CAREERS AND RECRUITMENT

Evidence for a mental health crisis in

graduate education
© 2018 Nature America, Inc., part of Springer Nature. All rights reserved.

Teresa M Evans1, Lindsay Bira2, Jazmin Beltran Gastelum3, L Todd Weiss4 & Nathan L Vanderford4,5
With mental illness a growing concern within graduate education, data from a new survey should prompt both
academia and policy makers to consider intervention strategies.

T here is a growing cry for help from graduate

students across the globe who struggle with
significant mental health concerns1. Despite
trainee population, we deployed a comprehen-
sive survey that included clinically validated
scales for anxiety (GAD07) and depression
with perceived work–life balance and mentor-
ship quality.
The transgender and/or gender-noncon-
increased discussion of the topic, there remains (PHQ09) via social media and direct email. forming population faces an increased risk
a dire need to resolve our understanding of the We surveyed a total of 2,279 individuals (90% of depression and anxiety6. Further, women
mental health issues in the trainee population. PhD students and 10% Master’s students). consistently are more likely to suffer from
Recent research on mental health in the Respondents were from 26 countries and mental health disorders than men7. Our
trainee population has focused on within-insti- 234 institutions (Supplementary Tables 1 results corroborate these findings within the
tution cohorts, such as the 2014 UC Berkeley and 2) and represented diverse fields of study graduate trainee population; both transgender/
report2, which found that 43–46% of graduate including, biological/physical science (38%), gender-nonconforming and female gradu-
students in the biosciences were depressed, and engineering (2%), humanities/social sciences ate students are significantly more likely to
the 2015 University of Arizona report3, which (56%) and “other” (4%). The data presented experience anxiety and depression than their
found that a majority of doctoral students here demonstrate that the graduate trainee male graduate student counterparts (Fig. 1b).
reported “more than average” current stress community has a considerable prevalence Our study found that the prevalence of anxi-
or “tremendous” stress and endorsed school of individuals with anxiety and depression. ety and depression in transgender/gender-
and education-related issues as the most sig- Although this is a convenience sample in which nonconforming graduate students was 55%
nificant contributors to their stress. Although respondents who have had a history of anxi- and 57%, respectively, compared with their
these studies demonstrate the mental health ety or depression may have been more apt to nongender minority counterparts (43% and
concerns in this population, more research is respond to the survey, the data should prompt 41% in females and 34% and 35% in males,
needed to better define the prevalence of men- both academia and policy makers to consider respectively).
tal health issues and the role of key variables intervention strategies. Work–life balance is associated with physical
such as gender, mentorship relationships and and mental well-being, and among academic
perceived work–life balance on susceptibility Mental health crisis in the graduate faculty there are long work hours and vary-
to mental health struggles in the trainee popu- student population ing degrees of work–life balance as a result8.
lation. Our results show that graduate students are However, little is known about work–life bal-
In order to address gaps in the understand- more than six times as likely to experience ance in the graduate trainee population. Our
ing of mental health prevalence in the graduate depression and anxiety as compared to the gen- respondents were asked if they agree with the
eral population. Forty-one percent of graduate statement, “I have a good work–life balance.”
1Graduate School of Biomedical Sciences students scored as having moderate to severe Of the graduate students who experienced
and Department of Pharmacology, UT anxiety on the GAD07 scale as compared to moderate to severe anxiety, 56% did not agree
Health San Antonio, San Antonio, Texas, 6% of the general population, as demonstrated with this statement versus 24% who agreed
USA. 2Department of Psychiatry, UT Health previously4. Additionally, 39% of graduate stu- (Fig. 1c). Additionally, of those graduate stu-
San Antonio, San Antonio, Texas, USA. dents scored in the moderate to severe depres- dents with depression, 55% did not agree with
3Department of Biology, St. Mary’s University, sion range in our study, as compared to 6% of the statement versus 21% who agreed. These
San Antonio, Texas, USA. 4Markey Cancer the general population measured previously results show that good work–life balance is sig-
Center, University of Kentucky, Lexington, with the same scale5 (Fig. 1a). In order to nificantly correlated with better mental health
Kentucky, USA. 5Department of Toxicology & better understand the factors influencing this outcomes.
Cancer Biology, University of Kentucky College significantly higher prevalence of anxiety and Lastly, the principal investigator (PI)/
of Medicine, Lexington, Kentucky, USA. depression in this population compared to the advisor relationship with graduate students
e-mail: [email protected] or nathan. general population, we also examined their affects the quality of training in graduate
[email protected] prevalence across genders and in association education9. Therefore, it is alarming to find

