to personal development to better manage the stress.” / “Try to blow it off.”

Note. N = 112.

When asked more specifically how important it was to them to manage their stress, several themes emerged. An overwhelming 91% of participants said that managing their stress was important to them. A few participants said that managing their stress was somewhat important (3%) and 1% said that it was not important but should be. Some participants did not specify the level of importance but nonetheless noted that they were managing their stress (3%) or not managing their stress (6%). Only 5% said that managing their stress was not important.

Hypotheses 1 and 2 focused on the relationships between specific attitudes toward stress and outcomes of recovery experiences, relaxation remorse, WFC and perceived stress. Correlations among these study variables, as well as work overload, gender and tenure, are provided in Table 6. There were significant positive correlations between all stress badge perceptions and stress-stigma subscales with relaxation remorse and WFC. These correlations were small to moderate in magnitude. Correlations with the different types of recovery experiences were more varied, but there were several significant negative relationships, particularly with stress-related social comparisons correlating with psychological detachment, control and mastery (but not relaxation) and stress-related impression management correlating with control, but not the other three forms of recovery. Interestingly, stress as achievement did not significantly correlate with any recovery experiences. Both self- and other- stress stigma negatively correlated with all four recovery experiences. Self- and other-stress stigma positively correlated with perceived stress, but the stress badge dimensions were not significantly correlated with perceived stress levels.

Table 6.

Correlations among all study variables and partial correlations, controlling for workload, gender and tenure.

	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
1. Gender	—
2. Tenure	−.24*	—
3. Work overload	.16	−.21*	(.88)
4. Stress achievement	.00	−.13	.21*	(.92)	.48*	.33*	.21*	.29*	−.10	−.09	−.05	−.18	.33*	.12	.05
5. SRSC	.27*	−.26*	.06	.46*	(.89)	.39*	.25*	.28*	−.14	−.03	−.35*	−.25*	.31*	.39*	.11
6. SRIM	.13	−.15	.15	.32*	.43*	(.89)	.19	.15	.02	.04	−.14	−.06	.36*	.25*	−.10
7. Stigma from other	.20	−.21*	.20*	.23*	.28*	.21*	(.96)	.49*	−.15	−.20	−.33*	−.22*	.29*	.26*	.22*
8. Self-stigma	.30*	−.26*	.25*	.31*	.32*	.18	.56*	(.94)	−.19	−.24*	−.28*	−.19	.44*	.21*	.24*
9. Detachment	−.22*	.25*	−.39*	−.18	−.23*	−.10	−.23*	−.28*	(.85)	.58*	.26*	.12	−.47*	−.18	−.25*
10. Relaxation	−.30*	.26*	−.24*	−.12	−.16	−.10	−.26*	−.32*	.66*	(.84)	.31*	.11	−.43*	−.06	−.15
11. Control	−.06	.12	−.08	−.06	−.37*	−.19*	−.32*	−.26*	.34*	.41*	(.86)	.40*	−.31*	−.27*	−.21*
12. Mastery	−.31*	.21*	−.10	−.17	−.32*	−.14	−.27*	−.28*	.23*	.27*	.42*	(.90)	−.22*	−.14	−.10
13. Relaxation remorse	.23*	−.14	.37*	.36*	.34*	.39*	.35*	.49*	−.56*	−.49*	−.35*	−.29*	(.95)	.28*	.30*
14. WFC	.26*	−.28*	.52*	.20*	.38*	.31*	.35*	.36*	−.35*	−.23*	−.27*	−.26*	.41*	(.89)	.27*
15. Perceived stress	.14	−.25*	.40*	.13	.16	.04	.30*	.35*	−.38*	−.26*	−.27*	−.17	.40*	.45*	(.72)

Note. Bivariate correlations (N _range = 99–131) among all variables are displayed below the diagonal. Partial correlations (N = 93), controlling for gender, tenure, and work overload are displayed above the diagonal. Cronbach’s α is reported on the diagonal in parentheses. SRSC: Stress related social comparisons; SRIM: stress-related impression management; WFC: work–family conflict.

*p < .05.

