Grappling with graduate student mental health and suicide

In the spring of 2016, Scripps Research Institute California graduate student Anna Owensby texted a friend: “I have this feeling right now that there isn’t really a place for me ... People like me aren’t supposed to get a Ph.D., we are addicts or homeless or in jail.”

Owensby, 26, was a fourth-year graduate student. In her first years at Scripps, she was briefly hospitalized to evaluate her mental health, and she changed lab groups. Her new adviser, Scripps molecular medicine professor Dennis W. Wolan, had assisted in finding her a therapy group that seemed to be helping her. The group was not in Owensby’s insurance network, but Scripps was paying for the sessions. She passed her Ph.D. candidacy exam at the end of 2015, at which time her committee members indicated that she was progressing appropriately and that they had no nonacademic concerns about her.

Then, in the space of just a few weeks in spring 2016, Owensby was banned from Wolan’s lab and told not to contact him or her lab mates. She was encouraged to resign from Scripps and informed that Wolan would not give her a job recommendation.

On April 14, 2016, Owensby died by suicide.

There is no one situation or event that makes a person suicidal, experts in suicide risk and prevention emphasize. “We always want to come up with one cause, but it doesn’t work that way,” says Jill Harkavy-Friedman, vice president of research at the American Foundation for Suicide Prevention. People end their lives when a number of conditions and factors come together to make someone feel overwhelmed and unable to cope.

Nonetheless, there are concrete steps that academic advisers, departments, and schools can take to better support their graduate students’ mental health and reduce the risk of losing people like Owensby.

Risk factors

In 2015, the rate of death by suicide for people ages 20 to 34 in the U.S. was between 15 and 16 people per 100,000, according to the Centers for Disease Control & Prevention. Suicide is one of the five most common causes of death for people ages 15 to 44 years.

Long-term factors that increase the risk that someone might attempt suicide include a psychiatric diagnosis, substance use disorder, childhood trauma, and family history of suicide.

Few studies have investigated mental health problems and suicide in Ph.D. students specifically. One survey of 3,659 students in Belgium suggests that 32% of Ph.D. students are at risk of having or developing a psychiatric disorder such as depression (Res. Policy 2017, DOI: 10.1016/j.respol.2017.02.008). A separate survey of 301 graduate students at Emory University determined that more than 34% likely had moderate to severe depression, while 7.3% reported suicidal thoughts, and 2.3% reported having plans for suicide (Acad. Psychiatry 2014, DOI: 10.1007/s40596-014-0041-y). “The rates of mental health issues in this population are elevated pretty significantly above the general population,” says Nathan L. Vanderford, assistant dean for academic development in the University of Kentucky College of Medicine.

In addition to the long-term factors that increase suicide risk, short-term issues such as life transitions and stress also play a role, experts say. Not unlike starting college, the transition to graduate school can involve significant life upheaval—but without the large number of support staff and peers that undergraduates encounter when moving to campus.

Graduate students instead might move to a new location to live in an apartment alone or with a rarely seen roommate and then interact mostly with a relatively small set of classmates and lab mates in one or two buildings. As time goes on, other events can disrupt a student’s world: An adviser loses a grant, doesn’t get tenure, or chooses to move, or a student fails to pass a milestone such as an oral exam. Meanwhile, family and friends might not understand the unique experience of graduate school.

And the stakes are high. “Students are putting so much effort and energy into pursuing the degree that every single step can carry a lot of weight. Certainly there’s fear of failure or perception of failure,” says Christa Labouliere, a clinical psychology professor and administrator for a suicide prevention program at New York State Psychiatric Institute, which is part of Columbia University Medical Center.

Anna’s story

A year after Anna Owensby died, her mother, Victoria Owensby, sobs when she tries to talk about her. “She was so funny, her heart was so kind, and she was beautiful and brilliant,” Victoria says.

Anna Owensby was valedictorian of her high school class and a talented cellist who enjoyed playing music composed by Johann Sebastian Bach. After high school, she initially attended Cleveland Institute of Music. Students at the institute take nonmusic electives at nearby Case Western Reserve University. After a general chemistry class, Owensby reconsidered her career path—“she didn’t think she could make money playing the cello,” her mother says—and Owensby wound up switching to a chemistry major at Case.

As an undergraduate, Owensby stood out for her enthusiasm and work ethic, says Case chemistry professor Gregory Tochtrop, who taught Owensby organic chemistry before recruiting her to do research in his lab. “She was incredibly smart,” Tochtrop says. After discussing a research idea, she would track down and read all the same papers a more seasoned scientist would look up. “She had the mind-set of ‘I’m going to learn something about this so that I can contribute to it,’ ” he says.

