Work–Life Balance in Academic Medicine: Narratives of Physician-Researchers and Their Mentors

Erin A. Strong, MS1, Rochelle De Castro, MS2,3, Dana Sambuco, MPA2,3, Abigail Stewart, PhD4,5, Peter A. Ubel, MD6,7, Kent A. Griffith, MS8, and Reshma Jagsi, MD, DPhil2,3

1University of Michigan Medical School, Ann Arbor, MI, USA; 2Center for Bioethics and Social Sciences in Medicine, University of Michigan, Ann Arbor, MI, USA; 3Department of Radiation Oncology, University of Michigan, Ann Arbor, MI, USA; 4Women’s Studies Program, University of Michigan, Ann Arbor, MI, USA; 5Department of Psychology, University of Michigan, Ann Arbor, MI, USA; 6Fuqua School of Business, Duke University, Durham, NC, USA; 7Sanford School of Public Policy, Duke University, Durham, NC, USA; 8Biostatistics Unit, University of Michigan Comprehensive Cancer Center, Ann Arbor, MI, USA.

BACKGROUND: Leaders in academic medicine are often selected from the ranks of physician-researchers, whose demanding careers involve multiple professional commitments that must also be balanced with demands at home. OBJECTIVE: To gain a more nuanced understanding of work–life balance issues from the perspective of a large and diverse group of faculty clinician-researchers and their mentors. DESIGN: A qualitative study with semi-structured, in- depth interviews conducted from 2010 to 2011, using inductive analysis and purposive sampling. PARTICIPANTS: One hundred former recipients of U.S. National Institutes of Health (NIH) K08 or K23 career development awards and 28 of their mentors. APPROACH: Three researchers with graduate training in qualitative methods conducted the interviews and thematically coded verbatim transcripts. KEY RESULTS: Five themes emerged related to work–life balance: (1) the challenge and importance of work–life balance for contemporary physician- researchers, (2) how gender roles and spousal dy- namics make these issues more challenging for women, (3) the role of mentoring in this area, (4) the impact of institutional policies and practices intended to improve work–life balance, and (5) per- ceptions of stereotype and stigma associated with utilization of these programs. CONCLUSIONS: In academic medicine, in contrast to other fields in which a lack of affordable childcare may be the principal challenge, barriers to work–life balance appear to be deeply rooted within profes- sional culture. A combination of mentorship, inter- ventions that target institutional and professional culture, and efforts to destigmatize reliance on flexibility (with regard to timing and location of work) are most likely to promote the satisfaction and success of the new generation of clinician-research- ers who desire work–life balance.

KEY WORDS: gender; work–life balance; academic medicine; medical profession; qualitative research.

J Gen Intern Med 28(12):1596–603

DOI: 10.1007/s11606-013-2521-2

© Society of General Internal Medicine 2013

BACKGROUND

A growing number of recent medical school graduates, both men and women, seek balance between their professional and personal lives.1–3 This trend may be influenced by broader societal changes, such as the modern movement of women into the paid labor force and the increasing expectation that men participate in parenting and house- work.2,4,5 Men and women in the current generation of physicians in particular have been found to possess a strong sense of responsibility for family life and parenting, and thus, may experience discontent or tension if work demands impinge upon a more a balanced lifestyle.6–10 These developments within the medical profession are impor- tant to understand, in part because of their potential impact upon patients, whose experiences and outcomes depend upon a dedicated cadre of clinicians and researchers.11,12

Women, who now constitute a substantial and growing proportion of the physician workforce,13–15 may face unique challenges in balancing the competing demands of professional duty and family responsibility.11,16,17 Com- pared to other faculty, women faculty with children have been found to be more likely to perceive problems with institutional policies affecting work–life balance, such as meetings held after hours or a lack of on-site childcare.16

Prior evidence suggests a paucity of role models in academic medicine who have successfully managed both career and family life,11,17 which may partly explain some women’s attrition from academic medicine and failure to succeed at the same rate as their male colleagues.16,18–22

Thus, gender differences in experiences related to work–life balance are particularly worthy of further investigation.

Received December 13, 2012 Revised May 13, 2013 Accepted May 31, 2013 Published online June 14, 2013

1596

OBJECTIVE

Leaders in academic medicine are often selected from the ranks of physician-researchers, whose demanding careers involve commitments to research, clinical care, teaching, and administrative responsibilities that must also be balanced with demands at home. In order to gain a more nuanced understanding of work–life balance issues from the perspective of a large and diverse group of faculty clinician- researchers and their mentors, we conducted a qualitative analysis of interviews with men and women who received prestigious K-series career development awards from the NIH and their academic mentors. In this way, we sought to provide thick description of how promising clinician- researchers of both genders feel about their competing responsibilities and how they perceive institutional climate, policy, and practice related to work–life balance. We deliberately selected recipients of K08 and K23 awards, because these awards are made to individuals who hold clinical doctorates and so are likely to face multiple competing demands at work that must, in turn, be balanced with responsibilities at home.

DESIGN & PARTICIPANTS

After approval by the University of Michigan institutional review board (IRB), we conducted in-depth, semi-structured interviews with selected recipients of K08 and K23 awards from the NIH between 1997 and 2009, as well as their academic mentors. Purposive sampling was utilized to collect a diverse set of viewpoints. We included both men and women, with oversampling for racial and ethnic minorities. We chose individuals from a variety of medical specialties and academic institutions. We selected those who remained at their original institution at time of K award, those who had changed institutions, and those who had left academic positions (as determined by internet searching), as well as those who had gone on to attain R01 funding and those who had not (as determined through the NIH RePORTER database).23

The interview protocol included closed and open-ended questions relating to a number of domains, including questions about the participant’s family structure, the effects of family or personal demands upon career success, and relevant institu- tional policies and practices. Three researchers with graduate training in qualitative

methods conducted telephone interviews with 128 respond- ents, at which point the criterion of thematic saturation was met. Of the 128 participants, 54 were members of matched mentor–mentee pairs. Tables 1 and 2 display the character- istics of all study participants. The tape-recorded interviews were transcribed verbatim

by an independent professional transcriptionist. The average interview lasted 52 min.

APPROACH

One of the three researchers independently reviewed and coded all transcripts using standard techniques of thematic analysis,24 and QSR NVivo Version 8.0.332.0 SP4 (Don- caster, Victoria, Australia) software. Some potential codes were anticipated; others were generated de novo. After independent coding, the researchers discussed identified themes and quotations coded as exemplary. Validity was

Table 1. Participant Characteristics (n=128)

Interviewee type K-Awardee 100 (78 %) Mentor 28 (22 %)

Gender Women 76 (59 %) Men 52 (41 %)

Race-Ethnicity* White/Caucasian 99 (77 %) Black/African-American 7 (5 %) Hispanic/Latino 3 (2 %) Asian/Asian-American 18 (14 %) Not reported 5 (4 %)

*Percentage exceeds 100 because some participants were allowed to report more than one race/ethnicity

Table 2. K-Awardee Characteristics (n=100)

K Award type K08 38 K23 62

Degree type MD or DO 56 MD or DO with PhD 16 Other clinical doctorate 28

Specialty Medical 40 Surgical 3 Families, women, or children 15 Hospital-based 13 Other 28 Not reported 1

Institution type (at time of K award) Public 45 Private 54 Nonprofit 1

Received R01 or equivalent funding Yes 38 No 62

Current career status Academic institution 80 Government 2 Independent research institution 1 Industry 7 Nonprofit 3 Private practice 7

Marital status Divorced 4 Living with partner 3 Married 80 Separated 1 Unmarried 10 Not reported 2

Parental status Yes 81 No 14 Not reported 5

1597Strong et al.: Work–Life Balance in Academic MedicineJGIM

established through investigator triangulation.25 Differences in interpretations were regularly discussed and arbitrated. An iterative process was employed whereby major themes were coded and discussed, followed by minor themes. Six major thematic clusters emerged, one of which

related to time and balance. The present analysis focuses on the major theme of work–life balance in academic medicine, which fell within this cluster. The other thematic clusters were mentoring26; rejection and resilience27; nego- tiation and resources28; unequal treatment, conflict, and discrimination; and goals and aspirations. Of note, many of the identified themes were cross-cutting in nature. In addition, a brief quantitative analysis was performed.

The percentage of words devoted to work–life balance was calculated for each interviewed subject. Comparisons were made between genders for both awardees and mentors separately using the Wilcoxon rank-sum test statistic, as well as between individuals receiving their K awards between 1997 and 2003 (an “early cohort”) and those who received awards since 2004 (a “more recent cohort”). In addition, for the K awardees, we assigned a code to each transcript, indicating whether or not work–life balance issues appeared to be a significant personal concern, and we then compared whether the frequency of this code differed by gender.

RESULTS

The final data set included transcripts totaling 1,108 pages (513,730 words, excluding interview questions). Among K- award recipients, the median percentage of words devoted to discussions of work–life balance was greater for females than males (19.1 % vs. 8.9 %, p<0.0001). Among the smaller sample of mentors, a similar pattern was evident, but the gender difference was not statistically significant (19.4 % of female mentors’ words vs 11.5 % of male mentors’ words, p=0.09). Overall, 63 % of female K award recipients we interviewed indicated that work–life balance was a significant personal concern, compared to only 33 % of men (p=0.01). In the early cohort of K awardees, there were 26 (55 %) female

and 21 (45 %) male K awardees. In the more recent cohort, there were 41 (77 %) female and 12 (23 %) male K awardees. Overall, regardless of gender, there was a trend (p<0.10) toward a higher percentage of words related to work–life balance in the more recent cohort (median=17.3 % vs 11.7 %, p=0.068), Five emergent subthemes related to work–life balance

were identified: (1) the challenge and importance of work– life balance for contemporary physician-researchers, (2) how gender roles and spousal dynamics make these issues more challenging for women, (3) the role of mentoring in this area, (4) the impact of institutional policies and practices, especially with regard to the inherent flexibility

of academic work, and (5) perceptions of stereotype and stigma associated with utilization of programs intended to improve work–life balance.

The Challenge and Importance of Work–Life Balance for the Physician-Researcher

Both men and women discussed work–life balance as an important challenge, particularly when parenting young children.

“Work–life balance was a struggle the whole time my kids were at home…now that they are both in college…I still make trade-offs between having a social life and my professional work, but they are less acute…” (Male K-Awardee)

Some respondents described a generational difference in expectations and perspective, with the new generation of early career physicians, both men and women, more concerned about spending time with family and less willing to allow their work to be their sole priority.

“… I’m in the first generation of people that refuse to do it the way all the older men have done it who were successful, and it has been difficult to do it differently. It was important for me to be home most evenings for dinner with my kids and see them go to bed…. I don’t know that was a priority for those who were mentors to me, so I…felt like I didn’t have very good role models myself.” (Female, K-Awardee)

Women, in particular, indicated that work–life balance was integral to their satisfaction and part of their definition of success.

“The thing that I prize the most, and I know I share this with other women in academic medicine, is really being able to have both the family and the successful career…. it’s crazy busy with the two, but without the two, it just wouldn’t be as satisfying and fulfilling.” (Female, K-Awardee)

Societal Expectations, Gender Roles, and Spousal Dynamics

A number of respondents felt that work–life balance issues were especially challenging for women, in part due to societal gender roles.

“…I think it’s the social conditioning…. It’s a social thing that we expect women to be more into the child raising thing and men to be less…” (Male, K-Awardee)

1598 Strong et al.: Work–Life Balance in Academic Medicine JGIM

Both male and female respondents noted a tendency for women to feel distressed, guilty, or judged when faced with the competing expectations of career and motherhood.

“…perhaps guilt that comes with not being there…is felt by women more than men…I do believe there’s a lot of angst and worry that comes with not being there for your kids, working…” (Male, Mentor)

A number of respondents commented on the role of the spouse, noting that those who had stay-at-home spouses were advantaged, whereas those who were members of dual-career couples were more challenged, particularly when both careers were of equal status.

“I’m just thinking of my own parents’ marriage, where the husband’s career…comes first and the wife sees herself as…supporting her husband’s career…. The few other successful scientists I see, they often have little ‘wifey-poo’ at home who, if they have to stay until 9 o’clock to finish writing something, “Oh, no problem. I’ll give the kids a bath. I’ll do this…” That’s just not…what my life is. My wife comes home, and…she’s exhausted. She needs me to come home…. It’s just difficult if you don’t have a stay-at-home spouse.” (Male, K-Awardee)

Several women commented on the importance of being able to trust their husbands to contribute equally to the household and childrearing responsibilities.

“…My husband is very supportive…I’ve sort of from the beginning kind of made it clear that he actually has to take responsibility and not just help and he does…so I feel like I’m lucky in that way.” (Female, K-Awardee)

The Role of Mentoring in Achieving Balance

A number of female respondents commented that they had modeled themselves after, or had received support and advice, from female mentors or senior women who understood the challenges of having to respond to the demands of both a career and a personal life.

“I think I have learned from my mentors. In particular, two female mentors…I’ve learned a lot in terms of working to be as productive as you can be with your academic career…but realizing that you still have to fit in the rest of your life as well, especially being a woman….” (Female, K-Awardee)

Others perceived a lack of senior women who could role model successful work–life balance.

“…this is still a field where there aren’t a lot of women in general and I think less than half the women in our

faculty have kids, so I feel like we’re still learning how to do this without a lot of role models. I think it’s very different to be a mom with three kids than to watch the dads with three kids…” (Female, K-Awardee)

One female respondent indicated that she felt disadvantaged despite having supportive male mentors, mainly because they could not truly relate to or give appropriate advice concerning the work–life balance issues that she personally faced.

“…I think my long-term mentor and my chairman… they have had a fundamentally different work–life balance because they both had wives who stayed home full-time and took care of their kids. And so even though they are at least emotionally and psychologi- cally supportive…they weren’t able to give me really strong advice about how to do that because they never had to deal with it.” (Female, K-Awardee)

Several female respondents noted specifically that work– life balance was an issue for both men and women, and they considered advising on this subject to be an important mentoring role.

“I think both men and women these days are interested in trying to balance work and personal life so that is often what becomes part of the goal is ‘How do you develop a sustainable balance between work and life, family life, or other parts of yourself?” (Female, Mentor)

Of note, certain male mentors also described feeling a responsibility to promote work–life balance and remain sensitive to generational and gender differences in particular.

“…Early parts of child rearing are particularly challenging for women…. You have to say, “You’re a woman…let’s talk about when you have babies… let’s plan how it’s going to affect what’s going on during the course of your career development….”… As a male in particular, you have to be compassion- ate because you can’t pretend that you understand all of it because you don’t as a male…” (Male Mentor)

Institutional Policies and Practices: The Flexibility of Academic Work

When discussing institutional policies and practices related to the maintenance of work–life balance, respondents described various approaches towards scheduling, modified workloads, and leave.

“That is definitely a strength of my department…a specific person has had a child and wants to go to half

1599Strong et al.: Work–Life Balance in Academic MedicineJGIM

time, we figure out how to do that. There is a formal mechanism for extending tenure track for those who have family reasons for doing that…my department is very proactive; that’s kind of the highest priority thing for faculty…” (Male, K-Awardee)

Some described having express support from their institution for utilizing flexible scheduling.

“I think they are supportive both on paper and also in reality. So, for example, allowing me to work from home. I didn’t have to do it under the table like I’ve heard some people do. We actually had it all written out what days I would be here and what days I would be there….” (Female, K-Awardee)

Other respondents indicated that they relied upon the flexibility of academic scheduling on an ad hoc basis only, or even deliberately hid this aspect of their time manage- ment approach from others.

“You don’t want to…tell too many people when your kids are sick or if you have to sneak out to something at their school…Truthfully, I feel like when guys say they need to go to a soccer game and they do that everybody thinks they’re…incredible; women, nobody’s going to pat you on the back when you have to run out to do something…you just either suck it up and don’t care what people think or…edit where you are going and what you’re doing sometimes.” (Female, K-Awardee)

Stereotypes and Stigma

The individuals selected for this study were high-achieving individuals who have previously demonstrated their commit- ment to academic careers by attaining prestigious K-series career development awards. Nevertheless, when discussing gender discrimination or bias, a number of female respondents felt that women were more likely to be perceived as prioritizing family over work, even when this was not accurate.

“Thereis…justsortofanattitudeabout…whatthewomen’s priorities really are and what happens to women when they have babies…there is some willingness to accommodate, but there is also just a lot of pressure to show that…it’s not interfering with my work….” (Female, K-Awardee)

Some also expressed the concern that even when women make the same attempts to balance work and home activities as men, women’s actions were viewed differently.

“When a female faculty says, I need to leave at five o’clock, she’s looked on differently than when a

male faculty says I need to leave at five o’clock…. The male faculty leaving at five o’clock to go to soccer practice is thought of to be a great father… whereas, the female faculty saying I need to leave at five o’clock to drive my son to soccer, is looked on as she’s not serious; she’s on a mommy track….” (Female, K-Awardee)

One male mentor observed a perceived risk in fostering the careers of women based on the assumption that all women have the potential to become mothers and change their priorities.

“I think there is an innate fear in all mentors that…a woman has a higher likelihood of dropping out of full-time academia and so that whole work that was put into moving that person’s career forward can be stopped or stunted at any given time, whereas it is much less likely with a man.” (Male, Mentor)

As a result, even when encouraged by their superiors to take time off for family or to utilize programs and other sources of support, women sometimes described being hesitant to do so for fear of projecting this stereotypic image to their colleagues, senior faculty, and administrators.

“…My clinical division director…was the one who said…“it’s OK to leave and go to your children’s performances or plays or whatever in the middle of the day.” …I didn’t think that was OK. … I was the first woman in our [department]…so there were a lot of ways in which I didn’t want to rock the boat; I didn’t want to be seen as not pulling my weight.” (Female, K-Awardee)

“I don’t want to call myself out as someone who needs [additional support]…the men I’m working with don’t ever perceive or vocalize family issues are a problem…that they need that time to be better dads…. Because then it’s just one more thing that makes it harder to compete when you’re trying to move up the ladder…” (Female, K-Awardee)

DISCUSSION

Participants in this large qualitative study of NIH career development award recipients and their mentors vividly described the ways in which they experienced and managed challenges related to work–life balance. Their views support a number of observations of previous studies concerning the impact of both gender and generation and provide additional insights.