We also wanted to determine whether these attitudes toward stress and stressors still correlated with recovery, relaxation remorse, WFC and perceived stress after controlling for work overload and demographic variables. Thus, we first examined partial correlations, controlling for work overload, gender and tenure (see also Table 6). All three stress badge subscales and both stress-stigma subscales still maintained a significant positive correlation with relaxation remorse, and all but stress as achievement significantly correlated with WFC. Again, partial correlations with recovery experiences were more varied, with only stress-related social comparison from the stress badge measures correlating with less mastery activities and less control in non-work time. Stress stigma from others also correlated with less mastery and control. Self-stress stigma correlated with less relaxation and less control. Both stress stigma subscales remained significant correlates of perceived stress.

In sum, Hypothesis 1 received mixed support. The stress badge subscales were correlated with relaxation remorse and WFC (with the exception of stress achievement) after controlling for workload and demographic variables, but relationships with recovery experiences were few, and there were no significant relationships with perceived stress. Hypothesis 2 was largely supported where the stress stigma subscales were correlated with relaxation remorse, WFC, perceived stress, and two of four recovery experiences when also accounting for control variables.

For exploratory purposes, hierarchical regression analyses determined which of the stress badge perceptions and stress stigma subscales explained unique variance in our outcomes, again controlling for work overload, gender and tenure (see Table 7). In the model predicting relaxation remorse, the stress badge and stress stigma variables explained an additional 26% of variance in relaxation remorse, beyond the control variables. Only work overload, stress-related impression management, and self-stress stigma were significant unique predictors. In the model predicting WFC, the focal predictors explained 13% additional variance in WFC beyond the control variables. Only work overload and stress-related social comparisons exhibited significant unique relationships.

Table 7.

Multiple regression analyses for stress-related attitudes predicting recovery experiences, relaxation remorse and WFC.

	B	SE	β	sr 2	Model R2	ΔR2
Outcome: Detachment
Step 1					.17
Gender	−0.27	0.17	−.16	.02
Tenure	0.01	0.01	.10	.01
Workload	−0.28*	0.09	−.32	.09
Step 2					.22	.05
Gender	−0.18	0.18	−.10	.01
Tenure	0.00	0.01	.06	.003
Workload	−0.26*	0.09	−.29	.07
Stress achievement	−0.02	0.07	−.03	.001
SRSC	−0.05	0.06	−.10	.01
SRIM	0.08	0.09	.10	.01
Stress stigma from others	−0.05	0.09	−.07	.003
Stress self-stigma	−0.13	0.12	−.13	.01
Outcome: Relaxation activities
Step 1					.16
Gender	−0.35*	0.14	−.26	.06
Tenure	0.01	0.01	.15	.02
Workload	−0.10	0.07	−.15	.02
Step 2					.22	.06
Gender	−0.30*	0.14	−.22	.04
Tenure	0.01	0.01	.11	.01
Workload	−0.06	0.07	−.09	.01
Stress achievement	−0.03	0.06	−.06	.002
SRSC	0.02	0.05	.04	.001
SRIM	0.05	0.07	.08	.01
Stress stigma from others	−0.07	0.07	−.11	.01
Stress self-stigma	−0.15	0.10	−.19	.02
Outcome: Mastery
Step 1					.14
Gender	−0.57*	0.18	−.33	.10
Tenure	0.00	0.01	.06	.004
Workload	−0.06	0.09	−.06	.004
Step 2					.22	.08
Gender	−0.45*	0.18	−.26	.05
Tenure	0.001	0.01	.01	.0001
Workload	−0.03	0.09	−.03	.001
Stress achievement	−0.04	0.07	−.05	.003
SRSC	−0.10	0.06	−.19	.02
SRIM	0.05	0.09	.06	.003
Stress stigma from others	−0.11	0.09	−.14	.01
Stress self-stigma	−0.07	0.12	−.07	.003
Outcome: Control
Step 1					.01
Gender	−0.08	0.17	−.05	.002
Tenure	0.004	0.01	.06	.003
Workload	−0.03	0.08	−.04	.001
Step 2					.23	.22
Gender	0.16	0.16	.10	.01
Tenure	−0.0014	0.01	−.02	.0004
Workload	−0.02	0.08	−.02	.0004
Stress achievement	0.12	0.07	.21	.03
SRSC	−0.18*	0.06	−.37	.09
SRIM	−0.01	0.08	−.01	.00004
Stress stigma from others	−0.16*	0.08	−.23	.03
Stress self-stigma	−0.13	0.11	−.14	.01
Outcome: Relaxation Remorse
Step 1					.15*
Gender	0.50	0.32	.15	.02
Tenure	0.000	0.01	−.002	.000
Workload	0.54*	0.16	.33	.10
Step 2					.40*	.26*
Gender	0.19	0.29	.06	.003
Tenure	0.01	0.01	.09	.01
Workload	.33*	0.15	.20	.03
Stress achievement	0.14	0.12	.12	.01
SRSC	0.05	0.10	.05	.002
SRIM	.36*	0.14	.23	.04
Stress stigma from others	0.06	0.14	.04	.001
Stress self-stigma	.62*	0.20	.33	.06
Outcome: WFC
Step 1					.35*
Gender	0.26	0.16	.14	.02
Tenure	−0.01	0.01	−.13	.02
Workload	0.48*	0.08	.50	.24
Step 2					.48*	.13
Gender	0.06	0.16	.03	.001
Tenure	−0.01	0.01	−.08	.010
Workload	.46*	0.08	.49	.20
Stress achievement	−0.08	0.06	−.12	.01
SRSC	.18*	0.05	.31	.06
SRIM	0.09	0.08	.10	.01
Stress stigma from others	0.10	0.08	.12	.01
Stress self-stigma	0.06	0.11	.06	.002
Outcome: Perceived stress
Step 1
Gender	0.02	0.13	.01	.00	.20*
Tenure	−0.01	0.01	−.19	.03
Workload	0.26*	0.06	.36	.12
Step 2
Gender	−0.07	0.13	−.06	.00	.28*	.08
Tenure	−0.01	0.01	−.15	.02
Workload	0.22*	0.06	.33	.09
Stress achievement	−0.01	0.05	−.02	.00
SRSC	0.04	0.05	.10	.01
SRIM	−0.11	0.06	−.17	.02
Stress stigma from others	0.08	0.06	.13	.01
Stress self-stigma	0.13	0.09	.17	.02