But Owensby had difficulties as well. Growing up in Greenville, S.C., she was a tough child to parent, often losing emotional control when faced with even minor setbacks, her family says. One of Owensby’s aunts, Amy Forton, is a mental health counselor and says Owensby exhibited severe anxiety. “She wasn’t depressed, though,” Forton believes. “She was happy and interested in things in life.”

Nevertheless, Owensby’s difficulty regulating her emotions was significant enough that her mother sought psychological help for her daughter. The therapists “just said she was really, really smart and overly perfectionist and sensitive,” her mother says. “They never said anything about any mental illness, ever.”

There were mental health issues in Owensby’s family. Her father had alcoholism and died of liver disease while she was in college. Tochtrop believes that the challenges of growing up with her father’s alcoholism and then losing him led Owensby to look for father figures. Owensby tried to push for a more personal connection in an otherwise professional relationship with him, Tochtrop says. Early in college, Owensby would turn to Tochtrop to talk for hours about her life history and problems. “It wasn’t anything inappropriate, but I think she was looking for things she didn’t have,” Tochtrop says.

When Owensby was a junior, Tochtrop—at the time an assistant professor who had a wife in law school, a baby, and an approaching tenure decision—felt he had to draw a line. He told Owensby that he couldn’t help her with personal matters, but he was happy to support her career. Owensby seemed to take the message well, and after the conversation “we still had a really good professional relationship,” he says.

Owensby used Case counseling services, Tochtrop says, but he doesn’t know further details. Once, he noticed that she had been cutting herself. Despite her personal challenges, Owensby graduated from Case having earned several university awards, including one for best thesis in undergraduate chemistry research.

Owensby started at Scripps in 2012. Details of her time there are based on materials provided by and interviews with her family and Tochtrop, who stayed in close touch after she left Case. Scripps refused C&EN’s request to interview institute personnel about Owensby and the school’s programs to support students’ mental health. In a statement, Vice President for Communications and Development Cara Miller said, “All universities are faced with the challenge of wanting to provide the utmost support for members of their community who may be in need of assistance, while also maintaining their privacy and adhering to legal requirements.” Owensby’s mother has filed a complaint with Scripps’s accrediting commission because although the institute has given her Owensby’s educational records, it has withheld disciplinary records.

During her first years at graduate school, Owensby diagnosed herself with borderline personality disorder (BPD). In January 2015, she began a form of cognitive behavioral therapy with the support of her adviser, Wolan, and Scripps. The director of the therapy center that Owensby used, Milton Z. Brown of the Dialectical Behavior Therapy Center of San Diego, did not respond to voice mail messages left by C&EN.

The National Institute of Mental Health describes BPD as “a serious mental disorder marked by a pattern of ongoing instability in moods, behavior, self-image, and functioning.” People with BPD often exhibit impulsive actions and have unstable relationships. They go through episodes of anger, depression, and anxiety that can last only a few hours or as long as days. Because Brown did not return C&EN’s calls, C&EN was unable to determine whether Owensby’s therapists agreed with her self-diagnosis.

An unfortunate consequence of Wolan’s support in the face of Owensby’s mental health challenges may have been that Owensby again sought personal fulfillment in a professional relationship. “Looking back, I am just so incredibly grateful for the times you gave me when I felt valued and wanted, as they were the happiest moments of my life,” she wrote to Wolan in a letter she last edited a few days before she died. “I slowly came to love you like the family I had always wanted.”

On Sunday, March 13, 2016, Owensby seemed to be doing well at Scripps. Wolan had texted Owensby, “You are working harder than you likely ever have in your life. And—more importantly—holding it together. You’ve come an incredibly long way. Very proud of you!” according to a document Owensby later prepared for attorneys.

Then Owensby’s life quickly derailed. According to Owensby’s notes, on Wednesday, March 16, Wolan called her into his office to ask if she’d gone through the drawers of a lab mate with whom Owensby had had a contentious relationship. (Six months earlier, Owensby had suspected the lab mate of sabotaging her research materials.)

On Thursday, March 17, Owensby was unexpectedly escorted from her lab to meet with James R. Williamson, then dean of graduate studies at Scripps and now executive vice president for research and academic affairs. According to Owensby’s notes about the meeting, Williamson told her that Wolan had requested her removal because she was disruptive and had gone through other lab members’ personal property. Williamson told Owensby not to return to the lab and had her identification badge and keys confiscated. Then she was escorted to her car.

On Friday, March 18, Scripps director of graduate studies, Dawn Eastmond, contacted Owensby to ask about laboratory materials needed to continue experiments. Eastmond also told her not to contact Wolan or any of her lab mates.