1600 Strong et al.: Work–Life Balance in Academic Medicine JGIM

Some participants discussed the arrival of a new generation of physicians who hold different values than their predecessors concerning their professional and private lives.6–10 Male physicians who are more active in family life, particularly those whose wives also work, appear to face difficulties that are not generally encountered by men with stay-at-home wives. Several of our respondents, both men and women, observed that important accommodations are sometimes limited by assumptions that faculty members can rely on support from stay-at-home spouses or spouses who only work part-time. Consistent with the existing literature on this subject, many

respondents also recognized that women, particularly those with children or other family responsibilities, face unique challenges.9,11,16,29–31 Although both men and women prior- itized personal and family life, gendered societal expectations of women’s roles, within and outside the workplace, continue to have a substantial impact. Previous studies have shown that dual-career couples generally adhere to traditional gendered arrangements.32 Consistent with evidence that female medical faculty spend more time on child care than their male colleagues,16 our findings show that many women in academic medicine feel they shoulder a majority of the family responsibilities despite their demanding full-time careers. Considerable work has evaluated ways in which working

women can balance roles at work and at home. Although policy-makers have promoted initiatives to promote affordable childcare,33–36 the physicians in our study had relatively high incomes, and securing childcare was generally not their concern. Instead, participants, particularly female, tended to feel guilt over not spending enough time with their families while simultaneously feeling a need to further develop their research careers. Prior literature has suggested cultural expect- ations may lead women to feel particularly distressed about work–life conflict and that advancement for women is hindered by time constraints and a “rigid workcentric culture,” with inflexible academic routines and promotion processes.16,37–39

Of note, these findings should not be mistaken as a justification for why women should not be expected to succeed in academic medicine. The women in this study did not generally articulate choosing or prioritizing family over career. Our participants were selected because of their demonstrated commitment to their careers, and the conflicts they described did not seem to arise due to a lack of commitment to career or “opting out,” but rather a profound articulated commitment to both career and family. Given growing concerns about the adequacy of the physi-

cian-scientist pipeline in the context of the feminization of the profession, understanding these challenges is an important step in developing structural, institutional, and individual interven- tions that can facilitate flexibility and thereby promote the career success of both women and men.40 Prior evidence suggests that there are perceived problems with institutional policies that limit flexibility.16 Our study builds upon such work by demonstrating that although programs addressing challenges

for women in academic medicine and options for flexible schedules were generally felt to have value, several respondents indicated a further need to destigmatize their utilization. It is problematic that concerns about conforming to gender stereo- types were barriers to the utilization of certain institutional programs and policies intended to assist in this area. However, these concerns are not unfounded. Considerable research in the fields of management and psychology has shown that both men and women are perceived as less committed to their work after utilizing parental leave policies or even simply becoming a parent.41–48 Policies alone are insufficient unless institutions actively promote a culture that allows their utilization.49

Workplace flexibility has been defined as workers’ ability to “choose” when, where, and for how long they engage in work- related tasks.50 Our current findings suggest that it is still quite difficult to “choose” flexibility in an academic environment, particularly when stereotypes and stigma act as barriers to policy and program use. Of note, Villablanca and colleagues identified numerous other reasons for lack of policy use among biomedical sciences faculty, including concerns regarding heavier service loads, undue burden on colleagues, and inability to stop work on grant funded projects/research.51 In addition, Welch and colleagues noted that faculty at academic health centers in particular may be subject to conflicts between the policies of the parent university system, the partner hospital(s), and the faculty practice plan(s).52 Additional research is necessary to further explore such challenges. In the current study, female mentors with families served as

significant role models for both men and women. Prior studies have shown that there is a need for mentors in academic medicine who value having both a career and a family life and who can provide inspiration and advice based on personal experiences.17,53 In fact, the lack of mentors to serve in this role has been identified as prominent in the choice to leave academic medicine.17 In the current study, respondents expressed appreciation for mentors who could relate to them and who shared their desire for both meaningful careers and fulfilling personal lives. Some data suggest that different generations may have different values and/or strategies when balancing the competing demands of work and home.6–10

Future research is warranted to further explore this issue. This study has a number of strengths, including its rich

detailed narratives and approach to analysis, which conforms to standards for rigorous qualitative research.25,54 Our study was large for a qualitative study, including 128 respondents; the mean sample size for comparable studies published in top tier medical journals since 2000 is 36.55 The qualitative approach does have certain inherent limitations, however. Many social scientists would argue that the insights from qualitative studies should be viewed as complementary to those generated from larger survey studies, which are, in turn, limited in the depth to which they may explore participants’ experiences. By purposively sampling to ensure a diversity of viewpoints and continuing accrual until meeting the criterion

1601Strong et al.: Work–Life Balance in Academic MedicineJGIM

of thematic saturation, concerns about the generalizability of our findings should be minimized. Of note, we focused on individuals who had received, or who had mentored recipients of, highly selective NIH K-awards, and hence our results may not be generalizable to those who have different interests or aptitude. Still, we believe that the insights gleaned from this particular population have broader relevance for faculty seeking careers in academic medicine more generally. In sum, this study provides important insights about the ways

in which concerns about work–life balance affect those pursuing careers in academic medicine. Institutional leaders and policy- makers should carefully consider the insights offered by our participants, particularly regarding the concerns about gender stereotyping and the stigma associated with utilizing certain policies intended to provide support. A combination of mentor- ship, interventions that target institutional and professional culture, and efforts to destigmatize reliance on the flexibility of academic careers, is most likely to promote the satisfaction and success of the new generation of clinician-researchers who desire balance in their careers and their family lives.

Acknowledgements: Contributors: The authors wish to thank the K-award recipients and mentors who took the time to participate in this study.

Funding/Support: This work was supported by Grant 5 R01 HL101997-04 from the National Institutes of Health to Dr. Jagsi. The funding body played no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; or preparation, review, or approval of the manuscript.

Prior presentations: None.

Conflict of Interest: The authors declare that they do not have a conflict of interest.

Corresponding Author: Reshma Jagsi, MD, DPhil; Department of Radiation Oncology, University of Michigan, UHB2C490, SPC 5010; 1500 East Medical Center Drive, Ann Arbor, MI 48109-5010, USA (e- mail: [email protected]).

REFERENCES 1. Dorsey ER, Jarjoura D, Rutecki GW. Influence of controllable lifestyle

on recent trends in specialty choice by US medical students. JAMA. 2003;290:1173–8.

2. Dorsey ER, Jarjoura D, Rutecki GW. The influence of controllable lifestyle and sex on the specialty choices of graduating U.S. medical students, 1996–2003. Acad Med. 2005;80:791–6.

3. Gelfand DV, Podnos YD, Wilson SE, Cooke J, Williams RA. Choosing general surgery: insights into career choices of current medical students. Arch Surg. 2002;137:941–7.

4. Waite LJ, Nielsen M. The rise of the dual-earner family, 1963–1997. In: Hertz R, Marshall NL, eds. Working Families: the Transformation of the American Home. Berkeley: University of California Press; 2001:23–41.

5. Bianchi SM, Milkie MA, Sayer LC, et al. Is anyone doing the housework? Trends in the gender division of household labor. Social Forces. 2000;79:191–228.

6. Bickle J, Brown A. Generation X: implications for faculty recruitment and development in academic health centres. Acad Med. 2005;80:205–10.

7. Howell LP, Servis G, Bonham A. Multigenerational challenges in academic medicine: UC Davis’s response. Acad Med. 2005;80:527–32.

8. Howell LP, Joad JP, Callahan E, Servis G, Bonham AC. Generational forecasting in academic medicine: a unique method of planning for success in the next two decades. Acad Med. 2009;84:985–93.

9. Jovic E, Wallace JE, Lemaire J. The generational and gender shifts in medicine: an exploratory survey of internal medical physicians. BMC Heal Serv Res. 2006;6:55.

10. Lowenstein SR, Fernandez G, Crane LA. Medical school faculty discontent: prevalence and predictors of intent to leave academic careers. BMC Med Educ. 2007;7:37.

11. Levinson W, Tolle SW, Lewis C. Women in academic medicine. Combining career and family. N Engl J Med. 1989;321:1511–7.

12. Sibert K. Don’t Quit This Day Job. New York Times. 2011;June 11. Available at: http://www.nytimes.com/2011/06/12/opinion/12sibert.html?pagewanted=all. Accessed November 2, 2011.

13. Levinson W, Lurie N. When most doctors are women: what lies ahead? Ann Intern Med. 2004;141:471–9.

14. Jagsi RJ. Women’s leadership in the development of medicine. In: O’Connor K, ed. Gender and Women’s Leadership: a Reference Hand- book. Thousand Oaks: Sage; 2010:720–31.

15. American Association of Medical Colleges. (2011) Women in U.S aca- demic medicine and science: statistics and benchmarking report, 2009– 2010. Available at: https://www.aamc.org/members/gwims/statistics/. Accessed November 2, 2011.

16. Carr PL, Ash AS, Friedman RH, et al. Relation of family responsibilities and gender to the productivity and career satisfaction of medical faculty. Ann Intern Med. 1998;129:532–8.

17. Levine RB, Lin F, Kern DE, Wright SM, Carrese J. Stories from early- career women physicians who have left academic medicine: a qualitative study at a single institution. Acad Med. 2011;86:752–8.

18. Jagsi R, DeCastro R, Griffith KA, et al. Similarities and differences in the career trajectories of male and female career development award recipients. Acad Med. 2011;86:1415–21.

19. Jagsi R, Guancial EA, Worobey CC, et al. The “gender gap” in authorship of academic medical literature — a 35-year perspective. N Engl J Med. 2006;355:281–7.

20. Jagsi R, Motomura AR, Griffith KA, Rangarajan S, Ubel PA. Sex differences in attainment of independent funding by career development awardees. Ann Intern Med. 2009;151:804–11.

21. Jagsi R, Tarbell N, Henault LE, Chang Y, Hylek EM. The representa- tion of women on the editorial boards of major medical journals: a 35- year perspective. Arch Intern Med. 2008;165:544–8.

22. Pohlhaus JR, Jiang H, Wagner RM, Schaffer WT, Pinn VW. Sex differences in application, success, and funding rates for NIH extramural programs. Acad Med. 2011;86(6):759–67.

23. National Institutes of Health Research Portfolio Online Reporting Tools (RePORT). Reports, data, and analyses of NIH research activities. http:// projectreporter.nih.gov/reporter.cfm Accessed December 13, 2012.

24. Braun V, Clarke V. Using thematic analysis in psychology. Qual Res Psychol. 2006;3:77–101.

25. Giacomini MK, Cook DJ. Users’ guides to the medical literature: XXIII; qualitative research in health care A. are the results of the study valid? JAMA. 2000;284:357–62.

26. DeCastro R, Sambuco D, Ubrel PA, Stewart A, Jagsi R. Mentor networks in academic medicine: moving beyond a dyadic conception of mentoring for junior faculty researchers. Acad Med. 2013;88(4):488–96.

27. DeCastro R, Sambuco D, Ubel PA, Stewart A, Jagsi R. Batting 300 is good: perspectives of faculty researchers and their mentors on rejection, resilience, and persistence in academic medical careers. Acad Med. 2013;88:497–504.

28. Sambuco D, Dabrowska A, DeCastro R, Stewart A, Ubel PA, Jagsi R. Negotiation in academic medicine: narratives of faculty researchers and their mentors. Acad Med. 2013;88:505–11.

29. McGuire LK, Bergen MR, Polan ML. Career advancement for women faculty in a U.S. school of medicine: perceived needs. Acad Med. 2004;79:319–25.

30. Reed V, Buddeberg-Fischer B. Career obstacles for women in medicine: an overview. Med Educ. 2001;35:139–47.

31. Sonnad SS, Colletti LM. Issues in the recruitment and success of women in academic surgery. Surgery. 2002;132:415–9.

32. Moen P, Yu Y. Effective work/life strategies: working couples, work conditions, gender, and life quality. Soc Probl. 2000;47:291–326.

33. Connelly R. The effect of child care costs on married women’s labor force participation. Rev Econ Stat. 1992;74:83–90.

1602 Strong et al.: Work–Life Balance in Academic Medicine JGIM

34. Han W, Waldfogel J. Child care costs and women’s employment: a comparison of single and married mothers with pre-school-aged chil- dren. Soc Sci Q. 2001;82:552–68.

35. Kimmel J. Child care costs as a barrier to employment for single and married mothers. Rev Econ Stat. 1998;80:287–99.

36. Ribar DC. Child care and the labor supply of married women: reduced form evidence. J Hum Resour. 1992;27:134–65.

37. Milkie M, Peltola P. Playing all the roles: gender and the work-family balancing act. Journal of Marriage and the Family. 1999;61:476–90.

38. Pololi LH, Jones SJ. Women faculty: an analysis of their experiences in academic medicine and their coping strategies. Gender Medicine. 2010;7:438–50.

39. Yedidia MJ, Bickel J. Why aren’t there more women leaders in academic medicine? The views of clinical department chairs. Acad Med. 2001;76:453–65.

40. Committee on Maximizing the Potential of Women in Academic Science and Engineering, & Committee on science Engineering and Public Policy. Beyond bias and barriers: fulfilling the potential of women in academic science and engineering. Washington, DC;2007.

41. Allen TD, Russell JEA. Parental leave of absence: some not so family friendly implications. J Appl Soc Psychol. 1999;29:166–91.

42. Fletcher J, Bailyn L. Challenging the last boundary: reconnecting work and family. In: Arthur MB, Rousseau DM, eds. The Boundaryless Career: a New Employment Principle for a New Organizational Era. New York: Oxford University Press; 1996:256–67.

43. Hall DT. Moving beyond the mommy track: An organization change approach. Personnel. 1989;66:23–9.

44. Miller A, Tsiantar D. Mommy tracks. Newsweek. 1991;November 25:48–9. 45. Norman N, Tedeschi JT. Paternity leave: the unpopular benefit option.

Pers Adm. 1984;29:39–43.

46. Powell GN. One more time: do female and male managers differ? Acad Manag Exec. 1990;4:68–75.

47. Swiss DJ, Walker JP. Women and the Work/Family Dilemma. New York: John Wiley & Sons; 1993.

48. Kirby E, Krone K. The policy exists but you can’t really use it: communication and the structuration of work-family policies. J Appl Commun Res. 2002;30:50–77.

49. Hewlett SA. Executive Women and the Myth of Having it All. Harvard Business Review. 2002.

50. Hill EJ, Grzywacz JG, Allen S, Blanchard VL, Matz-Costa C, Shulkin S, Pitt-Catsouphes M. Defining and conceptualizing workplace flexibil- ity. Community, Work & Family. 2008;11:149–63.

51. Villablanca AC, Beckett L, Nettiksimmons J, Howell LP. Career flexibility and family-friendly policies: an NIH-funded study to enhance women’s careers in biomedical sciences. Journal of Women’s Health. 2011;20:1485–96.

52. Welch JL, Wiehe SE, Palmer- Smith V, Dankoski ME. Flexibility in faculty work–life policies at medical schools in the big ten conference. Journal of Women’s Health. 2011;20:725–32.

53. Levinson W, Kaufman K, Clark B, Tolle SW. Mentors and role models for women in academic medicine. West J Med. 1991;154:423–6.

54. Giacomini MK, Cook DJ. Users’ guides to the medical literature: XXIII. Qualitative research in health care B. what are the results and how do they help me care for my patients? JAMA. 2000;284:478–82.

55. Yamazaki H, Slingsby BT, Takahashi M, Hayashi Y, Sugimori H, Nakayama T. Characteristics of qualitative studies in influential jour- nals of general medicine: a critical review. Bioscience Trends. 2009;3:202–9.

1603Strong et al.: Work–Life Balance in Academic MedicineJGIM

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

ORIGINAL ARTICLE

For editorial comment, see page 1593; for a related article, se page 1694

From the Division of Hematology (T.D.S.), Division of Primary Car Internal Medicine (L.N.D Division of Biomedical Statistics and Informatic

Affiliations contin the end of this

1600

Changes in Burnout and Satisfaction With Work-Life Balance in Physicians and the

General US Working Population Between 2011 and 2014

e

e .),

s

ued at article.

Tait D. Shanafelt, MD; Omar Hasan, MBBS, MPH; Lotte N. Dyrbye, MD, MHPE; Christine Sinsky, MD; Daniel Satele, MS; Jeff Sloan, PhD; and Colin P. West, MD, PhD

Abstract

Objective: To evaluate the prevalence of burnout and satisfaction with work-life balance in physicians and US workers in 2014 relative to 2011. Patients and Methods: From August 28, 2014, to October 6, 2014, we surveyed both US physicians and a probability-based sample of the general US population using the methods and measures used in our 2011 study. Burnout was measured using validated metrics, and satisfaction with work-life balance was assessed using standard tools. Results: Of the 35,922 physicians who received an invitation to participate, 6880 (19.2%) completed surveys. When assessed using the Maslach Burnout Inventory, 54.4% (n¼3680) of the physicians reported at least 1 symptom of burnout in 2014 compared with 45.5% (n¼3310) in 2011 (P<.001). Satisfaction with work-life balance also declined in physicians between 2011 and 2014 (48.5% vs 40.9%; P<.001). Substantial differences in rates of burnout and satisfaction with work-life balance were observed by specialty. In contrast to the trends in physicians, minimal changes in burnout or satisfaction with work-life balance were observed between 2011 and 2014 in probability-based samples of working US adults, resulting in an increasing disparity in burnout and satisfaction with work-life balance in physicians relative to the general US working population. After pooled multivariate analysis adjusting for age, sex, rela- tionship status, and hours worked per week, physicians remained at an increased risk of burnout (odds ratio, 1.97; 95% CI, 1.80-2.16; P<.001) and were less likely to be satisfied with work-life balance (odds ratio, 0.68; 95% CI, 0.62-0.75; P<.001). Conclusion: Burnout and satisfaction with work-life balance in US physicians worsened from 2011 to 2014. More than half of US physicians are now experiencing professional burnout.

ª 2015 Mayo Foundation for Medical Education and Research n Mayo Clin Proc. 2015;90(12):1600-1613

M edicine is both a demanding and a rewarding profession. Physicians spend more than a decade in postsec-

ondary education, work substantially more hours than most US workers in other fields, and often struggle to effectively integrate their personal and professional lives.1 They engage in highly technical and intellectually demanding work that often requires complex, high-stakes decision making despite substantial uncertainty. These challenges are offset by meaningful rela- tionships with patients, the intellectual stimula- tion of the work, and the satisfaction of helping fellow human beings.2-4 Physicians are also well

Mayo Clin Proc. n December 2015;90( www.mayoclinicproceedings.org n

compensated relative to many professions, are part of a fraternity of supportive colleagues, and often enjoy the respect and appreciation of their community.