Note. WFC: work–family conflict; SRSC: stress-related social comparisons; SRIM: stress-related impression management.

*p < .05.

Concerning recovery experiences, we found that the focal variables did not explain significant incremental variance in psychological detachment. Only work overload emerged as a significant unique predictor. There was also no significant incremental variance explained in relaxation or mastery experiences; the only significant unique predictor in this analysis was gender. The focal variables did explain an additional 22% of the variance in perceived control over non-work time, with stress-related social comparisons and stigma from others exhibiting significant unique relationships. Finally, when examining perceived stress as an outcome, the focal variables did not explain incremental variance. The only significant predictor in this analysis was quantitative workload.

In sum, from these exploratory analyses, it appears that stress stigma, stress-related social comparisons, and stress-related impression management could explain unique variance above and beyond the other stress attitudes for some outcomes. Stress as achievement never emerged as unique predictor and may not have as much of an impact on these recovery-related outcomes.

A summary of the significance of the findings for Hypothesis 1 and 2 in each of the analyses discussed (bivariate correlations, partial correlations, and multiple regression) is available in Supplemental Table 1.

Our final hypotheses concerned individual differences in gender and tenure as potential influences to perceived stress, stress attitudes, and recovery. Table 6 shows that gender was correlated with several of our variables. We further considered these isolated mean differences in independent samples t-tests (Table 8). We found that women reported higher WFC, stress-related social comparisons, stress self-stigma and relaxation remorse than men. Men reported higher levels of recovery experiences including relaxation, detachment and mastery than women. Differences in work overload, stress as achievement, stress-related impression management, stress stigma from others, control recovery experiences and perceived stress were not significant. These results partially support Hypothesis 3, in that there were some differences in recovery and WFC, as well as stigma perceptions around stress, but differences in workload, stress and other stress attitudes were not significant. These trends suggest that men and women may feel similar work demands and outward pressures around stress, but women may experience more difficulty recovering and internalization of stress-related pressures.