282 VOLUME 36 NUMBER 3 MARCH 2018 NATURE BIOTECHNOLOGY

 CAREERS AND RECRUITMENT

a c Anxiety
health will result in a competitive advan-
Depression
60 tage for institutions and increase retention,

Percentage of students
Anxiety 50 56 55 thus strengthening the bioscience workforce
41% 40 pipeline.
30

Depression 20 24
21
Call for cultural change. There is a grave need
39%
10 to educate faculty about the impact of gradu-
0
Unhealthy Healthy ate education on the mental health of gradu-
Work-life balance ate students. The NIH Office of Intramural
Training and Education recently used a train-
b d Disagree
the-trainers model at a highly successful event
Agree
Anxiety (%) Anxiety (%) where career development leaders trained
50
Provides mentorship the next generation of career development
© 2018 Nature America, Inc., part of Springer Nature. All rights reserved.

36
Male 34
Provides ample support
35
49
professionals on topics that are vital to their
Female 43 Positive emotional impact
34
48 success11. Using a similar train-the-trainers
Asset to career
53 model, faculty and administration could be
Transgender 55 37

Feel valued by mentor

55 trained by mental health professionals to
34
ensure that those in a position to directly sup-
Depression (%) Depression (%)
port graduate students are equipped to iden-
Provides mentorship 50 tify student mental health needs and provide
Male 35 33
adequate guidance and referrals as needed.
50
Provides ample support
Female 41
32
47
This model could also be used by career devel-
Positive emotional impact
31 opment professionals to provide faculty with
54
Transgender 57 Asset to career
35 training in the fundamental skills and knowl-
56
Feel valued by mentor
30 edge needed to mentor today’s PhD students in
the vast and ever-changing job market.
Figure 1 The prevalence of anxiety and depression within the population of graduate students studied. Many in academia have spoken out about
(a) Overall prevalence. (b) Prevalence of anxiety and depression by gender. (c) Effect of perceived work– their own mental illnesses and the stigma they
life balance. (d) Effect of relationship with mentor (see Supplementary Data). faced within the academic community. Fears of
not gaining tenure or judgments being cast by
that of graduate students who experienced graduate institutions and the US National colleagues are just a few of the major concerns
anxiety and/or depression, 50% did not agree Institutes of Health (NIH). The NIH has faced by those who suffer from mental health
with the statement that their PI or advisor provided a model through efforts such as disorders in academia12. Fundamentally, data
provides “real” mentorship as compared the Broadening Experiences in Scientific such as ours call for a shift in the culture within
to 36% and 33% who agreed, respectively Training (BEST) award program and within academia to eliminate the stigma and ensure
(Fig. 1d; see Supplementary Data). The the NIH campus through establishing the that students are not reluctant to communicate
majority of those who experienced anxiety Office of Intramural Training and Education openly with their faculty advisors.
(49%) and depression (50%) disagreed with that houses many programs to support the Work–life balance is hard to attain in a
the statement that their PI/advisor pro- career development of intramural trainees. culture where it is frowned upon to leave the
vides ample support versus 35% and 32% Although these resources have served as a laboratory before the sun goes down. The
who agreed. When asked if their PI/advisor scaffold for the development of local pro- stress of increased pressure to produce data
positively impacts their emotional or mental grams at many graduate institutions, there in order to compete for funding has increased
well-being, 48% with anxiety and 47% with are still many universities that lack adequate exponentially, and science fields are feeling
depression did not agree versus 34% and career and professional development pro- immense pressure13. Faculty and administra-
31% who agreed. Further, the majority of grams. tors must set a tone of self-care as well as an
graduate students experiencing anxiety and/ Career development encompasses many efficient and mindful work ethic in order to
or depression did not agree with the state- skills that are vital to graduate student suc- move to a healthier work and education envi-
ments that their PI/advisor is an asset to their cess but is often not included under this ronment.
career (53% and 54% versus 37% and 35% umbrella as mental health10. We believe that
who agreed) or that they feel valued by their our data support the need for a more com- Call to action. These data demonstrate a criti-
mentor (55% and 56% versus 34% and 30% prehensive development of more adequate cal need for additional studies that investigate
who agreed). These data indicate that strong, career development offices across institu- intervention strategies that could address the
supportive and positive mentoring relation- tions to ensure that they can serve to edu- mental health crisis in the graduate trainee
ships between graduate students and their cate students about mental health and refer population. It is only with strong and validated
PI/advisors correlate significantly with less those who need mental health support. interventions that academia will be able to pro-
anxiety and depression. Establishing this infrastructure is a founda- vide help for those who are traveling through
tional step for institutions to ensure that stu- the bioscience workforce pipeline.
Intervention strategies dents are prepared to become the biomedical
Enhanced access to mental health support. workforce of tomorrow. Further, creating Conclusions
There has been a keen focus on building the programs and interventions within career Our studies show a high prevalence of anxiety
biomedical workforce of tomorrow among development offices that promote mental and depression in a diverse graduate student