The cumulative effect of these forces on the personal and professional satisfaction of each physician is unique. Although future physicians begin medical school with mental health profiles better than those of college graduates pursuing other fields,5 this profile is reversed 1 to 2 years into medical school.6

Once in practice, physicians have generally high degrees of satisfaction with their career choice but experience high degrees of

12):1600-1613 n http://dx.doi.org/10.1016/j.mayocp.2015.08.023 ª 2015 Mayo Foundation for Medical Education and Research

INCREASING BURNOUT AMONG US PHYSICIANS, 2011 TO 2014

professional burnout and dissatisfaction with work-life integration.1,7 Burnout is a syn- drome of emotional exhaustion, loss of mean- ing in work, feelings of ineffectiveness, and a tendency to view people as objects rather than as human beings.8 Burnout has profound implications for individual physicians and their families.9,10 In addition, burnout appears to impact the quality of care physicians pro- vide11-16 and physician turnover,17,18 which have profound implications for the quality of the health care delivery system.15,19,20

In 2011, we conducted a national study measuring burnout and other dimensions of well-being in US physicians as well as the gen- eral US working population.1 At the time of that study, approximately 45% of US physi- cians met criteria for burnout. Substantial vari- ation in the rate of burnout was observed by specialty, with the highest rates observed among many specialties at the front line of access to care (eg, family medicine, general internal medicine, and emergency medicine). Burnout among physicians also varied by career stage, with the highest rate among mid- career physicians.21 Burnout was more com- mon among physicians than among the general US working population, a finding that persisted after adjusting for age, sex, hours worked, and level of education.1

The landscape of medicine continues to rapidly evolve. Technology, legislation, and market forces have contributed to consolida- tion of medical practices, fluctuating reim- bursement, new care delivery models, increased productivity expectations for physi- cians, and more widespread use of electronic medical records over the past several years.22

The study of US physicians we first reported on in 2011 was designed to reevaluate the well-being and satisfaction of US physicians approximately every 3 years to assess changes in burnout and satisfaction with work-life bal- ance (WLB) over time. Here, we report results of the 2014 survey in comparison to the 2011 findings.

PATIENTS AND METHODS The 2014 survey used methods similar to those of the 2011 study.1 At both time points, we assessed a range of personal and profes- sional characteristics as well as personal well- being in several dimensions (described below).

Mayo Clin Proc. n December 2015;90(12):1600-1613 n http://dx.do www.mayoclinicproceedings.org

Participants Physician Sample. A sample of physicians from all specialty disciplines was assembled using the American Medical Association (AMA) Physician Master File (PMF). The PMF is a nearly complete record of all US phy- sicians independent of AMA membership. To ensure an adequate sample of physicians from each specialty area, we oversampled phy- sicians in fields other than family medicine, general pediatrics, general internal medicine, and obstetrics/gynecology. Canvassing e-mails stating the purpose of the study (eg, to better understand the factors that contribute to satis- faction in US physicians), along with an invita- tion to participate and a link to the survey, were sent to 94,032 physicians in August 2014 with 3 reminder requests sent over the ensuing 6 weeks. The 35,922 physicians who opened at least 1 invitation e-mail were considered to have received the invitation to participate in the study.23 Participation was voluntary, and all responses were anonymous.

Population Control Sample. For comparison to physicians, we surveyed a probability-based sample of individuals from the general US pop- ulation in October 2014. Although the initial population comparison (December 2010) used modest oversampling of individuals younger than 34 years (to allow comparison to medical students and residents),1 the 2014 population survey oversampled individuals between the ages of 35 and 65 years to better match the age range of practicing US physicians. The population survey was conducted using the Knowledge Panel, a probability-based panel (http://www. knowledgenetworks.com/knpanel/index.html and http://www.knowledgenetworks.com/ganp/ reviewer-info.html) designed to be representa- tive of the US population. On the basis of the intent to compare workers in other fields to physicians, only employed individuals were surveyed. The Mayo Clinic Institutional Review Board reviewed and approved the study.

Study Measures Both the physician and population controls pro- vided information on demographic characteris- tics (age, sex, and relationship status), hours worked per week, burnout, symptoms of depression, suicidal ideation, and satisfaction with WLB. Physician professional characteristics

i.org/10.1016/j.mayocp.2015.08.023 1601

MAYO CLINIC PROCEEDINGS

1602

were ascertained by asking physicians about their practice. Population controls also provided information about the highest level of education completed and occupation.

Burnout. Burnout among physicians was measured using the Maslach Burnout Inventory (MBI), a validated 22-item questionnaire consid- ered the criterion standard tool for measuring burnout.8,24-26 Consistent with convention,27-29

we considered physicians with a high score on the depersonalization and/or emotional exhaus- tion subscales of the MBI as having at least 1 manifestation of professional burnout.8

Although the 22-item MBI is the criterion standard for the assessment of burnout,8 its length and the expense of administration limit feasibility for use in long surveys addressing multiple content areas or in large population samples. Thus, to allow comparison of burnout between physicians and population controls, we measured burnout in both groups using 2 single-item measures adapted from the full MBI (ie, physicians completed the full MBI and the 2-item instrument; controls completed just the 2-item instrument). These 2 items correlated strongly with the emotional exhaustion and depersonalization domains of burnout as measured by the full MBI in a sample of more than 10,000 individuals30,31 with an area under the receiver operator characteristic curve of 0.94 and 0.93 for emotional exhaustion and deper- sonalization, respectively, for these single items relative to the full MBI. This approach has also been used in previous large-scale national studies of US physicians collectively enrolling more than 20,000 physicians.1,32

Symptoms of Depression and Suicidal Ideation. Symptoms of depression among physicians were assessed using the 2-item Pri- mary Care Evaluation of Mental Disorders,33 a standardized and validated assessment for depression screening that performs as well as longer instruments.34 It should be noted that this tool has a high sensitivity but lower spec- ificity such that approximately 1 of every 4 individuals screening positive would meet criteria for major depression if they were to undergo full psychiatric assessment. Recent suicidal ideation was evaluated by asking par- ticipants, “During the past 12 months, have you had thoughts of taking your own life?” This

Mayo Clin Proc. n December 2015;90(

item was designed to measure somewhat recent, but not necessarily active suicidal idea- tion.35 These questions have been used exten- sively in other studies and allow ready comparison to the prevalence of suicidal idea- tion in other studies of the US population36-38

and US physicians.1,39

Satisfaction With WLB. Satisfaction with WLB was assessed by the item “My work schedule leaves me enough time for my personal/family life” (response options: strongly agree, agree, neutral, disagree, and strongly disagree).1 In- dividuals who indicated “strongly agree” or “agree” were considered to be satisfied with their WLB, whereas those who indicated “disagree” or “strongly disagree” were considered to be dissatisfied with their WLB.

Statistical Analyses Standard descriptive summary statistics were used to characterize the physician and control samples. Associations between variables were evaluated using the Kruskal-Wallis test (contin- uous variables) or the chi-square test (categorical variables), as appropriate. All tests were 2-sided with type I error rates of .05. Multivariate analysis of differences across physician specialties was per- formed using logistic regression. Similarly, a pooled multivariate logistic regression analysis of physicians and population controls was per- formed to identify demographic and professional characteristics associated with the dependent out- comes. For all comparisons with population con- trols, physician data were restricted to responders who were between the ages of 29 and 65 years and not retired to match the age of the population sample. Comparisons between physicians in 2011 and 2014 were made using the chi-square test or the Kruskal-Wallis tests, as appropriate. These data were not paired and were treated as in- dependent samples. Comparisons in the propor- tions of burnout and satisfaction with WLB between physicians and populations controls in 2011 relative to 2014 were performed using Breslow-Day tests. All analyses were done using SAS version 9 (SAS Institute Inc).

RESULTS

Well-being of US Physicians Of the 35,922 physicians who received an invi- tation to participate, 6880 (19.2%) completed

12):1600-1613 n http://dx.doi.org/10.1016/j.mayocp.2015.08.023 www.mayoclinicproceedings.org

TABLE 1. Demographic Characteristics of Responding Physicians Compared With All US Physicians

Characteristic 2014 Responders

(N¼6880) All US physicians 2014 (n¼835,451)

2011 Responders (N¼7288)

Sex Male 4497 (67.5%) 557,063 (66.8%) 5241 (71.9%) Female 2162 (32.5%) 277,271 (33.2%) 2046 (28.1%) Missing 221 1117 1

Age (y) Median 56 51.5a 55 <35 332 (5.0%) 59,849 (7.2%) 321 (4.5%) 35-44 1223 (18.4%) 219,394 (26.3%) 1299 (18.0%) 45-54 1416 (21.3%) 219,492 (26.3%) 1842 (25.6%) 55-64 2193 (33.0%) 211,056 (25.3%) 2586 (35.9%) �65 1491 (22.4%) 125,660 (15.0%) 1162 (16.1%) Missing 225 75

Primary careb

Primary care 1596 (23.3%) 277,425 (32.1%)a 1907 (26.4%) Nonprimary care 5249 (76.7%) 585,507 (67.9%)a 5326 (73.6%)

Specialty Anesthesiology 236 (3.5%) 309 (4.3%) Dermatology 164 (2.4%) 174 (2.4%) Emergency medicine 355 (5.2%) 333 (4.6%) Family medicine 540 (7.9%) 752 (10.4%) General surgery 259 (3.8%) 276 (3.8%) General surgery subspecialtyc 381 (5.6%) 374 (5.2%) Internal medicine-general 453 (6.6%) 578 (8.0%) Internal medicine subspecialtyb 784 (11.5%) 1019 (14.1%) Neurology 246 (3.6%) 252 (3.5%) Neurosurgery 58 (0.9%) 82 (1.1%) Obstetrics and gynecology 246 (3.6%) 312 (4.3%) Ophthalmology 241 (3.5%) 199 (2.8%) Orthopedic surgery 239 (3.5%) 269 (3.7%) Otolaryngology 165 (2.4%) 193 (2.7%) Other 255 (3.7%) 329 (4.6%) Pathology 170 (2.5%) 184 (2.5%) Pediatrics-general 362 (5.3%) 286 (4.0%) Pediatric subspecialtyc 321 (4.7%) 239 (3.3%) Physical medicine and rehabilitation 170 (2.5%) 97 (1.3%) Preventive medicine, occupational medicine, or environmental medicine 112 (1.6%) 76 (1.1%)

Psychiatry 566 (8.3%) 488 (6.8%) Radiation oncology 64 (0.9%) 55 (0.8%) Radiology 261 (3.8%) 216 (3.0%) Urology 119 (1.7%) 136 (1.9%) Missing 66 60

Hours worked per week Median 50 (40-60) 50 (40-60) <40 1172 (17.4%) 985 (14.3%) 40-49 1340 (19.9%) 1459 (21.1%) 50-59 1667 (24.7%) 1852 (26.8%) 60-69 1526 (22.6%) 1659 (24.0%) 70-79 535 (7.9%) 455 (6.6%) �80 509 (7.5%) 497 (7.2%) Missing 131 381

No. of nights on call per week Median (interquartile range) 1 (0-3) 1 (0-3)

Primary practice setting Private practice 3605 (52.6%) 4087 (57.7%)

Continued on next page

INCREASING BURNOUT AMONG US PHYSICIANS, 2011 TO 2014

Mayo Clin Proc. n December 2015;90(12):1600-1613 n http://dx.doi.org/10.1016/j.mayocp.2015.08.023 www.mayoclinicproceedings.org

1603

TABLE 1. Continued

Characteristic 2014 Responders

(N¼6880) All US physicians 2014 (n¼835,451)

2011 Responders (N¼7288)

Primary practice setting, continued Academic medical center 1625 (23.7%) 1494 (21.1%) Veterans hospital 104 (1.5%) 184 (2.6%) Active military practice 58 (0.8%) 65 (0.9%) Not in practice or retired 160 (2.3%) 89 (1.3%) Other 1303 (19%) 1164 (16.4%) Missing 25 205

aAs of March 11, 2015. bPhysicians in subspecialty areas were intentionally oversampled to provide an adequate number of responses from physicians from each specialty to allow comparison across specialties. Primary care specialties include the following: Internal medicine-general, general practice, family medicine, obstetrics/gynecology, and pediatrics-general. cFor further subspecialty breakdown, see the Supplemental Material, available online at http://www.mayoclinicproceedings.org.

MAYO CLINIC PROCEEDINGS

1604

surveys. The demographic characteristics of participants relative to all 835,451 US physi- cians were generally similar, although partici- pants were slightly older (Table 1). The 2014 participants were also similar to the 2011 par- ticipants (Supplemental Table 1, available on- line at http://www.mayoclinicproceedings.org) other than a slight increase in women physi- cians (2011: 28.1%; 2014: 32.5%), consistent with the increased proportion of women among US physicians overall (2011: 30.7%; 2014: 33.2%). Analysis of early responders compared with late responders (a standard approach to evaluate for response bias) by age, sex, and specialty found no statistically significant differences when comparing sex and specialty (primary care vs nonprimary care) and only a minor difference in age (median, 56.0 years vs 57.0 years), providing further evidence that the sample was generally representative of US physicians from a demographic perspective.

Rates of burnout, symptoms of depression, suicidal ideation in the last 12 months, and satis- faction with WLB among participating physi- cians are summarized in Table 2. When assessed using the full MBI, 46.9% of US physi- cians had high emotional exhaustion, 34.6% high depersonalization, and 16.3% a low sense of personal accomplishment in 2014. In aggre- gate, 54.4% of the physicians had at least 1 symptom of burnout based on a high emotional exhaustion score and/or a high depersonaliza- tion score. Only 40.9% of the physicians felt that their work schedule left enough time for personal/family life, with 14.6% neutral and 44.5% disagreeing with this assertion.

Mayo Clin Proc. n December 2015;90(

Whencompared with 2011, rates ofburnout among physicians were higher (54.4% vs 45.5%; P<.001) in 2014 and satisfaction with WLB was lower (40.9% vs 48.5%; P<.001). In contrast, minimal differences were observed in the pro- portion of physicians reporting symptoms of depression (39.8% vs 38.2%; P¼.04) and no difference in the rates of suicidal ideation was observed (6.4% vs 6.4%; P¼.98).

As in 2011, substantial variation in the prev- alence of burnout was observed by specialty. Compared with 2011, the prevalence of burnout was higher for all specialty disciplines in 2014 (Figure 1, A). Family medicine (51.3% vs 63.0; P<.001), general pediatrics (35.3% vs 46.3%; P¼.005), urology (41.2% vs 63.6%; P<.001), orthopedic surgery (48.3% vs 59.6%; P¼.01), dermatology (31.8% vs 56.5%; P<.001), phys- ical medicine and rehabilitation (47.4% vs 63.3%; P¼.01), pathology (37.6% vs 52.5%; P¼.006), radiology (47.7% vs 61.4%; P¼.003), and general surgery subspecialties (42.4% vs 52.7%; P¼.005) each experienced a more than 10% increase in burnout.

Substantial variation in satisfaction with WLB was also observed by specialty. Satisfaction with WLB was lower in 2014 for all specialty dis- ciplines with the exception of obstetrics and gynecology and general surgery (Figure 1, B). Categorization of the 24 specialty disciplines based on whether the prevalence of burnout and satisfaction with WLB in their specialty was aboveorbelowtheprevalenceofallUSphysicians in each dimension is shown in Figure 1, C.

We next conducted multivariate analysis to identify factors associated with burnout

12):1600-1613 n http://dx.doi.org/10.1016/j.mayocp.2015.08.023 www.mayoclinicproceedings.org

TABLE 2. Physician Career Satisfaction, Burnout, Depression, and Quality of Life 2014 Relative to 2011

Variable 2014 2011 P

Burnout indicesa

Emotional exhaustion Median 25.0 21.0 <.001 % low score 2299 (34.1%) 3041 (42.2%) <.001 % intermediate score 1283 (19.0%) 1433 (19.9%) % high score 3165 (46.9%) 2734 (37.9%)

Depersonalization Median 7.0 5.0 <.001 % low score 2951 (44.0%) 3601 (50.1%) <.001 % intermediate score 1434 (21.4%) 1476 (20.5%) % high score 2325 (34.6%) 2116 (29.4%)

Personal accomplishment Median 41 42 <.001 % high score 4064 (61.2%) 4758 (66.6%) <.001 % intermediate score 1495 (22.5%) 1495 (20.9%) % low score 1085 (16.3%) 887 (12.4%)

Burned outb 3680 (54.4%) 3310 (45.5%) <.001 Depression Screen positive for depression 2715 (39.8%) 2753 (38.2%) .04

Suicidal ideation Suicidal ideation in the last 12 mo 438 (6.4%) 466 (6.4%) .98

Career satisfaction Would choose to become a physician again 4476 (67.0%) 5081 (70.2%) <.001 Would choose the same specialty again 4727 (70.8%) 5119 (70.8%) .94

Satisfaction with work-life balance Work schedule leaves me enough time for

my personal and/or family life Strongly agree 706 (10.6%) 1233 (17.0%) <.001 Agree 2012 (30.3%) 2279 (31.5%) Neutral 973 (14.6%) 1046 (14.4%) Disagree 2004 (30.1%) 1775 (24.5%) Strongly disagree 956 (14.4%) 911 (12.6%) Missing 229 44

aAs assessed using the full Maslach Burnout Inventory. Per the standard scoring of the MBI for health care workers, physicians with scores of �27 on the Emotional Exhaustion subscale, �10 on the Depersonalization subscale, or �33 on the Personal Accomplishment subscale are considered to have a high degree of burnout in that dimension. bHigh score on Emotional Exhaustion and/or Depersonalization subscales of the Maslach Burnout Inventory (see Methods).

INCREASING BURNOUT AMONG US PHYSICIANS, 2011 TO 2014

and satisfaction with WLB. Age, sex, specialty, hours worked per week, and practice setting were independently associated with both burnout and satisfaction with WLB (all P<.05; Supplemental Table 2, available online at http://www.mayoclinicproceedings.org).

Comparison of Physicians With the General US Working Population To compare the professional experience of practicing physicians relative to working US adults, 5313 nonretired physicians aged 29 to 65 years were compared with 5392 employed, nonphysician population control subjects aged 29 to 65 years (Table 3). The overall prevalence of burnout on the 2-item burnout measure for the general US working population was similar to that for the 2011 sample (28.4% vs 28.6%; P¼.85). Satisfaction with WLB for the general US working population in 2014 was slightly more favorable than for the 2011 sample (61.3% vs 55.1%; P<.001).