Table 8.

Independent samples t tests comparing men and women on workload, stress attitudes, recovery, relaxation remorse and WFC.

	Men M (SD)	Women M (SD)	t	df	Cohen’s d
Work overload	3.11 (.92)	3.42 (1.03)	−1.59	97	−.32
Stress as achievement	4.33 (1.38)	4.33 (1.28)	−.04	106	−.007
Stress-related social comparison	3.40 (1.39)	4.26 (1.78)	−2.84*	106	−.55
Stress-related impression management	2.14 (1.05)	2.40 (1.01)	−1.31	106	−.25
Stress stigma from others	2.54 (1.00)	2.98 (1.24)	−1.97*	97	−.40
Stress self-stigma	3.14 (.88)	3.66 (.75)	−3.11*	97	−.63
Detachment	2.90 (.77)	2.52 (.94)	2.33*	106	.45
Relaxation	3.62 (.63)	3.21 (.68)	3.24*	106	.63
Control	3.75 (.78)	3.66 (.82)	.64	106	.12
Mastery	3.69 (.77)	3.14 (.93)	3.38*	106	.65
Relaxation remorse	4.16 (1.57)	4.88 (1.54)	−2.39*	106	−.46
Work–family conflict	2.74 (.94)	3.22 (.86)	−2.66*	97	−.54
Perceived stress	2.56 (.60)	2.74 (.70)	−1.45	102	−.29

Note. WFC: work–family conflict.

*p < .05.

Table 6 also shows that tenure was negatively correlated with workload, stress-related comparisons, stress-stigma (self and others), WFC and perceived stress. Tenure positively correlated with several types of recovery experiences, including detachment, relaxation and mastery. Tenure did not significantly correlate with views of stress as achievement, stress-related impression management, control recovery experiences and relaxation remorse. In sum, there was support for Hypothesis 4 in regard to most outcomes considered. Attorneys newer to an organization were more likely to have higher perceived workloads and concerns with others’ evaluations of their stress, and less recovery experiences. However, views of stress as a part of achievement and managing one’s appearance of stress, along with control of recovery time and relaxation remorse, may be similarly experienced, regardless of tenure.

Discussion

These results indicate that the sheer amount of stress experienced by attorneys and attitudes around stress may prohibit successful recovery and lead to general expectations that being “stressed” is just part of the job. In examining our first research question, most of the attorneys surveyed (91%) recognized that managing their stress is important, but many noted various barriers to managing their stress successfully. Some felt that managing stress seemed near impossible, with remarks such as “I feel that I’m on a treadmill and can’t get off”, or, regarding managing stress, “I know that I have to or it [stress] will consume me. However, it seems to be consuming me even though I know it shouldn’t.” Qualitative findings also highlighted high stress levels, particularly in the context of the pandemic, consistent with other US and global surveys of attorneys during this timeframe (ABA, 2021; IBA, 2021). Quantitative analyses complemented open-ended remarks, such as workload and attitudes attorneys held about stress related to their abilities to engage in quality recovery.

Support for Hypothesis 1 was mixed, based on the unique subscales of the stress badge perceptions. Significant correlations with relaxation remorse were present for all subscales and correlations, even after controlling for perceived workload and demographics. Correlations with WFC were present for all subscales prior to accounting for control variables and remained significant for all subscales except stress as achievement. Thus, even if recovery is occurring, these attitudes may reduce attorneys’ quality of recovery and work–family relationships. This is not surprising, given the recovery paradox where those with high demands find it difficult to recover well (Sonnentag et al., 2017). Attorneys in particular may have difficulty avoiding guilt during relaxation time, if they feel relaxation interferes with their work progress or with their image as a high-achieving professional. Some of the guilt that arises from social pressures around stress may be related to the perceived need to always be available that many attorneys expressed, which was heightened even more than normal during the pandemic.