Owensby considered hiring an attorney to challenge Scripps’s actions but decided she couldn’t afford one. At a meeting on Thursday, March 31, 2016, with Williamson and Eastmond, Owensby learned more details of why she’d been shut out of Scripps. In addition to the accusation about going through lab mates’ property, Wolan had discovered that in 2014 Owensby and another lab member had accessed Wolan’s computer and gone through his web browser history.

In her notes for possible attorneys, Owensby maintained that the other, more senior, lab member was the instigator and she was a bystander. She “felt an enormous amount of social pressure to laugh along with” the other lab member when they were searching Wolan’s computer because of a fear of harassment from that person, Owensby wrote. Owensby also alleged that the other lab member had a history of inappropriate behavior, including using lab resources without Wolan’s knowledge to generate materials for a project the lab member planned to undertake elsewhere.

In texts to a friend after the meeting, Owensby wrote about the next steps Scripps planned to take: “They say they have conflicting testimony. They will basically feel it out and [Eastmond], [Williamson], and a [Scripps] lawyer will make a decision. [Eastmond] told me she envisions that either I resign or get terminated.”

Over the next couple of weeks, Owensby continued to text the same friend, expressing concern about how long Scripps would continue to pay her and whether she would have enough money for rent. She also said that she was applying for jobs, but “even McDonald’s wants to talk to your previous supervisor,” she wrote. “I don’t know what I’m gonna do.”

Owensby also met with Brown, one of her therapists. He later told her family that he tried to help her problem solve throughout the entire course of events. Around Tuesday, April 5, Owensby called Eastmond from Brown’s office to ask about getting a recommendation from Wolan. Owensby texted a friend that Eastmond said Wolan didn’t want anything to do with her.

On Wednesday, April 13, 2016, Eastmond called Owensby to suggest that she resign from Scripps. Owensby’s texts to a friend say that Scripps would pay her until the end of April. After Owensby died, Forton, her aunt, spoke with Wolan and Eastmond. According to Forton’s notes about the meeting, Eastmond says she informed Owensby that Scripps would continue to pay for her health insurance through June.

On Wednesday and Thursday, April 13 and 14, Owensby tried several times to contact Wolan. She also spoke for the last time with Tochtrop, who says he thinks about the conversation nearly every day. “She was worried about her career and what she could do,” Tochtrop recalls. “I really tried to emphasize how incredibly smart and talented she was. I said that I would do anything I could to make sure she had a safe landing and found a position.”

Meanwhile, Owensby also spoke with and texted friends. “Can you watch Curie for a little bit tonight,” she texted one on Wednesday, referring to her beloved dog. “I need to have some time alone.”

The friend wasn’t able to take Curie that evening, and Owensby repeated the request on Thursday. Wolan and Eastmond each contacted Owensby’s therapist that day, concerned about Owensby’s behavior. No one contacted Owensby’s family.

Thursday evening, Owensby killed herself.

“We deeply mourn Anna’s sudden passing,” Williamson said in a Scripps obituary. “She was extremely bright and truly passionate about science. Our sympathies go out to her family, close friends and colleagues. We are all struggling to come to terms with what has happened, and we are doing our best to come together to generate a support network for those most affected.”

Cultivating mental health

Given Owensby’s mental health history, she was at high risk for attempting suicide after Scripps personnel barred her from her lab and encouraged her to resign. Ideally, Scripps personnel and Owensby’s therapists would have been aware of her risk for suicide.

Generally, however, “predicting suicide is hard,” says Sidra J. Goldman-Mellor, who studies suicide risk as a professor of public health at the University of California, Merced. “For any given person in a program, it’s going to be really difficult to know if they’re at high risk for suicide, because a lot of people hide it. What that means is that it’s important to make sure there is support for all students.”

That support should go beyond sending an annual e-mail with a list of resources, suicide risk and prevention experts say. Additional steps to take might include holding events such as stress-management workshops or sponsoring thesis-writing support groups. “You want to make it a culture in the department to acknowledge that graduate school is hard and there are things you can do to help cope with it,” Columbia’s Labouliere says. “Maybe then people won’t get to the point of feeling trapped.”

Some chemistry departments are making efforts to reduce stress where they can, foster more open discussions about mental health, and encourage students to talk with each other.

When Philippe Buhlmann, a chemistry professor at the University of Minnesota, became his department’s director of graduate studies, he made it his mission in that role to promote mental health. He started by working with UMN’s mental health services to develop a survey of chemistry graduate students to figure out their primary stressors. The most intense ones turned out to be the interrelated needs to publish and to find a job upon graduation. “We can’t eliminate those,” Buhlmann says. “We want 100% of our students to get a job.”

But Buhlmann and colleagues did identify some measures that the department could take to reduce students’ stress. For example, the department reconsidered annual student progress reports and self-evaluations. Without guidance, some students are harder on themselves than they should be. “They come up with their own scale that by all objective assessment is twisted,” Buhlmann says. Now, after the students self-evaluate, their advisers must also weigh in to concur or correct the student’s assessment, clarifying expectations and improving communication in the process.