Compared with population controls, physi- cians were older (median, 53 years vs 52 years; P<.001), more likely to be men (62.2% vs 54.4%), and more likely to be married (82.9% vs 67.5%; P<.001). Similar to the 2011 find- ings, physicians worked a median of 10 hours more per week than US workers in general (50 vs 40 hours), with 41.8% of the physicians and 6.4% of the controls working 60 hours or more per week (P<.001 for both). On the 2- item burnout measure, physicians had higher rates of emotional exhaustion (43.2% vs 24.8%; P<.001), depersonalization (23.0% vs 14.0%; P<.001), and overall burnout (48.8% vs 28.4%; P<.001) (Figure 2, A). After adjusting for age, sex, relationship status, and hours worked per week, physicians remained at increased risk for burnout compared with the population (odds ratio, 1.97; 95% CI, 1.80- 2.16; P<.001) (Figure 2, B). Physicians also had a lower rate of satisfaction with WLB than did the general US working population (36.0% vs 61.3%; P<.001). After adjusting for age, sex, relationship status, and hours worked per week, physicians remained less likely to be satis- fied with WLB compared with the population (odds ratio, 0.68; 95% CI, 0.62-0.75; P<.001).

DISCUSSION Burnout is a pervasive problem among physi- cians that appears to be getting worse. Our

Mayo Clin Proc. n December 2015;90(12):1600-1613 n http://dx.do www.mayoclinicproceedings.org

findings suggest a 10% increase in the preva- lence of burnout among US physicians over the last 3 years. More than half of the US phy- sicians in our survey had symptoms of burnout when assessed using the full MBI, with increased rates of burnout observed across all specialties. A substantial erosion in satisfaction with WLB has also been observed among US physicians over the past 3 years, despite no increase in the median number of hours worked per week. In contrast to the in- crease in burnout and decrease in satisfaction

i.org/10.1016/j.mayocp.2015.08.023 1605

a

a

a

a

a

a

a

a

a

a

a

0 10 20 30 40 50 60 70 80

Preventive medicine/occupational medicine

Other

Radiation oncology

General pediatrics

Pediatric subspecialty

Psychiatry

Neurosurgery

Opthalmology

General surgery

Obstetrics and gynecology

Pathology

General surgery subspecialty

Internal medicine subspecialty

Otolaryngology

Mean burnout among all physicians participating

Anesthesiology

Dermatology

Neurology

General internal medicine

Orthopedic surgery

Radiology

Family medicine

Physical medicine and rehabilitiation

Urology

Emergency medicine

% Reporting burnout

2011 2014

A

FIGURE 1. Burnout (A) and satisfaction with WLB (B) by specialty 2014 vs 2011. For 1A and 1B, specialty discipline is shown on the y axis and burnout (A) and satisfaction with WLB (B) are shown on the x axis. For 1C, satisfaction with WLB is shown on the y axis and burnout on the x axis. GIM ¼ general internal medicine; OBGYN ¼ obstetrics and gynecology; PM&R ¼ physical medicine and rehabilitation; Prev ¼ Preventive medicine, occupational medicine, or environmental medicine; WLB ¼ work-life balance. aP<.05 from com- parison 2014 to 2011.

MAYO CLINIC PROCEEDINGS

1606

with WLB, minimal or no changes were observed in the prevalence of symptoms of depression or suicidal ideation.

It is notable that the increase in burnout and decrease in satisfaction with WLB in physicians

Mayo Clin Proc. n December 2015;90(

over the last 3 years runs counter to trends in the general US working population over the same interval. These disparate trends have resulted in a further widening in the rates of burnout and satisfaction with WLB among

12):1600-1613 n http://dx.doi.org/10.1016/j.mayocp.2015.08.023 www.mayoclinicproceedings.org

a

a

a

a

a

a

a

a

a

a

a

a

a

0 20 40 60 80

Neurosurgery

Urologic surgery

Neurology

General surgery subspecialty

Family medicine

Internal medicine subspecialty

Orthopedic surgery

General internal medicine

Anesthesiology

Pediatric subspecialty

General surgery

Otolaryngology

Radiology

Mean satisfaction

Pathology

Obstetrics and gynecology

Radation oncology

Physical medicine and rehabilitation

Dermatology

General pediatrics

Psychiatry

Emergency medicine

Opthalmology

Other

Preventive medicine/occupational medicine

% Satisfied that work leaves enough time for personal and/or family life

2011 2014

B

FIGURE 1. (continued).

INCREASING BURNOUT AMONG US PHYSICIANS, 2011 TO 2014

Mayo Clin Proc. n December 2015;90(12):1600-1613 n http://dx.doi.org/10.1016/j.mayocp.2015.08.023 www.mayoclinicproceedings.org

1607

Pediatric subspecialty Anesthesiology General surgery

Radiation oncology OBGYN

Otolaryngology Pathology

Psychiatry

Opthalmology

PM&R

General intermal medicine

Radiology

Family medicine

Emergency medicine

Dermatology

Prev/occupational medicine

Other

General pediatrics

GIM subspecialty General surgery subspecialty

NeurosurgeryAverage burnout

Average satisfaction WLB

Orthopedic surgery

Neurology

Urologic surgery

60

55

50

45

40

35

% S

at is

fie d

w ith

w or

k- lif

e ba

la n

ce

30

25

20

15 35 40 45 50 55

% Burned out 60 65 70 75

C

FIGURE 1. (continued).

MAYO CLINIC PROCEEDINGS

1608

physicians relative to the US working popula- tion, even after adjustment for differences in hours worked, age, sex, and relationship status.

What are the possible solutions to this prob- lem? More than 75% of the physicians are now employed by large health care organizations and meaningful progress will require an effective response at both the individual level and the or- ganization or system level.40 Health care organi- zations should focus on improving the efficiency and support in the practice environment,41-43

select and develop leaders with the skills to fos- ter physician engagement,44 help physicians optimize “career fit,”45 and create an environ- ment that nurtures community, flexibility, and control, all of which help cultivate meaning in work.2,3,41,42,46 Given the high number of hours worked by physicians as well as the unpredict- able nature of work hours in some settings (eg, surgery, hospital-based care), health care organizations must also establish principles that help facilitate work-life integration.47,48

Organizational approaches to help physicians self-calibrate and promote their own wellness may also be beneficial.49,50

Mayo Clin Proc. n December 2015;90(

There are also a number of steps physicians can take at the individual level to promote their own wellness. This often begins by identi- fying personal and professional values and determining how they will be prioritized when conflicts between personal and professional re- sponsibilities arise.51-53 This exercise requires self-awareness, limit setting, and reframing.51,53

Training in mindfulness-based stress reduction, which involves self-awareness, a focus on the present, and intentionality in thoughts and ac- tions, has also been shown to be an effective approach to reduce physician stress and burnout.54-56 Scientific studies have also identi- fied the habits and qualities that promote resil- ience in challenging situations, which are skills that can be learned and developed.57,58 Atten- tion to self-care, developing personal interests, and protecting and nurturing relationships are also essential.42,47,51,59

Our study is subject to several limitations. First, most of the physicians did not even open the e-mails informing them of the study and hence never received the invitation to participate. The participation rate among those

12):1600-1613 n http://dx.doi.org/10.1016/j.mayocp.2015.08.023 www.mayoclinicproceedings.org

TABLE 3. Comparison of Employed Physicians in the Sample Aged 29 to 65 y With a Probability-Based Sample of the Employed US Population Aged 29 to 65 y

Characteristic Physicians N¼5313 Population N¼5392 P Sex

Male 3291 (62.2%) 2934 (54.4%) <.001 Female 1996 (37.8%) 2458 (45.6%)

Age (y) Median 53 52 <.001 29-34 324 (6.1%) 526 (9.8%) <.001 35-44 1220 (23.0%) 1076 (20.0%) 45-54 1411 (26.6%) 1550 (28.7%) 55-65 2358 (44.4%) 2240 (41.5%)

Relationship status Single 632 (11.9%) 1300 (24.1%) <.001 Married 4387 (82.9%) 3642 (67.5%) Partnered 223 (4.2%) 354 (6.6%) Widowed/widower 52 (1.0%) 96 (1.8%) Missing 19 0

Hours worked per week Mean � SD 55 � 16.7 40 � 11.3 <.001 Median 50 40 <.001 <40 627 (11.9%) 1412 (26.2%) <.001 40-49 1042 (19.7%) 2927 (54.4%) 50-59 1400 (26.5%) 702 (13.0%) 60-69 1285 (24.4%) 268 (5.0%) 70-79 477 (9.0%) 36 (0.7%) �80 445 (8.4%) 39 (0.7%) Missing 37 8

Highest level of education completed Less than high school graduate 174 (3.2%) High school graduate 1159 (21.5%) Some college, no degree 1054 (19.5%) Associate degree 657 (12.2%) Bachelor’s degree 1341 (24.9%) Master’s degree 745 (13.8%) Professional or doctorate degree (other than MD/DO) 262 (4.9%) Missing 0

Occupation Professionala 2397 (45%) Health careb 390 (7.3%) Servicec 342 (6.4%) Salesd 414 (7.8%) Office and administrative support 428 (8.0%) Farming, forestry, fishing 22 (0.4%) Precision production, craft and repaire 341 (6.4%) Transportation and material moving 158 (3.0%) Armed services 26 (0.5%) Other 804 (15.1%) Missing 107

Distress Burnoutf

Emotional exhaustiong

Never 491 (9.4%) 718 (13.3%) <.001 A few times a year 1075 (20.5%) 1566 (29.1%) Once a month or less 663 (12.6%) 736 (13.7%) A few times a month 750 (14.3%) 1027 (19.1%) Once a week 626 (11.9%) 356 (6.6%)

Continued on next page

INCREASING BURNOUT AMONG US PHYSICIANS, 2011 TO 2014

Mayo Clin Proc. n December 2015;90(12):1600-1613 n http://dx.doi.org/10.1016/j.mayocp.2015.08.023 www.mayoclinicproceedings.org

1609

TABLE 3. Continued

Characteristic Physicians N¼5313 Population N¼5392 P Distress, continued

A few times a week 908 (17.3%) 634 (11.8%) Every day 736 (14.0%) 344 (6.4%) Missing 64 11 % High scoref 2270 (43.2%) 1334 (24.8%) <.001

Depersonalizationh

Never 1454 (27.7%) 2368 (44.3%) <.001 A few times a year 1308 (24.9%) 1255 (23.5%) Once a month or less 647 (12.3%) 492 (9.2%) A few times a month 631 (12.0%) 487 (9.1%) Once a week 461 (8.8%) 223 (4.2%) A few times a week 555 (10.6%) 311 (5.8%) Every day 193 (3.7%) 214 (4.0%) Missing 64 42 % High scoref 1209 (23.0%) 748 (14.0%) <.001

Burned outi 2550 (48.8%) 1529 (28.4%) <.001 Suicidal ideation

Suicidal ideation in the past 12 mo 383 (7.2%) 213 (4.0%) <.001 Work-life balance

Work schedule leaves me enough time for my personal/family life:

Strongly agree 402 (7.6%) 1227 (22.8%) <.001 Agree 1500 (28.4%) 2071 (38.5%) Neutral 782 (14.8%) 1012 (18.8%) Disagree 1738 (32.9%) 817 (15.2%) Strongly disagree 865 (16.4%) 249 (4.6%) Missing 26 16

aBusiness/financial, management, computer/mathematical, architecture/engineering, lawyer/judge, life/physical/social sciences, community/ social services, teacher nonuniversity, teacher college/university, and other. bNurse, pharmacist, paramedic, laboratory technician, nursing aide, orderly, and dental assistant. cProtective service, food preparation/service, building cleaning/maintenance, and personal care/service. dSales representative, retails sales, and other sales. eConstruction and extraction, installation/maintenance/repair, precision production (machinist, welder, backer, printer, and tailor). fAs assessed using the single-item measures for emotional exhaustion and depersonalization adapted from the full Maslach Burnout Inventory. Area under the receiver operating characteristic curve for the emotional exhaustion and depersonalization single items relative to that of their respective full Maslach Burnout Inventory domain score in previous studies was 0.94 and 0.93 and the positive predictive value of the single-item thresholds for high levels of emotional exhaustion and depersonalization was 88.2% and 89.6%, respectively.30 gIndividuals indicating emotional exhaustion symptoms weekly or more often have median emotional exhaustion scores of >30 on the full MBI and have a >75% probability of having a high emotional exhaustion score as defined by the Maslach Burnout Inventory (�27). hIndividuals indicating depersonalization symptoms weekly or more often have median depersonalization scores of >13 on the full Maslach Burnout Inventory and have a >85% probability of having a high depersonalization score as defined by the Maslach Burnout Inventory (�10). iHigh score (�weekly) on emotional exhaustion and/or depersonalization scales.

MAYO CLINIC PROCEEDINGS

1610

who opened the invitation e-mails was only 19%. Although the participation rate is gener- ally consistent with other national survey studies of physicians,7,60,61 it is lower than that of physician surveys in general.62 We did not use monetary or other incentives to improve participation.63 Nonetheless, several cross-sectional studies have failed to identify significant differences between responding and nonresponding physicians,64 with evidence

Mayo Clin Proc. n December 2015;90(

that nonresponse may be of less concern in physicians surveys than in surveys of the gen- eral public.65 We found no statistically signifi- cant differences between early responders and late responders (a standard approach to eval- uate for response bias) with respect to sex or specialty (primary care vs nonprimary care) and minimal differences by age (median, 56 years vs 57 years), providing support that re- sponders were representative of US physicians.

12):1600-1613 n http://dx.doi.org/10.1016/j.mayocp.2015.08.023 www.mayoclinicproceedings.org

100% B

u rn

ed o

u t (

%)

80%

60%

40%

20%

0%

2011 2014

Year

Breslow-day P value <.001

Population Physicians

100%

S at

is fie

d w

ith W

L B

(% )

80%

60%

40%

20%

0%

2011 2014

Year

Breslow-day P value <.001

Population Physicians

Burnout

Satisfaction with WLB

A

B

FIGURE 2. Changes in burnout and satisfaction with WLB in physicians and population year are shown on the x axis. Burnout (A) and satisfaction with WLB (B) are shown on the y axis. WLB ¼ work-life balance.

INCREASING BURNOUT AMONG US PHYSICIANS, 2011 TO 2014

Second, our survey was anonymous and we were unable to assess changes over time at the individual physician level. Third, although the age of individuals in the comparison sample of population controls was generally similar to that of physicians, they were more likely to be women. This was expected because of the demographic characteristics of US physicians and was adjusted for in the multivariate

Mayo Clin Proc. n December 2015;90(12):1600-1613 n http://dx.do www.mayoclinicproceedings.org

analysis; however, it remains possible that other unmeasured confounders exist.

Our study also has several important strengths. The physician sample was derived from the AMA PMF, which is a near complete registry of all US physicians. The sample included physicians from all specialty disci- plines, practice settings, and environments. Overall, the characteristics of participating

i.org/10.1016/j.mayocp.2015.08.023 1611

MAYO CLINIC PROCEEDINGS

1612

physicians appear similar to those of both US physicians overall and the 2011 comparison sample of physicians. The same validated instruments were used to study physicians in both 2011 and 2014, facilitating direct compar- ison. We also studied a sample of population controls at both time points to allow compari- son of the physician experience with the gen- eral US working population and provide context to how the physician experience com- pares to that of US workers in general.

CONCLUSION Burnout and satisfaction with WLB among US physicians are getting worse. American medicine appears to be at a tipping point with more than half of US physicians experi- encing professional burnout. Given the extensive evidence that burnout among physicians has effects on quality of care, pa- tient satisfaction, turnover, and patient safety, these findings have important implications for society at large.11-20 There is an urgent need for systematic application of evidence- based interventions addressing the drivers of burnout among physicians. These interventions must address contributing factors in the practice environment rather than focusing exclusively on helping physicians care for themselves and training them to be more resilient.

SUPPLEMENTAL ONLINE MATERIAL Supplemental material can be found online at http://www.mayoclinicproceedings.org. Sup- plemental material attached to journal articles has not been edited, and the authors take re- sponsibility for the accuracy of all data.

Abbreviations and Acronyms: AMA = American Medical Association; MBI = Maslach Burnout Inventory; PMF = Physician Master File; WLB = work-life balance

Affiliations (Continued from the first page of this article.): (D.S., J.S.), and Division of General Internal Medi- cine (C.P.W.), Mayo Clinic, Rochester, MN; and American Medical Association, Chicago, IL (O.H., C.S.).

Grant Support: The work was supported by the Mayo Clinic Program on Physician Well-being.

Potential Competing Interests: Dr Shanafelt is co-inventor of the Physician Well-being Index. Mayo Clinic holds the copyright on this technology and accordingly Mayo Clinic and Dr Shanafelt have a potential financial interest in this technology. The Physician

Mayo Clin Proc. n December 2015;90(

Well-Being Index has been licensed to a commercial entity, although no royalties have been received to date.

Correspondence: Address to Tait D. Shanafelt, MD, Department of Internal Medicine, Mayo Clinic, 200 First St SW, Rochester, MN 55905 ([email protected]).

REFERENCES 1. Shanafelt TD, Boone S, Tan L, et al. Burnout and satisfaction

with work-life balance among US physicians relative to the general US population. Arch Intern Med. 2012;172(18): 1377-1385.

2. Horowitz CR, Suchman AL, Branch WT Jr, Frankel RM. What do doctors find meaningful about their work? Ann Intern Med. 2003;138(9):772-775.

3. Clever LH. Some things have not changed. Ann Intern Med. 2000;132(1):85-89.

4. McMurray JE, Williams E, Schwartz MD, et al. Physician job satisfaction: developing a model using qualitative data. SGIM Career Satisfaction Study Group. J Gen Intern Med. 1997; 12(11):711-714.

5. Brazeau CM, Shanafelt T, Durning SJ, et al. Distress among matriculating medical students relative to the general popula- tion. Acad Med. 2014;89(11):1520-1525.

6. Dyrbye LN, West CP, Satele D, et al. Burnout among U.S. medical students, residents, and early career physicians relative to the general U.S. population. Acad Med. 2014;89(3):443-451.

7. Shanafelt TD, Balch CM, Bechamps GJ, et al. Burnout and career satisfaction among American surgeons. Ann Surg. 2009; 250(3):463-471.

8. Maslach C, Jackson S, Leiter M. Maslach Burnout Inventory Manual. 3rd ed. Palo Alto, CA: Consulting Psychologists Press; 1996.

9. Dyrbye LN, Schwartz A, Downing SM, Szydlo DW, Sloan JA, Shanafelt TD. Efficacy of a brief screening tool to identify medical students in distress. Acad Med. 2011; 86(7):907-914.

10. Shanafelt TD, Boone SL, Dyrbye LN, et al. The medical marriage: a national survey of the spouses/partners of US physicians. Mayo Clin Proc. 2013;88(3):216-225.

11. West CP, Huschka MM, Novotny PJ, et al. Association of perceived medical errors with resident distress and empathy: a prospective longitudinal study. JAMA. 2006; 296(9):1071-1078.