Surprisingly, none of the stress badge subscales were significantly correlated with perceived stress. This is contrary to prior work with the stress-related impression management and social comparison subscales in particular (Black & Britt, 2022). It could be that having these stress badge attitudes does not necessarily equate to more stress. Alternatively, many attorneys with a stress badge mentality may also underreport the stress they feel, or simply not be accurately aware of their stress. Consistent with this possibility, studies from other occupational contexts have suggested that high stress work environments can foster underreporting of perceived stress, even when stress-related health concerns are high (Faghri et al., 2015).

Correlations with actual recovery experiences were surprisingly inconsistent. Black and Britt (2022) previously found that all three subscales significantly correlated with detachment and relaxation activities. In our study, none of these perceptions correlated with relaxation. Only stress-related social comparison correlated with detachment, but this did not remain significant after accounting for controls. These non-significant findings could be a nuance of our sample, who reported overall low levels of detachment. As demonstrated in the qualitative data, many attorneys struggled to detach from the concerns and stresses of their work in non-work time, so there may be range restriction for relationships with detachment. Low detachment is concerning, considering the strong benefits it has for psychological health (e.g. Sonnentag & Fritz, 2007). It may also be that detachment and relaxation are better predicted in this population by other individual differences (e.g. mindfulness, leisure-related values) or situational factors (e.g. work schedule, family structure).

Stress-related social comparisons related to fewer recovery experiences of mastery and control. Stress-related impression management related to less control, but only prior to accounting for control variables. This was the first study, to the authors’ knowledge, to consider these other types of recovery experiences in connection to stress badge perceptions. It is likely that these more socially experienced comparisons and outward displays do leave individuals feeling less in control of their non-work time. Specifically, if one needs to “keep up” with the stress of those around them, they may feel like they have less of a choice in what they do in their free time. Furthermore, those with high stress-related social comparisons may be hesitant to take part in activities that are not productive for work and not partake in typical mastery activities, such as hobbies that promote personal growth and learning, but take away from work progress. Considering the qualitative remarks, comparisons may be particularly high for attorneys working their way up in an organizational hierarchy or with high standards for billable hours.

The general lack of relationships for stress as achievement with study outcomes was an interesting finding to highlight. Prior use of this subscale by Black and Britt (2022) did find this subscale to relate less consistently to health outcomes than the other subscales. It could be that stress as achievement is not as purely maladaptive as other beliefs of the need to compare or display one’s stress. Some attorneys may indeed view stress as a part of achievement, but also have good coping abilities that allow them to still manage their stress well. Further investigation into moderators, like coping skills or available resources, are warranted for more insight.

Adding to prior research that has correlated stress stigma with burnout (Klinefelter et al., 2021), we found stress stigma from oneself and others to be a consistent correlate with recovery- and stress-related outcomes. In support of Hypothesis 2, those who felt stigma around their stress (both personally and from others) reported higher levels of perceived stress and WFC, recovered less often and, when engaging in recovery, felt more remorse. Even when controlling for gender, organizational tenure and quantitative workload, stress stigma from oneself and from others correlated with relaxation remorse, WFC, perceived stress, and two of the four recovery experiences. The exceptions were that both stress stigma subscales did not remain significant correlates of detachment and mastery once accounting for the control variables. This may be similarly driven by the generally low detachment among our sample. Non-significant correlations with mastery are less clear, but it may be that these perceptions of stigma around having too much stress are less impactful than comparisons to others, which may more directly impact behavioural choices in how one spends leisure time for hobbies.

In our most stringent analysis, socially derived views of stress (i.e. stress-related social comparisons, stress stigma from others) correlated with less control of non-work recovery time likely as a result of participants worrying about how busy they should be based on perceptions of their co-workers. The multiple regression analyses also highlighted the strong contribution of stress-related impression management and stigma from oneself as a contributor to relaxation remorse. These beliefs that one needs to display an image of working hard and feelings of insufficiency when stress cannot be adequately handled may be particularly hard to overcome during recovery time, making good relaxation very difficult. Overall, these data demonstrate that while the amount of work is a very important stressor, the subsequent harmful internalization of stress badge perceptions can permeate several domains of non-work life and affect one’s ability to ever truly “leave the office” and recover at the end of the day. This finding is well situated within the literature, as Sonnentag and colleagues (2010) found that when job demands are too high, recovery is negatively affected.