NATURE BIOTECHNOLOGY VOLUME 36 NUMBER 3 MARCH 2018 283

 CAREERS AND RECRUITMENT

sample. The strikingly high rates of anxiety and D. Weiss, Graduate School of Biomedical Sciences 6. Dhejne, C., Van Vlerken, R., Heylens, G. & Arcelus, J.
(GSBS) at UT Health San Antonio for his guidance; Int. Rev. Psychiatry 28, 44–57 (2016).
depression support a call to action to estab- 7. Eaton, N.R. et al. J. Abnorm. Psychol. 121, 282–288
and the GSBS for providing partial funding for the
lish and/or expand mental health and career (2012).
study. Lastly, the authors dedicate this body of work
development resources for graduate students 8. Hogan, V. et al. Ir. J. Psychol. 35, 133–150 (2014).
to all graduate trainees who persevere in silence to 9. Tenenbaum, H. et al. J. Vocat. Behav. 59, 326–341
through enhanced resources within career overcome mental health struggles. (2001).
development offices, faculty training and a 10. Fuhrmann, C.N. Hum. Gene Ther. 27, 871–879
change in the academic culture. COMPETING FINANCIAL INTERESTS (2016).
The authors declare no competing financial interests. 11. NIH Office of Intramural Training & Education. How
to Teach and Advise on Career Development Topics
Note: Any Supplementary Information and Source Data 1. Gewin, V. Nature 490, 299–301 (2012). for the Next Generation of Biomedical Scientists: a
files are available in the online version of the paper. 2. UC Berkeley Graduate Assembly. Graduate Student Train-the-Trainers Event https://www.training.nih.gov/
Happiness and Well-being Report http://ga.berkeley. train_the_trainers_2016 (NIH, 2016).
edu/wellbeingreport (2014). 12. Pryal, K.R.G. Disclosure Blues: Should You Tell
ACKNOWLEDGMENTS Colleagues About Your Mental Illness? ChronicalVitae
3. Smith, E. & Brooks, Z. Graduate Student Mental Health
The authors thank D. Story, Markey Cancer Center (University of Arizona, 2015). https://chroniclevitae.com/news/546-disclosure-blues-
© 2018 Nature America, Inc., part of Springer Nature. All rights reserved.

Research Communications Office, for graphic design 4. Löwe, B. et al. Med. Care 46, 266–274 (2008). should-you-tell-colleagues-about-your-mental-illness
assistance; P. Chambers of Versatile PhD for her 5. Kocalevent, R.D., Hinz, A. & Brähler, E. Gen. Hosp. (13 June 2014).
comments on and aid in distributing the study survey; Psychiatry 35, 551–555 (2013). 13. Powell, K. Nature 538, 446–449 (2016).

284 VOLUME 36 NUMBER 3 MARCH 2018 NATURE BIOTECHNOLOGY