12. Shanafelt TD, Balch CM, Bechamps G, et al. Burnout and med- ical errors among American surgeons. Ann Surg. 2010;251(6): 995-1000.

13. West CP, Tan AD, Habermann TM, Sloan JA, Shanafelt TD. Association of resident fatigue and distress with perceived medical errors. JAMA. 2009;302(12):1294-1300.

14. Firth-Cozens J, Greenhalgh J. Doctors’ perceptions of the links between stress and lowered clinical care. Soc Sci Med. 1997; 44(7):1017-1022.

15. Grol R, Mokkink H, Smits A, et al. Work satisfaction of general practitioners and the quality of patient care. Fam Pract. 1985; 2(3):128-135.

16. Haas JS, Cook EF, Puopolo AL, Burstin HR, Cleary PD, Brennan TA. Is the professional satisfaction of general internists associated with patient satisfaction? J Gen Intern Med. 2000;15(2):122-128.

17. Shanafelt T, Sloan J, Satele D, Balch C. Why do surgeons consider leaving practice? J Am Coll Surg. 2011;212(3):421-422.

18. Shanafelt TD, Raymond M, Kosty M, et al. Satisfaction with work-life balance and the career and retirement plans of US oncologists. J Clin Oncol. 2014;32(11):1127-1135.

19. Wallace JE, Lemaire JB, Ghali WA. Physician wellness: a missing quality indicator. Lancet. 2009;374(9702):1714-1721.

20. Dyrbye LN, Shanafelt TD. Physician burnout: a potential threat to successful health care reform. JAMA. 2011;305(19): 2009-2010.

12):1600-1613 n http://dx.doi.org/10.1016/j.mayocp.2015.08.023 www.mayoclinicproceedings.org

INCREASING BURNOUT AMONG US PHYSICIANS, 2011 TO 2014

21. Dyrbye LN, Varkey P, Boone SL, Satele DV, Sloan JA, Shanafelt TD. Physician satisfaction and burnout at different career stages. Mayo Clin Proc. 2013;88(12):1358-1367.

22. Friedberg MW, Chen PG, Van Busum KR, et al. Factors affecting physician professional satisfaction and their implications for patient care, health systems, and health policy. 2013. http://www.rand.org/ content/dam/rand/pubs/research_reports/RR400/RR439/RAND_ RR439.pdf. Accessed December 31, 2015.

23. (AAPOR): Aafpor. Standard definitions: final dispositions of case codes and outcome rates for surveys. 2011. http://www.aapor. org/Content/aapor/AdvocacyandInitiatives/StandardsandEthics/ StandardDefinitions/StandardDefinitions2011.pdf. Accessed December 3, 2012.

24. Rafferty JP, Lemkau JP, Purdy RR, Rudisill JR. Validity of the Maslach Burnout Inventory for family practice physicians. J Clin Psychol. 1986;42(3):488-492.

25. Lee RT, Ashforth BE. A meta-analytic examination of the cor- relates of the three dimensions of job burnout. J Appl Psychol. 1996;81(2):123-133.

26. Leiter MP, Durup J. The discriminant validity of burnout and depression: a confirmatory factor analytic study. Anxiety Stress Copin. 1994;7(4):357-373.

27. Thomas NK. Resident burnout. JAMA. 2004;292(23):2880-2889. 28. Shanafelt TD, Bradley KA, Wipf JE, Back AL. Burnout and

self-reported patient care in an internal medicine residency program. Ann Intern Med. 2002;136(5):358-367.

29. Rosen IM, Gimotty PA, Shea JA, Bellini LM. Evolution of sleep quantity, sleep deprivation, mood disturbances, empathy, and burnout among interns. Acad Med. 2006;81(1):82-85.

30. West CP, Dyrbye LN, Sloan JA, Shanafelt TD. Single item mea- sures of emotional exhaustion and depersonalization are useful for assessing burnout in medical professionals. J Gen Intern Med. 2009;24(12):1318-1321.

31. West CP, Dyrbye LN, Satele DV, Sloan JA, Shanafelt TD. Concurrent validity of single-item measures of emotional exhaustion and depersonalization in burnout assessment. J Gen Intern Med. 2012;27(11):1445-1452.

32. West CP, Shanafelt TD, Kolars JC. Quality of life, burnout, educational debt, and medical knowledge among internal med- icine residents. JAMA. 2011;306(9):952-960.

33. Spitzer RL, Williams JB, Kroenke K, et al. Utility of a new pro- cedure for diagnosing mental disorders in primary care. The PRIME-MD 1000 study. JAMA. 1994;272(22):1749-1756.

34. Whooley MA, Avins AL, Miranda J, Browner WS. Case-finding instruments for depression: two questions are as good as many. J Gen Intern Med. 1997;12(7):439-445.

35. Meehan PJ, Lamb JA, Saltzman LE, O’Carroll PW. Attempted sui- cide among young adults: progress toward a meaningful estimate of prevalence. Am J Psychiatry. 1992;149(1):41-44.

36. Kessler RC, Borges G, Walters EE. Prevalence of and risk factors for lifetime suicide attempts in the National Comorbidity Sur- vey. Arch Gen Psychiatry. 1999;56(7):617-626.

37. Kessler RC, Berglund P, Borges G, Nock M, Wang PS. Trends in suicide ideation, plans, gestures, and attempts in the United States, 1990-1992 to 2001-2003. JAMA. 2005; 293(20):2487-2495.

38. Cooper-Patrick L, Crum RM, Ford DE. Identifying suicidal ideation in general medical patients. JAMA. 1994;272(22):1757-1762.

39. Shanafelt TD, Balch CM, Dyrbye L, et al. Special report: suicidal ideation among American surgeons. Arch Surg. 2011;146(1):54-62.

40. Hawkins M. 2012 Review of physician recruiting incentives. 2012. http://www.merritthawkins.com/uploadedFiles/MerrittHawkins/ Pdf/mha2012incentivesurveyPDF.pdf. Accessed May 5, 2014.

41. Sinsky CA, Willard-Grace R, Schutzbank AM, Sinsky TA, Margolius D, Bodenheimer T. In search of joy in practice: a report of 23 high-functioning primary care practices. Ann Fam Med. 2013;11(3):272-278.

42. Linzer M, Levine R, Meltzer D, Poplau S, Warde C, West CP. 10 bold steps to prevent burnout in general internal medicine. J Gen Intern Med. 2014;29(1):18-20.

Mayo Clin Proc. n December 2015;90(12):1600-1613 n http://dx.do www.mayoclinicproceedings.org

43. Linzer M, Baier Manwell L, Mundt M, et al. Organizational climate, stress, and error in primary care: the MEMO study. In: Henriksen K, Battles JB, Marks ES, Lewin DI, eds. Advances in Patient Safety: From Research to Implementation (Volume 1: Research Findings). Rockville, MD: Agency for Healthcare Research and Quality (US); 2005.

44. Shanafelt TD, Gorringe G, Menaker R, et al. Impact of organi- zational leadership on physician burnout and satisfaction. Mayo Clin Proc. 2015;90(4):432-440.

45. Shanafelt TD, West CP, Sloan JA, et al. Career fit and burnout among academic faculty. Arch Intern Med. 2009;169(10):990-995.

46. West CP, Dyrbye LN, Rabatin JT, et al. Intervention to promote physician well-being, job satisfaction, and professionalism: a ran- domized clinical trial. JAMA Intern Med. 2014;174(4):527-533.

47. Shanafelt TD, Sloan JA, Habermann TM. The well-being of phy- sicians. Am J Med. 2003;114(6):513-519.

48. Shanafelt TD, West CP, Poland GA, LaRusso NF, Menaker R, Bahn RS. Principles to promote physician satisfaction and work-life balance. Minn Med. 2008;91(12):41-43.

49. Dyrbye LN, Satele D, Sloan J, Shanafelt TD. Utility of a brief screening tool to identify physicians in distress. J Gen Intern Med. 2013;28(3):421-427.

50. Shanafelt TD, Kaups KL, Nelson H, et al. An interactive individual- ized intervention to promote behavioral change to increase per- sonal well-being in US surgeons. Ann Surg. 2014;259(1):82-88.

51. Quill TE, Williamson PR. Healthy approaches to physician stress. Arch Intern Med. 1990;150(9):1857-1861.

52. Clever LH. A checklist for making good choices in tryingdor tranquildtimes. West J Med. 2001;174(1):41-43.

53. Shanafelt T, Chung H, White H, Lyckholm LJ. Shaping your career to maximize personal satisfaction in the practice of oncology. J Clin Oncol. 2006;24(24):4020-4026.

54. Krasner MS, Epstein RM, Beckman H, et al. Association of an educational program in mindful communication with burnout, empathy, and attitudes among primary care physicians. JAMA. 2009;302(12):1284-1293.

55. Fortney L, Luchterhand C, Zakletskaia L, Zgierska A, Rakel D. Abbreviated mindfulness intervention for job satisfaction, qual- ity of life, and compassion in primary care clinicians: a pilot study. Ann Fam Med. 2013;11(5):412-420.

56. Beckman HB, Wendland M, Mooney C, et al. The impact of a program in mindful communication on primary care physicians. Acad Med. 2012;87(6):815-819.

57. Southwick SM, Charney DS. Resilience: The Science of Mastering Life’s Greatest Challenges. Cambridge, UK: Cambridge University Press; 2012.

58. Zwack J, Schweitzer J. If every fifth physician is affected by burnout, what about the other four? Resilience strategies of experienced physicians. Acad Med. 2013;88(3):382-389.

59. Shanafelt TD, Oreskovich MR, Dyrbye LN, et al. Avoiding burnout: the personal health habits and wellness practices of US surgeons. Ann Surg. 2012;255(4):625-633.

60. Allegra CJ, Hall R, Yothers G. Prevalence of burnout in the U.S. oncology community: results of a 2003 survey. J Oncol Pract. 2005;1(4):140-147.

61. Kuerer HM, Eberlein TJ, Pollock RE, et al. Career satisfaction, practice patterns and burnout among surgical oncologists: report on the quality of life of members of the Society of Sur- gical Oncology. Ann Surg Oncol. 2007;14(11):3043-3053.

62. Asch DA, Jedrziewski MK, Christakis NA. Response rates to mail surveys published in medical journals. J Clin Epidemiol. 1997;50(10):1129-1136.

63. Kasprzyk D, Montaño DE, St Lawrence JS, Phillips WR. The effects of variations in mode of delivery and monetary incentive on physicians’ responses to a mailed survey assessing STD prac- tice patterns. Eval Health Prof. 2001;24(1):3-17.

64. Kellerman SE, Herold J. Physician response to surveys: a review of the literature. Am J Prev Med. 2001;20(1):61-71.

65. Puleo E, Zapka J, White MJ, Mouchawar J, Somkin C, Taplin S. Caffeine, cajoling, and other strategies to maximize clinician sur- vey response rates. Eval Health Prof. 2002;25(2):169-184.

i.org/10.1016/j.mayocp.2015.08.023 1613

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Work–life balance is defined here as an indi-vidual’s ability to meet both their work and family commitments, as well as other non- work responsibilities and activities. Along with Sturges and Guest (2004), we use a deliberately broader term than previous research on work–family balance (eg Saltzstein, Ting & Saltzstein 2001), or work–family conflict (eg Frone, Russell & Cooper 1992), in growing recognition of the desire of all employed people, regardless of marital or parental status, to achieve

a healthier and more satisfying balance of their roles and responsibilities.

There is a plethora of research demonstrating the importance of work–life balance (or work–family conflict) for the health and well- being of individuals and families (eg Eby, Casper, Lockwood, Bordeaux & Brinley 2005; Frone et al 1992; Pocock 2003). Work–life balance is related to reduced stress and greater life satisfaction, with some indication that the relationship is strength- ening over time (Allen, Herst, Bruck & Sutton

226677

Copyright © eContent Management Pty Ltd. Journal of Management & Organization (2008) 14: 267–284.

Work–life balance or work–life alignment?

A test of the importance of work–life balance for employee engagement and intention

to stay in organisations

LLOOUUIISSEE PP PPAARRKKEESS Voice Project, School of Psychology, Macquarie University, Sydney NSW, Australia

PPEETTEERR HH LLAANNGGFFOORRDD Voice Project, School of Psychology, Macquarie University, Sydney NSW, Australia

AABBSSTTRRAACCTT In an Australian sample of over 16,000 employees we assessed whether employees are satisfied with their ability to balance work and other life commitments. We tested the hypothesis that work–life balance is important for engaging and retaining employees in the context of other aspects of organisational climate. We also explored how individual and organisational vari- ables were related to work–life balance aiding further development of theory integrating work with other aspects of life. Results showed that of 28 organisational climate factors, work–life balance was least related to employee engagement and intention to stay with an organisation. We discuss implications for how organisations position work–life balance strategies, particular- ly in relation to social responsibility and wellness, rather than the solution to employee commit- ment and retention.

Keywords: work–life balance; employee engagement; retention; social responsibility; organisational climate

Volume 14, Issue 3, July 2008 JJOOUURRNNAALL OOFF MMAANNAAGGEEMMEENNTT && OORRGGAANNIIZZAATTIIOONN

2000). In addition, although role-conflict theorists attributed this relationship to the ‘balance’ between multiple roles as a buffer for negative experiences (Marks & MacDermid 1996), it appears to be the place of work in the rest of life which has a major impact. Whereas an emphasis on family (in terms of time, involvement and sat- isfaction) is related to higher quality of life, imbal- ance caused by a greater emphasis on work has a negative effect on quality of life, mediated by increased work–family conflict and stress (Frone et al 1992; Greenhaus, Collins & Shaw 2003).

There is some evidence of increased pressure on work–life balance in Australia’s relatively long working hours (ranked second longest in the OECD), the increase in non-standard working hours, a growing number of employees who would prefer fewer hours than they currently work and the shift from male-breadwinner to dual-earner couples and single-parent households (Bittman & Rice 2002; Jacobs & Gerson 2001; Thornthwaite 2004). It is generally contested both in research literature and popular media that working men and women in Australia are dissat- isfied with their current levels of work–life bal- ance and are actively seeking to attain a ‘better’ balance through changes in work hours and arrangements, moving to ‘family-friendly’ organi- sations, or more radical ‘sea change’ type shifts (eg De Cieri, Holmes, Abbott & Pettit 2005; Fox 2006; Russell & Bowman 2000; Schmidt 2006; Smith 2006; Thornthwaite 2004).

Traditionally, work–life balance has been seen as an issue for individual employees, with organi- sational efforts at improving work–life balance focusing on programs aimed to help employees better manage their home life (for example, childcare or counselling). However, with growing awareness of the current skills shortage and war for talent, a subtle shift has been observed in the arguments for work–life balance, from respond- ing to individual employee needs to a broader based business case (Russell 2002; Thorthwaite 2004). Proponents argue that work–life balance contributes to employee engagement (job satis-

faction and organisation commitment), which in turn contributes to higher productivity and lower organisational turnover (Grawitch, Gottschalk & Munz 2006). For example, De Cieri et al (2005: 92) argue that any organisation aiming to increase competitive advantage must ‘develop the capability to attract, motivate and retain a highly skilled, flexible and adaptive workforce’ by ‘an approach to HR and work–life balance strategies that cater for the diverse needs of the workforce’. However, Lewis, Rapoport and Gamble (2003: 830) contend that the ‘business case, in many sit- uations is deeply flawed, and there is a need to be more honest about this’. Indeed, some assertions that work–life balance will increase the motiva- tion and retention of a talented workforce have been made on the basis of anecdotal, rather than empirical, evidence (eg Pocock 2005).

Consistent with this strategy to attract and retain a diverse workforce, work–life balance is often considered more important for women (who continue to bear the burden of domestic duties), older employees and the younger ‘work to live’ generation ‘Y’ (De Cieri et al 2005; Pocock 2005; Schmidt 2006). Behson (2002) found some evidence to suggest that family- friendly work cultures were slightly more impor- tant for satisfaction and commitment for women than men, and parents than non-parents. Howev- er, there is emerging in the literature a consistent lack of findings in this area. For example, when controlling for organisational variables Sturges and Guest (2004) found age, gender, marital sta- tus and dependent children had no effect on work/non-work conflict. Similarly, Greenhaus et al (2003) did not find any moderating effects of gender, parenthood and career aspirations on relations between work–family balance and quali- ty of life.

Given the high level of interest in work–life balance among researchers, practitioners and commentators, we aimed to: (1) empirically investi- gate employees’ satisfaction with work–life balance and (2) to test the impact of work–life balance (compared to other aspects of the work environ-

226688

Louise P Parkes and Peter H Langford

JJOOUURRNNAALL OOFF MMAANNAAGGEEMMEENNTT && OORRGGAANNIIZZAATTIIOONN Volume 14, Issue 3, July 2008

ment) on employee engagement. In addition, we wanted to explore how work–life balance is situ- ated in relation to other aspects of the work envi- ronment to help managers effectively integrate work–life balance strategies within a broader organisational context. Thus, our third aim was to examine individual and work variables that might moderate or mediate the relationship between work–life balance and work outcomes (Allen et al 2000).

LLiinnkkiinngg wwoorrkk––lliiffee bbaallaannccee ttoo oorrggaanniissaattiioonnaall oouuttccoommeess Despite a prevalent belief to the contrary, there are a number of studies that have failed to find a significant relationship between work–life bal- ance and organisational outcomes, or have reported mixed and even negative effects (Bruck, Allen & Spector 2002; Forsyth & Polzer- Debruyne 2007; Kossek & Ozeki 1998). In their meta-analysis Allen et al (2000) found moderate correlations on average between work–family con- flict and both job satisfaction (r = –0.24) and organisational commitment (r = –0.23). How- ever, individual study results were highly incon- sistent (ranging from +0.14 to –0.47). These inconsistencies in the existing literature linking work–life balance with organisational outcomes may stem from both theoretical and methodolog- ical short-comings.

First, finding a direct link between work–life balance and organisational outcomes has rarely been the result of rigourous theory development or testing (Eby et al 2005). The few exceptions have relied mostly on concepts such as the psy- chological contract (Rousseau 1995) and social exchange theory (Settoon, Bennett & Liden 1996), which predict for example, organisational citizenship behaviour in return for perceived use- fulness of work–family benefits (Lambert 2000), or inversely, dissatisfaction with and lower com- mitment to work due to perceived imbalance in workload and hours commensurate to rewards (Sturges & Guest 2004). However, while there may be a direct link between work–life balance

and employees’ satisfaction, commitment and intention to stay with an organisation (based on the above theories), this is likely to be small com- pared to the impact of other organisational fac- tors. For example, while extolling the virtues of flexibility in the workforce, Bond, Galinsky and Hill (2004) actually found that flexibility was least related to the item ‘I am willing to work harder than I have to, to help my company suc- ceed’ of their six criteria for ‘effective’ workplaces. In comparisons of workplaces high and low on flexibility, the percentage of highly engaged employees differed by 10%, as opposed to differ- ences of 20–36% for the five other effective work practices (these were job autonomy, learning opportunities and challenges on the job, supervi- sor and co-worker support for job success, and involvement in management decision-making).