When considering individual differences in gender predicted in Hypothesis 3, women attorneys had significantly higher levels of stress-related comparisons, stigma around their stress, and recovered less than men. Interestingly, women reported feeling self-stigma more frequently than stress stigma from others, which could be a result of higher internalization of cultural norms about the perceived acceptability of their own personal stress levels. Some of these pressures may be rooted in a greater need to “prove” themselves as a woman in the field, as expressed in open-ended remarks. It is also notable that men and women attorneys felt differing levels of stress-stigma and recovery, but the perceptions of quantitative workload, perceived stress, and even stress as achievement did not differ significantly. These non-significant differences suggest that even in the face of similar demands and attitudes toward high stress, women may experience more internalization and negative recovery outcomes.

Consistent with Hypothesis 4, we found that tenure within the organization negatively correlated with quantitative workload, perceived stress, stress comparison, stress stigma and WFC, while positively correlating with recovery experiences. So, as attorneys gain experience, rank and reputation within the firm, they may objectively have less work and subjectively conform less to harmful attitudes regarding stress, both of which can help decrease WFC. Interestingly, stress-related impression management was similar across genders and tenure levels, suggesting that these feelings toward displaying stress may be similar, regardless of these demographics. Control also did not vary across demographic variables, suggesting that control of leisure time may be driven by other factors, such as one’s work schedule or family situation.

Practical implications

Our findings have several important practical implications. First, our findings reinforce numerous other studies that document the very high stress and workload of attorneys. This high workload can be a major barrier to efforts to recover from workplace demands and maintain personal well-being. These findings point to the importance of stress-management interventions for attorneys. Such interventions may need to be appropriately tailored for this group, where many attorneys may be very motivated by their work, at the neglect of their personal health. Our findings, along with previous findings on the potential negative effects of over-engagement (Nickum & Desrumaux, 2022) and the limited effectiveness of stress-management strategies for reducing psychological strain (Brough & Boase, 2019), highlight the need to help attorneys not only understand how to manage stress but also why to manage their stress. Arguments for stress management can leverage attorneys’ high motivation toward their work by showing how taking time to recover can actually enhance, rather than detract from their performance.

Even more, interventions aimed at changes to the organizational and occupational cultures may be necessary to allow for more lasting changes. Targeted programs aimed at reducing stress, provided at the participants’ place of work, have been found to result in less fatigue and rumination (e.g. Querstret et al., 2016). The benefits are likely from the program itself and the support shown by the organization by facilitating de-stressing programs during the workday, as opposed to during the employees’ free time. Even prior to an intervention, targeting attitudes and norms around stress and the importance of recovery first may allow individual stress-management techniques to be utilized to a greater degree.

Several common practices mentioned in the qualitative remarks, such as pressures to be always available or to meet high billable hours, may reinforce stress as normative. Cultural practices that de-emphasize being “stressed” as the norm could include initiatives to limit work email usage after the workday via explicit rules or cultural norms, or even simple practices of accountability for taking time for self-care. As additional examples, some Australian law firms provided a wellness day to take time off in the midst of the pandemic (Niesche, 2021). Efforts to encourage attorneys to try take time off or recovery-supporting behaviours (e.g. requiring an out-of-office reply be set up during a day off), even if employees do not initially see the value, may result in attitude change over time. This idea is rooted in Self-Perception Theory (Bem, 1973), where individuals can understand their attitudes in light of their behaviours (e.g. if I temporarily psychologically detach from work and notice I feel less stressed, I may begin to have a more positive reaction to recovery efforts). Changes to organizational culture and practices are most successful when supported by leaders; thus, it will likely be most successful to engage with partners of firms to initiate and exemplify change.