Rather than a direct link, we suggest that work–life balance is more likely to benefit an organisation indirectly through those well-being factors found to be consistently and strongly asso- ciated with it, that is, work-related stress and burnout (Allen et al 2000; Eby et al 2005; Fox & Dwyer 1999). Other aspects of organisational cli- mate appear to interact with, and support, this relationship. For example, work–life balance is positively related to the perceived fairness and support of supervisors (Nielson, Carlson & Lankau 2001; White, Hill, McGovern, Mills & Smeaton 2003), organisational understanding of family needs (Allen 2001; Saltzstein et al 2001) and support for out-of-work activities and respon- sibilities (Sturges & Guest 2004). Yet Frone, Yard- ley and Markel (1997) found that support from supervisors and co-workers appeared to reduce work-to-family conflict primarily by reducing work distress and work overload.

Certainly, one of the strongest explanatory variables for work–life balance is the length of working hours, with work–life balance higher among those who work fewer hours (Dex & Bond 2005; Parasuraman & Simmers 2001; Sturges & Guest 2004; Thornthwaite 2004; White et al 2003). Although often considered at

226699

Work–life balance or work–life alignment?

Volume 14, Issue 3, July 2008 JJOOUURRNNAALL OOFF MMAANNAAGGEEMMEENNTT && OORRGGAANNIIZZAATTIIOONN

the individual level, longer and more intense working hours can be attributed to ‘high commit- ment’ management practices and expectations of prioritising work over other responsibilities (Frone et al 1997; Hand & Lewis 2002; Peetz et al 2003; Russell & Bowman 2000). There is some evidence that practices encouraging high commit- ment and performance such as performance appraisal systems, quality circles, training and career development can affect pay, promotion and effort in a way that negatively impacts work–fam- ily balance (Lewis, Gamble & Rapoport 2007; White et al 2003). According to role-conflict the- ory, high job involvement (and presumably high organisational commitment) entails greater time, effort and preoccupation with the work role which detracts from an employee’s ability to fulfil the demands of other roles (Frone et al 1992; Hammer, Allen & Grigsby 1997; Parasuraman & Simmers 2001). For example, Adams, King and King (1996) found that workers who reported higher levels of job involvement were more satis- fied with their jobs, but also reported higher lev- els of work–family conflict.

The complex relationship between work–life balance and other aspects of organisational cli- mate highlights the inadequate design of much of the research in this area. The majority of studies that have empirically examined the impact of work–life balance policies (or satisfaction with work–life balance) on work-related outcomes have tested this relationship in isolation. For example, Marks and MacDermid (1996) sur- veyed 65 employed wives and mothers in the United States. They found that role-balanced women (those scoring high on the single item: ‘Nowadays, I seem to enjoy every part of my life equally well’) were higher on ‘work productivity’ as measured by a single self-report item. The impact of role-balance was not compared to any other work-related variable (no others were meas- ured). When studied alone, work–life balance strategies, family-friendly culture and higher work-life balance can, and usually do, demon- strate a positive correlation with an employee’s

job satisfaction, organisational commitment and citizenship and decision to remain with an employer (Allen 2001; Burke 2001; Forsyth & Polzer-Debruyne 2007; Greenhaus, Collins, Singh & Parasuraman 1997; Haar & Spell 2004; Kossek & Ozeki 1998; Lambert 2000; Macran, Joshi & Dex 1996; Netemeyer, Boles & McMur- rian 1996; Saltzstein et al 2001; Thompson, Beauvais & Lyness 1999). However, the limited scope of these studies rarely allows for tests of mediation, nor do they assess the importance of work–life balance in the context of other aspects of the work environment.

To our knowledge, only three studies have examined the impact of work–life balance on employee outcomes in the context of other organisational variables. For example, in a sample of 147 employed American students, Behson (2002) examined the impact of family–friendly work cultures on work-family conflict, job satis- faction and organisational commitment in com- parison to three broader aspects of the work climate. While perceptions of family-supportive cultures affected work–family conflict, they did not significantly impact job satisfaction and com- mitment when controlling for perceived organisa- tional support, fair interpersonal treatment and trust in management.

Greenhaus et al (1997) investigated reasons for organisational departure among 310 accountants with moderate home responsibilities (either mar- ried or with children). They found that, rather than work–family conflict, work overload and career advancement aspirations predicted turnover intentions (other work-related variables included career development opportunities and advancement expectations). While women were more likely to leave than men, this was attributa- ble to differences in their career aspirations.

Finally, in a study of 280 graduates in the early stages of their career, Sturges and Guest (2004) found that while graduates professed work–life balance was very important to their intentions to stay with their organisation, work/non-work con- flict, fulfilment of psychological contract and

227700

Louise P Parkes and Peter H Langford

JJOOUURRNNAALL OOFF MMAANNAAGGEEMMEENNTT && OORRGGAANNIIZZAATTIIOONN Volume 14, Issue 3, July 2008

number of hours worked did not significantly predict organisational commitment. Only organi- sational support for out-of-work activities was positively related to organisational commitment. By contrast, which organisation the graduates worked for (five in total) and functional work area accounted for three times the variance in organisational commitment than all of the work–life balance variables. While not measured directly, this suggests that other aspects of the organisation and immediate work climate have a greater impact on organisational commitment than work-life balance.

HHyyppootthheesseess In addition to assessing employees’ satisfaction with work–life balance compared to other organ- isational climate factors, we aimed to test the fol- lowing hypotheses.

Hypothesis 1a. Work–life balance is positively related to overall employee engagement (including job satisfaction and intention to stay), however:

Hypothesis 1b. Other aspects of organisational climate are more strongly related to employee engagement than work–life balance.

To explore further proposed organisational correlates of work–life balance, we made the fol- lowing tentative predictions.

Hypothesis 2a. Work–life balance is positively related to work practices promoting individual and community health and well-being such as wellness (management of stress and workload),

the importance placed on safety, fair treatment and support from supervisors, help and sup- port from co-workers, support for diversity, and ethical and social responsibility.

Hypothesis 2b. Work–life balance is negatively related to organisational commitment and ‘high commitment’ work practices such as career opportunities, performance appraisal, and a focus on results.

MMEETTHHOODDSS

PPaarrttiicciippaannttss

This study was part of a larger project investigat- ing organisational climate in Australian organisa- tions conducted by The Voice Project at Macquarie University1.

In six waves of data collection over the years 2002 to 2006, samples of approximately ten employees from 1535 work units2 completed anonymous surveys (total of 16,813 respon- dents). Participation of organisations and their employees was voluntary, with consent required from the manager of each work unit and his or her participating employees. In return for their participation, managers received a report sum- marising the results for their work unit, bench- marking their results against all other organisations participating in the study in the same year.

Most of the participating organisations were in the private commercial sector (83% of sample), 11% were public sector, and 6% not-for-profit organisations. A broad range of industries were represented, with the largest being retail trade

227711

Work–life balance or work–life alignment?

Volume 14, Issue 3, July 2008 JJOOUURRNNAALL OOFF MMAANNAAGGEEMMEENNTT && OORRGGAANNIIZZAATTIIOONN

1 The ethical aspects of this research were approved by the Macquarie University Ethics Review Committee (Human Research). Voice Project is a research and consulting practice based at Macquarie University. Non-commercial research is conducted through Macquarie University and commercial research and consulting is conducted through Voice Project Pty Ltd, a legally independent company owned and directed by Dr Peter Langford, operating under licence of Access Macquarie Ltd, the commercial arm of Macquarie University.

2 Participants were recruited by business and organisational psychology students using the survey for an assignment as part of their course requirements. Instructions and consent forms requested that employees surveyed be part of a single work unit or department within an organisation. The exact number of different organisations is not known because organisations were given the option of anonymous participation, however from available data it could be estimated that approximately 20% of work units across all waves of data collection were from organisations already represented by other work units. It is very unlikely that samples were repeatedly surveyed over the years.

(21%), accommodation, hospitality, tourism, cafes and restaurants (12%), finance and insur- ance (9%), and information and communication technologies (8%). Approximately 82% of the organisations had the majority of their employees based in Australia or New Zealand.

Demographic variables were not collected in 2005, however, across the rest of the sample 52.5% of respondents were female, and 47.5% male. In terms of age, there was a larger propor- tion of employees in the 20–29 year age bracket (42%), probably reflecting our method of data collection as part of a business course for students.

In terms of employment type, 60% of partici- pants were full-time employees, 14% were part- time, and 15% long-term casuals. The remaining 11% were contract or short-term casuals. Most employees (56%) were in managerial or profes- sional occupations, while 44% were clerical or service workers, tradespeople, labourers and oth- ers. In terms of education, 36.5% of employees held a Bachelor degree or above.

MMeeaassuurreess ‘The Voice Climate Survey’3 measures 31 different aspects of organisational climate and employee outcomes. Our conceptualisation of organisational climate, and this measure, refers to the more visi- ble or tangible level of organisational culture rep- resented by shared perceptions of work structures and practices, which in turn reflect deeper levels culture such as shared values and beliefs (Hofstede 1991; Rousseau 1990). A total of 102 items are rated on Likert scales ranging from 1 (strongly dis- agree) to 5 (strongly agree), with an additional option of ‘don’t know/not applicable’. The survey shows strong factor structure and internal reliabili- ty, with an average alpha coefficient of 0.82 (Langford 2007). To give an idea of the content covered in the survey, the work practices aggregate into higher-order work systems labelled as Purpose (including practices such as direction, ethics and

role clarity), Property (including resources, facili- ties and technology), Participation (including employee involvement, recognition and develop- ment), People (teamwork, talent, motivation and initiative), Peace (wellness and work–life balance), and Progress (achieving objectives, successful change and innovation, and satisfied customers).

‘Employee engagement’ is a robust higher- order outcome measure included in the Voice Climate Survey (Langford, Parkes & Metcalf 2006) which measures a composite of job satis- faction, organisational commitment and inten- tion to stay on ten items (Cronbach reliability coefficient = 0.92). Construct validity for this measure of employee engagement has been demonstrated by significant correlations with organisational reports of annual turnover and absenteeism (Langford 2007).

‘Work–life balance’ is measured on four items: ‘I maintain a good balance between work and other aspects of my life’; ‘I am able to meet my family responsibilities while still doing what is expected of me at work’; ‘I have a social life out- side of work’; and ‘I am able to stay involved in non-work interests and activities’. It demonstrates good internal reliability with a Cronbach’s alpha coefficient of 0.86. The work–life balance climate scale has shown good divergent validity with an average correlation with other scales of r = 0.25 (compared to an average inter-correlation between climate scales of 0.41).

Demographic variables measured included age, gender, family structure, occupation type and work type (see Table 1 for categories), number of hours worked per week, number of overtime hours per week, salary band, seniority (on a scale of 1 = ‘front line worker’ to 9 = ‘senior executive’) and tenure with current employer.

RREESSUULLTTSS We have reported most results in user-friendly terms of average percentage favourable (% Fav),

227722

Louise P Parkes and Peter H Langford

JJOOUURRNNAALL OOFF MMAANNAAGGEEMMEENNTT && OORRGGAANNIIZZAATTIIOONN Volume 14, Issue 3, July 2008

3 The Voice Climate Survey is copyrighted for commercial use, but freely available to researchers for not-for-profit use. For a copy of the survey, detailed description of survey development and psychometric properties contact the authors.

that is, the average percentage of respondents responding ‘agree’ or ‘strongly agree’ to the work–life balance items. However, all tests of sta- tistical significance were performed on means scores. Given the size of the sample, even small differences were statistically significant. As such, results showing no significant differences between groups stand out for their consistency. Details of post-hoc Tukey HSD comparisons and signifi- cance levels are available from the first author.

SSaattiissffaaccttiioonn wwiitthh wwoorrkk––lliiffee bbaallaannccee Table 1 shows mean scores and percentage favourable figures for the overall sample and some sub-groups. Nearly three-quarters of the overall sample were satisfied with their work–life

balance (73% Fav). Compared to the other aspects of organisational climate, satisfaction with work–life balance ranked fifth highest (behind teamwork, support for diversity, results focus and role clarity). Employees were least satisfied with opportunities for career development, participa- tion and involvement in decision-making and cross-unit communication and co-operation.

We tested the robustness of these results for sub- populations grouped by age, gender, family struc- ture, occupation type (managers and professionals versus others) and work type. Although there were some statistical differences between groups, these were mostly fairly small. For example, managerial and professional employees rated their work–life balance slightly lower than other employees, and

227733

Work–life balance or work–life alignment?

Volume 14, Issue 3, July 2008 JJOOUURRNNAALL OOFF MMAANNAAGGEEMMEENNTT && OORRGGAANNIIZZAATTIIOONN

TTAABBLLEE 11:: RRAATTIINNGGSS OOFF SSAATTIISSFFAACCTTIIOONN WWIITTHH WWOORRKK––LLIIFFEE BBAALLAANNCCEE

nn MMeeaann 00..0000 %% FFaavv

TToottaall ssaammppllee 16784 3.96 0.80 73

OOccccuuppaattiioonn Manager/Professional 3313 3.94 0.80 73 Non-Manager/Professional 2588 4.04 0.80 75

GGeennddeerr Male 6015 3.91 0.80 71 Female 6643 4.03 0.78 76

AAggee Younger than 20 959 4.08 0.80 75 20–29 3948 3.99 0.80 74 30–39 2166 3.89 0.79 71 40–49 1473 3.92 0.79 73 50–59 765 4.05 0.75 79 60 or older 124 4.39 0.70 90

EEmmppllooyymmeenntt Full-time permanent 3902 3.88 0.79 71 TTyyppee Part-time permanent 909 4.08 0.75 77

Contract / fixed term 284 3.92 0.74 73 Long-term casual (>12 Months) 950 4.15 0.73 79 Short-term casual (<12 Months) 438 4.11 0.75 77

FFaammiillyy ttyyppee Single with no dependent children 6086 4.02 0.79 74 Single with dependent children 726 3.86 0.80 70 Married or de facto with no dependent children 2266 3.96 0.80 73 Married or de facto with dependent children 3212 3.93 0.78 73

Note: Not all of the sub-group category data were collected every year, hence some do not add to the total N. No demographic information was collected in 2005.

women were more satisfied than men with their work–life balance (P<0.001). Those between the ages of 30–49 were significantly lower on work–life balance than all other groups, and the over 60s enjoyed significantly higher work–life balance than all other groups. Parents had slightly lower work- life balance than non-parents (P<0.001), mostly due to lower ratings for single parents. Single employees with no dependents enjoyed the greatest work-life balance compared to couples (with and without children) and single parents. Part-time or casual employees reported significantly higher work–life balance than full-time permanent and contract/fixed term employees.

There was virtually no movement in overall satisfaction with work–life balance over the years 2002 to 2006, except for a slight decrease in satis- faction between 2003 and 2004 (from 75% Fav to 73% Fav, P<0.05).

RReellaattiioonnsshhiipp bbeettwweeeenn ddeemmooggrraapphhiicc vvaarriiaabblleess aanndd wwoorrkk––lliiffee bbaallaannccee Number of hours worked, overtime hours worked, salary, seniority and education level were all negatively correlated with work-life balance (r = –0.19, –0.18, –0.10; –0.08, –0.06, respec- tively, all P<.001).

To further investigate the impact of individual variables on satisfaction with work–life balance, we conducted a hierarchical regression analysis. Fol- lowing a logical order of influence, age and gender were entered in step one, parental status and employment type in step two, followed by hours worked and overtime hours in step three and all other individual variables in step four. Since differ- ent demographic data was collected from year to year, a single regression analyses with all of the vari- ables was not possible. A separate regression includ- ing education level and occupation type (managers/ professionals versus others) showed that neither of these variables accounted for additional variance in work–life balance after age, gender, parental status and hours worked (not shown here).

As can be seen in Table 2, the number of regu- lar and overtime hours, along with age, parental status and salary independently predicted work–life balance. Nevertheless, all of the indi- vidual variables entered only accounted for 7.5% of the variance in work–life balance.

Tests of mediation following Baron and Kenny (1986) demonstrated that employment type completely mediated the impact of gender on sat- isfaction with work–life balance4.

227744

Louise P Parkes and Peter H Langford

JJOOUURRNNAALL OOFF MMAANNAAGGEEMMEENNTT && OORRGGAANNIIZZAATTIIOONN Volume 14, Issue 3, July 2008

4 Not included here due to space limitations. Contact the first author for tables of these results.

TTAABBLLEE 22:: MMUULLTTIIPPLLEE RREEGGRREESSSSIIOONN CCOOEEFFFFIICCIIEENNTTSS FFOORR IINNDDIIVVIIDDUUAALL VVAARRIIAABBLLEESS PPRREEDDIICCTTIINNGG WWOORRKK–– LLIIFFEE BBAALLAANNCCEE

BBeettaa �RR 22

SStteepp 11 Gender 0.05* 0.002* Age 0

SStteepp 22 Gendera 0.02*** 0.03*** Age 0.09*** Parental statusb -0.11*** Employment typec 0.15

SStteepp 33 Gender 0 0.04*** Age 0.13*** Parental status -0.10*** Employment type -0.02 Hours worked -0.23*** Overtime hours -0.10***

SStteepp 44 Gender 0.01 0.003* Age 0.11*** Parental status -0.10*** Employment type 0 Hours worked -0.25*** Overtime hours -0.10*** Salary 0.07** Seniority 0.02 Tenure 0

Note: R2 = 0.075, P<0.001 a Males = 1, Female = 2 b Non-parents = 1, Parents = 2 c 1 = Full-time Permanent & Contract/Fixed Term,

2 = Part-time Permanent & Casual *P<0.05, **P<0.01, ***P<0.001

That is, women were more likely to be work- ing part-time or casual and their reported higher work–life balance could be completely attributed to their employment status. Further, while part- time/casual work was associated with higher work–life balance, this was entirely mediated by number of hours worked, as unsurprisingly, full- time employees worked significantly longer hours.

Similarly, older employees were also more like- ly to be parents. When parental status was con- trolled, age was a positive predictor of work–life balance. Thus, work–life balance generally increases with age, except for a drop between the ages 30–49 due to the impact of dependents.

The number of hours worked was the biggest predictor of work–life balance, even controlling for full-time versus part-time work status. While measured and analysed here as an individual vari- able, the number of hours worked is often claimed to be strongly influenced by organisa- tional culture. Analysis of the variance of hours worked within and between organisational units revealed that, indeed, there was greater variance between organisations than within organisations on hours worked (see Table 3).

IImmppoorrttaannccee ooff wwoorrkk––lliiffee bbaallaannccee ffoorr eemmppllooyyeeee eennggaaggeemmeenntt Correlations between all the different aspects of organisational climate and employee engagement are shown in Table 4. Consistent with Hypothe- ses 1a and 1b, although all of these were positive and significant, of the 28 climate scales, work–life balance showed the smallest correlation with engagement. Instead, engagement was high- ly correlated with the management of change and

degree of innovation, belief in the organisation’s mission and values, satisfaction with rewards and recognition, successfully achieving organisational objectives, participation and involvement in deci- sion-making, career opportunities, competence of and communication with leadership, and employee perceptions of customer satisfaction with goods and services.

We conducted a multiple regression analysis to test if work–life balance accounted for any vari- ance in engagement independent of the other aspects of organisational climate. When control- ling for other aspects of organisational climate, higher satisfaction with work–life balance actual- ly predicted lower engagement (see standardised betas in Table 4). In particular, wellness appeared to almost entirely mediate the positive relation- ship between work-life balance and engagement (partial correlation when controlling for wellness = –0.03, P<0.01)

To investigate differences in the importance of work–life balance for engaging different groups of employees, we conducted similar multiple regressions according to gender, age, occupation, employment type and family type. Given the similar patterns of ratings for age groups and employment types, some of the categories were combined for analysis (see Table 5). Work–life balance was not a significant positive predictor of engagement for any of the sub-groups. To test for significant differences between the groups, sepa- rate hierarchical regression analyses for each indi- vidual variable were conducted with the interaction term entered after the direct effects (not shown here, eg gender and work–life balance were enter in step one, followed by the interac-

227755

Work–life balance or work–life alignment?

Volume 14, Issue 3, July 2008 JJOOUURRNNAALL OOFF MMAANNAAGGEEMMEENNTT && OORRGGAANNIIZZAATTIIOONN

TTAABBLLEE 33:: AANNAALLYYSSIISS OOFF VVAARRIIAANNCCEE BBEETTWWEEEENN AANNDD WWIITTHHIINN OORRGGAANNIISSAATTIIOONNSS FFOORR NNUUMMBBEERR OOFF HHOOUURRSS WWOORRKKEEDD PPEERR WWEEEEKK

SSuumm ooff ssqquuaarreess ddff MMeeaann ssqquuaarree FF

Between groups 880204.96 894 984.57 11.68*** Within groups 718673.72 8526 84.29 Total 1598878.68 9420

*P<0.05, **P<0.01, ***P<0.001

tion term gender x work–life balance in step two). Interaction terms were significant for employment type (�R 2 = 0.001, P<0.05), parental status (�R 2 = 0.003, P<0.001) and age (<30 versus 30-49, �R 2 = 0.004, P<0.001; <30 versus 50+, �R 2 = 0.001, P<0.01). As indicated by the correlations in Table 5, work-life balance was more important for engaging full-time and contract employees than part-time and casual

employees, and for parents (both singles and cou- ples) than non-parents. Of the non-parents, work–life balance was more strongly predictive of engagement for couples than singles (�R 2 = 0.001, P<0.01). Finally, work–life balance was less important for engaging employees under 30 than for older age groups.

RReellaattiioonnsshhiipp bbeettwweeeenn wwoorrkk––lliiffee bbaallaannccee aanndd ootthheerr oorrggaanniissaattiioonnaall cclliimmaattee vvaarriiaabblleess To explore the relationship between various aspects of organisational climate and work–life balance, we conducted a multiple regression analy- sis with each of the climate scales, including the subscales of engagement: job satisfaction, organi- sational commitment and intention to stay. Corre- lations and regression results are shown in Table 6. Consistent with Hypothesis 2a, the strongest pre- dictors of satisfaction with work-life balance were wellness and flexibility, followed by teamwork, role clarity, resources, rewards and recognition, and safety. Job satisfaction was a small (but signifi- cant) predictor of work–life balance. Work–life balance was also significantly predicted by diversi- ty and supervision as hypothesised, but not ethics.

Consistent with Hypothesis 2b, organisation commitment was associated with lower work–life balance. Intention to stay with the organisation had the smallest correlation with work–life bal- ance. As hypothesised, performance appraisal and career opportunities were significant negative pre- dictors of lower work–life balance, as was consul- tation and involvement in decision-making.

DDIISSCCUUSSSSIIOONN De Cieri et al (2005) report that over 95% of Australian organisations have implemented some kind of work–life balance strategy. Nevertheless, few workplaces have a high proportion of employees reporting access to each work practice (Gray & Tudball 2002), and only 6% of organi- sations have over 80% of employees using the strategies (De Cieri et al 2005). There appears to be a disconnect between the rhetoric of work–life

227766

Louise P Parkes and Peter H Langford

JJOOUURRNNAALL OOFF MMAANNAAGGEEMMEENNTT && OORRGGAANNIIZZAATTIIOONN Volume 14, Issue 3, July 2008

TTAABBLLEE 44:: CCOORRRREELLAATTIIOONNSS AANNDD MMUULLTTIIPPLLEE RREEGGRREESSSSIIOONN CCOOEEFFFFIICCIIEENNTTSS FFOORR CCLLIIMMAATTEE VVAARRIIAABBLLEESS PPRREEDDIICCTTIINNGG EEMMPPLLOOYYEEEE EENNGGAAGGEEMMEENNTT

CCoorrrreellaattiioonn wwiitthh eennggaaggeemmeenntt aa BBeettaa

Change and innovation 0.58 0.15*** Mission and values 0.57 0.20*** Rewards and recognition 0.55 0.14*** Organisation objectives 0.52 0.08*** Involvement 0.51 0.09*** Career opportunities 0.51 0.17*** Leadership 0.51 0.01ns Customer satisfaction 0.50 0.05*** Supervision 0.47 0.04*** Ethics 0.47 0.03** Organisation direction 0.47 0.00ns Recruitment and selection 0.47 -0.02* Performance appraisal 0.46 0.00ns Wellness 0.45 0.09*** Resources 0.45 -0.01ns Learning and development 0.44 -0.03** Role clarity 0.44 0.08*** Motivation and initiative 0.42 0.08*** Cross-unit cooperation 0.41 -0.07*** Results focus 0.41 0.00ns Safety 0.41 0.02** Diversity 0.41 0.00ns Technology 0.40 0.01ns Processes 0.39 -0.07*** Talent 0.38 0.00ns Facilities 0.37 0.02** Teamwork 0.34 0.03*** Work-life balance 0.21 -0.08***

Note: R2 = 0.56, P<0.001 a All correlations were significant at the P<0.001

level *P<0.05, **P<0.01, ***P<0.001, ns = not significant

balance (with its associated organisational poli- cies) and the empirical evidence on the impor- tance of work–life balance to employees and their uptake of such strategies. Across a broad range of individuals and organisations, this study asked whether employees really are dissatisfied with their current work–life balance and how impor- tant it is for engaging and retaining employees. We also examined the relationship between work–life balance and other work practices in order to clarify the place of work–life balance strategies within organisations.

SSaattiissffaaccttiioonn wwiitthh wwoorrkk––lliiffee bbaallaannccee Contrary to the dire situation reported by others (eg Bond, Galinsky & Swanberg 1998), employ-

ees are in general quite happy with their work–life balance and instead rate other aspects of their work environment as less than ideal. Our results suggest that most organisations are provid- ing an environment that supports satisfactory work–life balance. Of the 28 management prac- tices rated, work–life balance came in as the fifth highest performing climate factor. That is, 73% of employees either agreed or strongly agreed that they were able to meet both their non-work and work responsibilities and have a good balance between their work and other aspects of their lives. In contrast, less than half of employees were satisfied with the organisation’s ability to provide career opportunities, to consult with and involve employees in decisions that affected them, or to

227777

Work–life balance or work–life alignment?

Volume 14, Issue 3, July 2008 JJOOUURRNNAALL OOFF MMAANNAAGGEEMMEENNTT && OORRGGAANNIIZZAATTIIOONN

TTAABBLLEE 55:: CCOORRRREELLAATTIIOONNSS AANNDD MMUULLTTIIPPLLEE RREEGGRREESSSSIIOONN CCOOEEFFFFIICCIIEENNTTSS FFOORR WWOORRKK––LLIIFFEE BBAALLAANNCCEE PPRREEDDIICCTTIINNGG EEMMPPLLOOYYEEEE EENNGGAAGGEEMMEENNTT AACCRROOSSSS DDEEMMOOGGRRAAPPHHIICC GGRROOUUPPSS

CCoorrrreellaattiioonn bbeettwweeeenn wwoorrkk––lliiffee bbaallaannccee aanndd

nn eennggaaggeemmeennttaa nn BBeettaa

TToottaall ssaammppllee 16696 0.21 12867 -0.08***

OOccccuuppaattiioonn Mgr/Prof 3304 0.21 1817 -0.06** Non-Mgr/Prof 2568 0.21 1248 -0.04ns

GGeennddeerr Male 5993 0.19 4494 -0.08*** Female 6607 0.21 4906 -0.07***

AAggee < 30 4890 0.15 3314 -0.08*** 30-49 3631 0.29 2476 -0.05** 50+ 881 0.26 609 0.03ns

EEmmppllooyymmeenntt Full-time permanent and 4169 0.27 3965 -0.05*** ttyyppee contract/fixed term

Part-time permanent 2286 0.17 2117 -0.10*** and casual

FFaammiillyy ssttaattuuss Single with no dependent 6057 0.16 4562 -0.09*** children Single with dependent 720 0.28 538 -0.02ns children Married or de facto with no 2256 0.25 1721 -0.06** dependent children Married or de facto with 3202 0.31 2284 0.00ns dependent children

a All correlations were significant at the P<0.001 level *P<0.05, **P<0.01, ***P<0.001

share information and knowledge between sec- tions in the organisation. These results are fairly consistent with other empirical data showing that in Australia, almost two-thirds of employees are

satisfied with the number of hours they currently work (Thornthwaite 2004). The contrast partial- ly reflects the presentation of work–life balance data from either glass half-empty versus half-full perspectives (or rather one-quarter empty and three-quarters full).

Despite some observations that there appear to be growing numbers of employees who are dissat- isfied (linked with desire for more family time, Thornthwaite 2004), we did not see any increase in dissatisfaction with work–life balance between the years 2002 and 2006, although this may be too small an interval to assess such trends. Simi- larly, our results showed very small differences in work–life balance for age, gender and family structure, with individual variables accounting for only a small proportion of variance in work–life balance (also demonstrated by Frone et al 1992; Greenhaus et al 1997). Work–life bal- ance was no more important for engaging women than men. Women enjoyed a slightly higher work–life balance due to their greater likelihood of part-time or casual employment, which has previously been identified as the ‘do-it-yourself ’ approach to creating a more family-friendly work place (Saltzstein et al 2001). This is also consis- tent with research showing that greater involve- ment with family (as opposed to work) is related to higher quality of life (Greenhaus et al 2003). Although not as exciting as the current debate about generational differences, our data also shows the unsurprising finding that satisfaction with work–life balance is harder to achieve and more important for middle-aged employees with children (especially single parents) and less important for engaging single employees and under 30s (generation ‘Y’). Again, these results are consistent with those of other researchers (eg Dex & Bond 2005).

One interesting finding regarding individual predictors of work–life balance was that although salary was negatively correlated with work–life balance overall, when other variables were con- trolled, particularly the numbers of hours worked, those on higher salaries enjoyed slightly

227788

Louise P Parkes and Peter H Langford

JJOOUURRNNAALL OOFF MMAANNAAGGEEMMEENNTT && OORRGGAANNIIZZAATTIIOONN Volume 14, Issue 3, July 2008

TTAABBLLEE 66:: CCOORRRREELLAATTIIOONNSS AANNDD MMUULLTTIIPPLLEE RREEGGRREESSSSIIOONN CCOOEEFFFFIICCIIEENNTTSS FFOORR OORRGGAANNIISSAATTIIOONNAALL CCLLIIMMAATTEE VVAARRIIAABBLLEESS PPRREEDDIICCTTIINNGG WWOORRKK––LLIIFFEE BBAALLAANNCCEE

CCoorrrreellaattiioonn wwiitthh wwoorrkk––lliiffee bbaallaannccee BBeettaa

Wellness 0.52 0.43*** Flexibilitya 0.43 0.25*** Teamwork 0.34 0.10*** Resources 0.31 0.06*** Role clarity 0.30 0.07*** Diversity 0.28 0.04*** Organisation objectives 0.28 0.04*** Talent 0.28 0.00ns Customer satisfaction 0.28 0.03** Rewards and recognition 0.27 0.06*** Processes 0.27 0.01ns Supervision 0.27 0.03** Motivation and initiative 0.27 0.03* Job satisfaction 0.27 0.03** Safety 0.26 0.05*** Change and innovation 0.25 0.01ns Ethics 0.24 0.01ns Mission and values 0.23 0.02ns Technology 0.23 0.00ns Recruitment 0.22 -0.03* Leadership 0.22 -0.03* Learning and development 0.21 -0.03** Organisation commitment 0.21 -0.08*** Results focus 0.21 -0.01ns Performance appraisal 0.21 -0.03** Facilities 0.20 -0.01ns Cross-unit cooperation 0.19 -0.03** Involvement 0.17 -0.03** Organisation direction 0.16 -0.03** Career opportunities 0.15 -0.03** Intent to stay 0.09 -0.07***

Note: R 2 = 0.34, P<0.001, n = 12851 (excluding flexibility) a Data on flexibility was only collected in 2006. We

repeated the analysis with only the 2006 data, which showed a virtually identical pattern of results to the overall sample. Results for flexibility are from this second analysis (n = 3334, R 2 = 0.35, P<0.001).

*P<0.05, **P<0.01, ***P<0.001, ns = not significant

higher work–life balance. This is consistent with suggestions that those employees in higher socioeconomic brackets can afford more help with their non-work responsibilities through resources such as childcare and household help (Allen et al 2000). However, because we meas- ured satisfaction with work–life balance rather than objective work–life balance, the results might also be attributed to social-exchange theo- ry. Employees receiving higher incomes may also be more satisfied with a lower objective level of work–life balance. Our finding that satisfaction with work–life balance was positively related to satisfaction with rewards and recognition also supports this interpretation of results.

In keeping with previous findings, the greater the number of hours worked each week, the lower the level of work–life balance (Dex & Bond 2005; Sturges & Guest 2004; White, et al 2003). Along with the previous researchers, we treated hours worked as an individual variable affected by factors such as full-time versus part- time status. However, the number of hours employees work was also strongly affected by the organisation they worked for. This provides sup- port for the idea that an organisation may be characterised by its ‘long hours’ culture.

Overall, the results for hours worked, together with the small impact of individual variables on work–life balance suggest that work–life balance can not be treated as an issue only for certain demographic sections of the workforce and strategies focussing on individual needs will have limited success. Instead, policies to improve work–life balance need to be targeted at the broader organisational level.

IIss wwoorrkk––lliiffee bbaallaannccee tthhee kkeeyy ttoo eennggaaggeemmeenntt aanndd rreetteennttiioonn?? Consistent with previous research, we found a small positive correlation between work–life bal- ance and employee engagement and each of its components: job satisfaction, organisational com- mitment and intention to stay (Allen et al 2000; Kossek & Ozeki 1998). However, the bigger pic-

ture tells us something much more important, that is, of the 28 organisational climate factors we included in this study, work–life balance was least related to employee engagement and intention to stay with an organisation. By comparison, aspects of the organisation such as effective change man- agement and belief in the mission and values of the organisation were strongly correlated with engagement.

When controlling all these other aspects of the organisational climate, employees with higher work–life balance actually reported lower engage- ment, particularly demonstrating less commit- ment to their organisations and less intention to stay. These results do not support the contention by Allen et al (2000) that a lack of balance between work and life may cause employees to flee the situation and seek alternative employ- ment in organisations with more supportive work cultures. Our interpretation of these results is that rather than work–life balance promoting engagement and retention, highly engaged employees will sometimes sacrifice work–life bal- ance to achieve organisational goals, especially if the organisation provides a supportive environ- ment in other ways. Our research is consistent with the study highly effective workplaces by Bond et al (2004). While these workplaces were outstanding in terms of their impact on job satis- faction, job engagement/commitment and reten- tion, nearly half of employees in these organisations were still only rated in the middle in terms of mental health. Bond et al attributed these results to stress. That is, work practices that are effective at engaging and retaining employees are not necessarily the same as those that opti- mise employee wellness (see also Langford, Parkes & Abbey 2006).

AAssppeeccttss ooff oorrggaanniissaattiioonnaall cclliimmaattee rreellaatteedd ttoo wwoorrkk––lliiffee bbaallaannccee Several aspects of organisational climate were sig- nificantly related to work–life balance as hypoth- esised. These included the management of workloads to reduce stress, providing flexible

227799

Work–life balance or work–life alignment?

Volume 14, Issue 3, July 2008 JJOOUURRNNAALL OOFF MMAANNAAGGEEMMEENNTT && OORRGGAANNIIZZAATTIIOONN

work arrangements, supportive supervisors and co-workers, support for diversity and a priority placed on a safety within the workplace. In con- trast to work–life balance, wellness was a direct predictor of employee engagement and analyses suggested that any positive association between work–life balance and engagement was mediated by wellness. One limitation of our present research is the cross-sectional nature of our data. While poor work–life balance might be tolerable in the passion and excitement of work now, lon- gitudinal studies may indicate if it leads to burnout and a drop in engagement later down the track. Thus the sustainability poor work–life balance is questionable in the long-term.

It is interesting that contrary to our hypothe- sis, a focus on achieving results for the organisa- tion was not significantly related to work–life balance. However, managerial and human resource practices promoting individual power and progress did predict lower work–life balance: performance appraisal, career opportunities and involvement in decision-making. These findings are important for understanding how to imple- ment the five ‘healthy workplace practices’ rec- ommended to businesses by Grawitch et al (2006). In addition to work–life balance, these include employee involvement and employee development (incorporating career development). Our results show that these practices can actually be in conflict with each other, and a ‘configura- tional’ approach to interrelationships between these work practices is required (Delery & Doty 1996). According to this approach, all practices must be considered together and be consistent with one another. Employee development and involvement can take various forms, some of which will be more consistent with other prac- tices (eg work–life balance) and thus more effec- tive in promoting healthy organisations.