Interventions may be particularly valuable for women and attorneys newer to a firm. Our findings, and previous research, suggest significant burdens for women attorneys, both observable (e.g. feeling the need to “prove” oneself to clients or partners) and internal (e.g. mental, emotional), as well as within and outside the physical workplace (e.g. Anker & Krill, 2021; Dinovitzer et al., 2009). When attempting to change organizational culture regarding stress and creating or strengthening work–life balance, emphasis should be placed on the particular challenges facing attorneys who are women, or, more broadly, gender, sexual and racial minorities, as well as young attorneys who are newer to the field. Creating added social support systems for such attorneys may help ameliorate the additional burdens faced by these groups, creating a healthier and more inclusive workplace. Unique support systems and efforts may also be needed, based on the area of practice. For instance, attorneys faced with a particularly high level of trauma in their work (e.g. asylum attorneys; Harris & Mellinger, 2022) may have unique needs for cultures that normalize talking about the heavy psychological demands of their work and providing appropriate support.

Limitations and future directions

We acknowledge several limitations in the present study that should be considered in interpreting and applying our findings. The study consists only of self-report data, which could be affected by participants wishing to answer in socially desirable ways. Due to stigma surrounding stress, some participants may have under-reported levels of stress, as well as negative effects. While the study was relatively equal in gender representation, the racial makeup was overwhelmingly white (92%). In the future, researchers should focus more on racial minority members, who may experience high stress and unique stressors within the legal field (ABA, 2021). Overall, future research should apply an intersectional lens to examine how multiple minoritized identities can affect workplace experiences.

Second, there may be self-selection bias concerning which participants chose to take part out of all of the attorneys who received the invitation. While we did receive diverse responses in terms of area of practice and time spent in practice, it is very likely that the most busy and stressed attorneys may not have taken the time to reply to the survey. Therefore, many of our findings could be a conservative estimate of some of the connections between stress attitudes and recovery. Related, there was some attrition, where a number of respondents completed open-ended questions at the start of the survey but did not complete all of the self-report quantitative measures. This may again introduce a bias into our results based on the characteristics of those who completed the full survey.

Lastly, we acknowledge that our sample was composed of attorneys from a single US state. Therefore, additional studies should replicate our results with attorneys from different regions of the US and other countries to account for differences in cultural norms. Studies may also want to venture into related but under-researched occupational groups in the legal profession. For instance, examining stress attitudes and recovery among the judiciary could be particularly relevant, given the unique stressors in these jobs. Despite calls to understand stress among judges (Chamberlain & Miller, 2008), there is little research in this area.

While not necessarily a limitation, the responses were impacted by the ongoing COVID-19 pandemic, which created a new realm of stress for the majority of participants. While it is well documented that stress had been high among attorneys long before the pandemic, future research is needed to understand the role of stress-related attitudes on recovery during more typical work periods. We also encourage other researchers to consider a broader range of variables that may limit recovery among attorneys. While stress-related stigma seemed to impact recovery experiences, the stress badge perceptions had fewer significant connections. Given a number of non-significant relationships with detachment in particular, correlates of detachment warrant further attention among attorneys, given the health benefits of detaching from work (Sonnentag & Fritz, 2007). Future research can determine whether detachment is equally as important in this population, where detachment may be difficult. If so, other drivers of detachment, such as individual personality differences, work arrangement characteristics, work culture, or perhaps non-work support could be explored further.

Conclusions

Prior research established that workplace stress can impact the well-being of attorneys, but little attention has been paid to attorney’s efforts to recover from demands. Attorneys may particularly struggle with the effects of stress because of the prevalence of expectations that being stressed and neglecting recovery are to some degree normative. Our findings suggest that stigma around stress and views of stress as a badge of honour can negatively impact recovery outside work and interactions between work and family. Future research and practice efforts must consider how to implement successful stress-management interventions in such contexts in order to promote healthy views of stress among high-achieving and high-stress occupations.

Funding Statement

This research was supported by an internal grant [University of Tennessee at Chattanooga Pre-Tenure Enhancement grant] provided to the second author.

Ethical standards

Declaration of conflicts of interest

Luke Wiley has declared no conflicts of interest.

Kristen Black has declared no conflicts of interest.

David Ross has declared no conflicts of interest.

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional research committee [University of Tennessee at Chattanooga; IRB protocol number IRB #20-112, designated as exempt] and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed consent

Informed consent was obtained from all individual participants included in the study.

Data availability statement

The data that support the findings of this study are available from the corresponding author, Kristen Black, upon reasonable request.