For example, organisations can encourage the perception that working long hours are necessary in order to progress within the organisation (Sturges & Guest 2004) and many employees believe that they are less likely to get ahead in

their careers if they use flexible workplace arrangements (Bond, Thompson, Galinsky & Prottas 2003). Moreover, while flexibility of work hours is strongly advocated as beneficial for achieving work–life balance (Bond et al 2004; Hill, Hawkins, Ferris & Weitzman 2001), some employees can actually use their discretion to spend more time working (White, et al 2003). Thus, it seems that organisations can create a ‘high commitment’ rather than ‘high perform- ance’ culture in which career-involved profession- als are willing to accept the work demands and sacrifice family life for the prestige and economic rewards that come with their career (for example, in accounting, Greenhaus, Parasuraman & Collins 2001). If organisations can create an equitable system of career development and opportunities to participate in decision-making regardless of the hours people work, we believe there will be less pressure for employees who want good work–life balance but don’t want to sacrifice these things.

WWhhaatt wwee ddiidd nnoott mmeeaassuurree

NNoonn--wwoorrkk oouuttccoommeess

We acknowledge that this study followed an indi- vidualistic and economic perspective by measur- ing employee satisfaction with work–life balance and its impact on organisational goals of employ- ee engagement and retention, rather than consid- ering the wider impact of actual work–life balance. For example, data provided by managers’ partners (Russell 2002) can show a very different picture of satisfaction with relationship quality and family life. Powerful evidence exists that long working hours have negative consequences for couple relationships, children, and communities (Pocock 2000; Voydanoff 2001).

There is still a narrow, although increasingly questioned, view that all business practices must be evaluated solely by their impact on economic bottom-line. Our results suggest that proponents of work–life balance should cease pursuing the ‘business case’, and instead persuade organisa-

228800

Louise P Parkes and Peter H Langford

JJOOUURRNNAALL OOFF MMAANNAAGGEEMMEENNTT && OORRGGAANNIIZZAATTIIOONN Volume 14, Issue 3, July 2008

tions to consider personal and societal needs. To the extent that employees and their families are key stakeholders of corporations, work–life bal- ance may be considered a key performance indi- cator of ethical corporate behaviour. Although there was only a small relationship between ethics and work–life balance in the present study, we recommend that work–life balance be understood in terms of constructing and managing work in a socially sustainable way (Lewis et al 2007).

AAttttrraaccttiioonn aanndd rreeccrruuiittmmeenntt While we failed to find a relationship between work–life balance and intention to stay, we did not examine the impact of work–life balance on recruitment. It is clear that although employees may have difficulty prioritising work–life balance, they still perceive it as a desirable goal (Sturges & Guest 2004). Honeycutt and Rosen (1997) found organisations were perceived as particularly attractive places to work if they offered flexible career paths and policies. It is possible that organ- isations may be able to successfully market them- selves as ‘employers of choice’ by offering cultures supportive of work–life balance.

AAnn aalltteerrnnaattiivvee:: WWoorrkk––lliiffee aalliiggnnmmeenntt Most work–life balance research has been in the context of work-family conflict and theoretically positioned work in opposition to family (Eby et al 2005). The term ‘work–life balance’ itself implies an exclusivity and inherent conflict between work and other life domains. Yet any conflict between work and other aspects of life is not solely dependent on objective time alloca- tion, but includes a person’s identification with each role and the extent to which time spent in each is role affirming (Edwards & Rothbard 1999; Thompson & Bunderson 2001). As such, Bruck, et al (2002) found that conflict as a result of time or stress were not significantly related to job satisfaction, but only ‘behaviour’ based con- flict, when behaviour that is effective and neces- sary at work would be counterproductive elsewhere. Such behaviour is conceivably the out-

working of values and beliefs, and is consistent with our result that an employee’s alignment with the values and purpose of their organisation is strongly related to satisfaction and engagement. While not denigrating the weighty social implica- tions of the work–life balance issues, the results of the present study suggest there is room for a more positive integration of work and life to allow the cross-pollination of values, passions, interests and abilities.

Together these results suggest that work–life balance policies are understood best in relation to other work practices and should prompt organi- sations to implement broader organisational strategies. Creating work–life alignment through congruent goals and values, fostering corporate social responsibility, looking after the health and safety of employees, improving reward and per- formance appraisal systems to more accurately reflect performance outcomes (rather than time in the office), developing fair and supportive supervisors, and facilitating participation and involvement in decision-making among all employees, would increase employee engagement and retention generally, reduce the impact of diversity and flow on to greater satisfaction with work–life balance.

RReeffeerreenncceess Adams GA, King LA and King DW (1996)

Relationships of job and family involvement, family social support, and work-family conflict with job and life satisfaction, Journal of Applied Psychology 81: 411–420.

Allen TD (2001) Family-supportive work environments: The role of organisational perceptions, Journal of Vocational Behavior 58: 414–435.

Allen TD, Herst DE, Bruck CS and Sutton M (2000) Consequences associated with work-to- family conflict: A review and agenda for future research, Journal of Occupational Health Psychology 5: 278–308.

Baron RM and Kenny DA (1986) The moderator- mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations, Journal of Personality and Social Psychology 51: 1173–1182.

228811

Work–life balance or work–life alignment?

Volume 14, Issue 3, July 2008 JJOOUURRNNAALL OOFF MMAANNAAGGEEMMEENNTT && OORRGGAANNIIZZAATTIIOONN

Behson SJ (2002) Which dominates? The relative importance of work–family organisational support and general organisational context on employee outcomes, Journal of Vocational Behavior 61: 53-72.

Bittman M and Rice JM (2002) The spectre of overwork: an analysis of trends between 1974 and 1997 using Australian time-use diaries, Labour and Industry, 12: 5–25.

Bond JT, Galinsky E and Hill EJ (2004) When work works - Flexibility: A critical ingredient in creating an effective workplace, Family and Work Institute, New York.

Bond JT, Galinsky E and Swanberg JE (1998). The 1997 National Study of the Changing Workforce, Families and Work Institute, New York.

Bond JT, Thompson C, Galinsky E and Prottas D (2003) Highlights of the [2002] National Study of the Changing Workforce, Families and Work Institute, New York.

Bruck CS, Allen TD and Spector PE (2002) The relation between work–family conflict and job satisfaction: A finer-grained analysis, Journal of Vocational Behavior, 60: 336–353.

Burke RJ (2001) Organizational values, work experiences and satisfactions among managerial and professional women, Journal of Management Development, 20: 346–353.

De Cieri H, Holmes B, Abbott J and Pettit T (2005) Achievements and challenges for work– life balance strategies in Australian organizations, International Journal of Human Resource Management, 16: 90–103.

Delery JE and Doty DH (1996) Modes of theorizing in strategic human resource management: Tests of universalistic, contingency, and configurational performance predictions, The Academy of Management Journal, 39: 802–835.

Dex S and Bond S (2005) Measuring work–life balance and its covariates, Work, Employment and Society 19: 627–637.

Eby LT, Casper WJ, Lockwood A, Bordeaux C and Brinley A (2005) Work and family research in IO/OB: Content analysis and review of the literature (1980–2002), Journal of Vocational Behavior, 66: 124–197.

Edwards JR and Rothbard NP (1999) Work and family stress and well-being: An examination of person-environment fit in the work and family domains, Organisational Behaviour and Human Decision Processes 77: 85–129.

Fox C (2006) Taking the guilt out of leaving early The Financial Review May 23.

Fox ML and Dwyer DJ (1999). An investigation of the effects of time and involvement in the relationship between stressors and work–family conflict, Journal of Occupational Health Psychology 4: 164–174.

Frone MR, Russell M and Cooper ML (1992) Antecedents and outcomes of work–family conflict: Testing a model of the work–family interface, Journal of Applied Psychology 77: 65–78.

Frone MR, Yardley JK and Markel KS (1997) Developing and testing an integrative model of the work-family interface, Journal of Vocational Behavior 50: 145–167.

Grawitch MJ, Gottschalk M and Munz DC (2006) The path to a healthy workplace: A critical review linking healthy workplace practices, employee well-being, and organizational improvements, Consulting Psychology Journal: Practice and Research 58: 129–147.

Gray M and Tudball J (2002) Family-Friendly Work Practices: Differences within and between workplaces (Research Report No 7): Australian Institute of Family Studies.

Greenhaus JH, Collins KM and Shaw JD (2003) The relation between work–family balance and quality of life, Journal of Vocational Behavior 63: 510–531.

Greenhaus JH, Collins KM, Singh R and Parasuraman S (1997) Work and family influences on departure from public accounting, Journal of Vocational Behavior 50: 249–270.

Greenhaus JH, Parasuraman S and Collins KM (2001) Career involvement and family involvement as moderators of relationships between work–family conflict and withdrawal from a profession, Journal of Occupational Health Psychology 6: 91–100.

Haar JM and Spell CS (2004) Programme knowledge and value of work–family practices and organisational commitment, International Journal of Human Resource Management 15: 1040–1055.

Hammer LB, Allen E and Grigsby TD (1997) Work-family conflict in dual-earner couples: Within-individual and crossover effects of work and family, Journal of Vocational Behavior 50: 185–203.

Hand K and Lewis V (2002) Fathers views on family life and paid work, Family Matters 61: 26–29.

Hill EJ, Hawkins AJ, Ferris M and Weitzman M (2001) Finding an extra day a week: The positive influence of perceived job flexibility on work and family life balance. Family Relations 50: 49–58.

228822

Louise P Parkes and Peter H Langford

JJOOUURRNNAALL OOFF MMAANNAAGGEEMMEENNTT && OORRGGAANNIIZZAATTIIOONN Volume 14, Issue 3, July 2008

Hofstede G (1991) Cultures and Organisations: Software of the Mind, McGraw-Hill, London.

Honeycutt TL and Rosen B (1997) Family friendly human resource policies, salary levels, and salient identity as predictors of organizational attraction, Journal of Vocational Behavior 50: 271–290.

Jacobs JA and Gerson K (2001) Overworked individuals of overworked families? Explaining trends in work, leisure and family time, Work and Occupations 28: 40.

Kossek EE and Ozeki C (1998) Work-family conflict, policies, and the job-life satisfaction relationship: A review and directions for organizational behavior– human resources research, Journal of Applied Psychology 83: 139–149.

Lambert SJ (2000) Added benefits: the link between work-life benefits and organisational citizenship behavior, The Academy of Management Journal 43: 801–815.

Langford PH (2007) Psychometrics of the Voice Climate Survey: Evidence for a lower and higher- order factor structure of work practices and outcomes, manuscript submitted.

Langford PH, Parkes LP and Abbey J (2006) Peace, passion and progress: Are they compatible? InPsych 28: 13–15.

Langford PH, Parkes LP and Metcalf L (2006) Developing a structural equation model of organi- sational performance and employee engagement, Proceedings of the Joint Conference of the Australian Psychological Society and the New Zealand Psychological Society, Auckland.

Lewis S, Gambles R and Rapoport R (2007) The constraints of a work-life balance approach: An international perspective, The International Journal of Human Resource Management 18: 360–373.

Lewis S, Rapoport R and Gambles R (2003) Reflec- tions on the integration of paid work and the rest of life, Journal of Managerial Psychology 18: 824–841.

Macran S, Joshi H and Dex S (1996) Employment after Childbearing: A survival analysis. Work, Employment and Society 10: 272–296.

Marks SR and MacDermid SM (1996) Multiple roles and the self: A theory of role balance. Journal of Marriage and the Family 58: 417–432.

Netemeyer RG, Boles JS and McMurrian R (1996) Development and validation of work-family conflict and family-work conflict scales, Journal of Applied Psychology 81: 400–410.

Nielson TR, Carlson DS and Lankau MJ (2001) The supportive mentor as a means of reducing work-family conflict, Journal of Vocational Behavior 59: 364–381.

Parasuraman S and Simmers CA (2001) Type of employment, work–family conflict and well- being: A comparative study, Journal of Organisational Behaviour 22: 551–568.

Peetz D, Townsend K, Russell B, Houghton C, Allan C and Fox A (2003) Race against time: extended hours in Australia, Australian Bulletin of Labour 29: 126–142.

Pocock B (2000) The Effect of Long Hours on Family and Community Life: A survey of existing literature. A report for the Queensland Department of Industrial Relations, Centre for Labour Research, Adelaide University.

Pocock B (2003) The Work–Life Collision: What work is doing to Australians and what to do about it, Federation Press, Sydney.

Pocock B (2005) Work–life ‘balance’ in Australia: Limited progress, dim prospects, Asia Pacific Journal of Human Resources 43:198–209.

Rousseau D (1990) Assessing organisational culture: The case for multiple methods, In B. Schneider (Ed.), Organisational Climate and Culture, Jossey-Bass, San Francisco CA.

Rousseau D (1995) Psychological Contracts in Organisations, Sage, Thousand Oaks CA.

Russell G (2002) Making a difference to work and family outcomes Family Matters 61: 76-78.

Russell G and Bowman L (2000) Work and Family: Current thinking, research and practice, Department of Family and Community Services, Canberra.

Saltzstein AL, Ting Y and Saltzstein GH (2001) Work–family balance and job satisfaction: the impact of family-friendly policies on attitudes of federal government employees, Public Adminis- tration Review 61: 452–467.

Schmidt L (2006) Law culture at a crossroads as X meets Y The Financial Review June 02.

Settoon R, Bennett N and Liden R (1996) Social exchange in organisations: Perceived organisa- tional support, leader–member exchange, and employee reciprocity, Journal of Applied Psychology 81: 219–227.

Smith F (2006) Perth nears perfection The Financial Review October 10.

Sturges J and Guest D (2004) Working to live or living to work? Work–life balance early in the career, Human Resource Management Journal 14: 5–16.

Thompson CA, Beauvais LL and Lyness KS (1999) When work–family benefits are not enough: The influence of work–family culture on benefit utilization, organizational attachment, and work-family conflict, Journal of Vocational Behavior 54: 392–415.

228833

Work–life balance or work–life alignment?

Volume 14, Issue 3, July 2008 JJOOUURRNNAALL OOFF MMAANNAAGGEEMMEENNTT && OORRGGAANNIIZZAATTIIOONN

228844

Louise P Parkes and Peter H Langford

JJOOUURRNNAALL OOFF MMAANNAAGGEEMMEENNTT && OORRGGAANNIIZZAATTIIOONN Volume 14, Issue 3, July 2008

Thompson JA and Bunderson JS (2001) Work– nonwork conflict and the phenomenology of time: Beyond the balance metaphor, Work and Occupations 28: 17.

Thornthwaite L (2004) Working time and work– family balance: A review of employees’ preferences, Asia Pacific Journal of Human Resources 42: 166–184.

Voydanoff P (2001) Incorporating community into work and family research: A review of basic

relationships, Human Relations 54: 1609-1637. White M, Hill S, McGovern P, Mills C and

Smeaton D (2003) ‘High-performance’ management practices, working hours and work–life balance, British Journal of Industrial Relations 41: 175–195.

Received 7 November 2006 Accepted 17 September 2007

eContent Management Pty Ltd, PO Box 1027, Maleny QLD 4552, Australia Tel.: +61-7-5435-2900; Fax. +61-7-5435-2911

www.e-contentmanagement.com

SSeeaa CChhaannggeess,, TTrreeee CChhaannggeess aanndd BBuusshh LLeessssoonnss:: PPoosstt--ccoommppuullssoorryy EEdduuccaattiioonn aanndd RRuurraall RReenneewwaall

AA ssppeecciiaall iissssuuee ooff RRuurraall SSoocciieettyy jjoouurrnnaall VVooll 1199((33)) DDeecceemmbbeerr 22000099

Guest edited by GGeeooffff DDaannaahheerr and RRoobbeerrttaa HHaarrrreevveelldd, Central Queensland University, and PPaattrriicckk DDaannaahheerr, University of Southern Queensland

Over the past 20 years, reforms to Australia’s post-compulsory educational systems have attempted to provide greater opportunities for communities in regional, coastal and rural Australia, delivering wider access to uni- versities and Registered Training Organisations (RTOs), including Institutes of Technical and Further Education (TAFE). At the same time, these communities have been undergoing significant transformation.

While technological, market and climatic variations have challenged some traditional regional industries and communities, others have experienced a ‘sea change’ and/or a ‘tree change’ phenomenon in which the influx of people from metropolitan centres has dramatically affected these communities’ demographic, cultural and eco- nomic life. This special issue of Rural Society will focus on the changing forms of post-compulsory education in rural Australia and their current and potential contributions to sustainable and transformative rural renewal.

wwwwww..rruurraallssoocciieettyyjjoouurrnnaall..ccoomm//aarrcchhiivveess//vvooll//1199//iissssuuee//33//ccaallll//

IInnnnoovvaattiioonn aanndd tthhee CCiittyy -- IInnnnoovvaattiivvee CCiittiieess AA ssppeecciiaall iissssuuee ooff IInnnnoovvaattiioonn:: MMaannaaggeemmeenntt,, PPoolliiccyy && PPrraaccttiiccee VVooll 1100((33)) DDeecceemmbbeerr 22000088

with two themes: Innovation in the City and Innovative Cities Guest edited by JJaannee MMaarrcceeaauu

Adjunct Professor, City Futures Research Centre, University of New South Wales, Sydney Innovation has become the catch cry of policymakers in most OECD countries as they seek to ensure continued prosperity in changed technological and competitive circumstances.

1. Innovation and the City: The role of ‘the city’ in innovation and differences between cities which stimulate innovation has returned to be a focus of public policies of many kinds aimed at economic growth, sustain- ability and social and cultural development – characteristics which create more innovative places, such as knowledge levels, size, governance mechanisms or economic base will be explored – as will institutional arrangements, governance, education, the organization of housing and transport and social and cultural services with critical new emphasis on ecologically sustainable city activities.

2. Innovative Cities: Recently city policymakers and observers have focused on ‘innovative cities’ where gover- nance mechanisms, population characteristics and policies make some cities innovative in many domains. Papers will address specific ‘city’ issues, whole cities or major sections of them such as: economic develop- ment and technological change; transport and city logistics; housing; education delivery and focus; sustain- ability, including the conservation of water and energy; cultural and artistic development; urban governance; urban finance and management - new approaches based on real cities.

wwwwww..iinnnnoovvaattiioonn--eenntteerrpprriissee..ccoomm//aarrcchhiivveess//vvooll//1100//iissssuuee//33//ccaallll//

SS PP EE CC II AA LL II SS SS UU EE SS OO NN UU RR BB AA NN AA NN DD RR UU RR AA LL RR EE NN EE WW AA LL

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.