The department also now requires that fourth-year students meet with their thesis committee members to talk about research progress and career aspirations. “Students were worried that this would be just another hoop” to jump through, Buhlmann says. “But I’m getting very positive feedback. It gives students other perspectives and pushes them to think more realistically” about how to reach their postgraduation goals.

Buhlmann also formed a student-run group called the Community of Chemistry Graduate Students (CCGS) that has a $3,000 budget to plan events promoting physical, mental, and social health. Activities so far have included hosting weekly runs and biweekly social hours, creating videos about depression, and holding panel discussions on topics such as stress reduction or how to write a résumé. “If we have too many things that are too strongly focused on mental health, then we don’t get good turnout,” says third-year chemistry graduate student and CCGS chair Evan Anderson. “If we mix things up, then we get better attendance overall.” A recent picnic had 150 attendees, which included more than half of the graduate students in the department.

Similar efforts are afoot in the department of chemistry and chemical biology at Harvard University. That department tried to implement changes after graduate student Jason Altom died by suicide in 1998—at the time, the department’s fourth suicide death in eight years. Some of those efforts, such as a biweekly catered dinner, didn’t stick.

A switch to multiprofessor doctoral committees did last—but different faculty approached committee responsibilities differently, such as how often they met with students. Starting with this fall’s incoming class, however, there will be some standardization: To alleviate the stress of having to choose committee members and corral them for meetings, those tasks will fall to the department’s director of graduate studies. Starting in their second year, students will have annual meetings with their committee in April, and the department has developed some basic guidelines about what those meetings should cover.

Additionally, the department has formed a Community Committee composed of faculty, staff, students, and postdocs who are charged with promoting department cohesiveness and communication. It has sponsored a Thanksgiving dinner for people remaining in the Cambridge, Mass., area during the holiday; sessions on mindfulness; chair massages; and “Fruitful Wednesdays” during nutritional awareness month, when students can gather and chat while picking up free fruit to eat. The committee does not yet have a budget because it is new, but “money is not a constraint; we will do things as they make sense,” says Elizabeth A. Lennox, director of laboratories and codirector of graduate studies in the department.

Meanwhile, the department’s Graduate Student & Postdoc Council has a $19,000 annual budget for social gatherings and what Lennox calls “fun but academically constructive events.” For example, the committee organized faculty to give talks about their experiences applying for academic jobs, running a lab, and rising from junior to senior faculty.

The increased efforts toward supporting mental health and building community were a priority set by professor Charles M. Lieber when he became department chair in 2015. A culture survey had shown that morale was down across the board in the department, something that Lieber attributes to a lack of transparency. Lieber emphasizes that expectations remain high for Harvard students, postdocs, and faculty. But promoting openness helps alleviate some of the issues created when people stay in their labs and offices and don’t talk with each other, he adds.

Crisis response

Although general programs to promote mental, physical, and social health are important, it is also critical for people to know how to respond when someone is in acute distress. Owensby was already connected with therapists when her life upended. But “most people who are depressed or suicidal don’t make it to a counseling center,” says Julie Cerel, a psychologist and professor at the University of Kentucky College of Social Work and president of the American Association of Suicidology. She says training people who are naturally part of students’ environment—“gatekeepers” such as lab managers, faculty, and other departmental staff—can help identify those in need of intervention. In departments in which people aren’t used to discussing mental health, training can also provide scripts they can use to begin conversations.

Such training, however, doesn’t mean that faculty or other department members become therapists. “There are professional limitations and personal boundaries that should not be crossed,” UMN’s Buhlmann emphasizes. He was a pilot trainee in a now-expanding effort at UMN to train mental health advocates within each academic unit. The advocates are taught a “4R” approach: recognize when people are in trouble, assume the role of listener, respond by giving them options, and direct them to appropriate resources.

Additionally, Buhlmann brought in UMN mental health services staff to train graduate student teaching assistants to recognize the signs of stress and mental health problems in chemistry undergraduates and how to direct them to appropriate help. These skills could also transfer to interactions with peers.

At Harvard, Lennox arranged for the school’s mental health services to conduct mental health awareness training this spring for the lab group administrators within her department. These administrators might be physically present in labs and see members in person more often than faculty do, Lennox says.

Lennox is also working with mental health services to develop a mental health curriculum to add to the department’s routine training for students and postdocs. She’d like to include topics such as signs of mental distress, suicide risk awareness, and campus resources.

As part of an overall campus approach, schools should also ensure that on-campus clinicians are trained in suicide prevention response and how to implement measures that can quickly ensure someone’s safety, Labouliere says. Those interventions include determining how to restrict access to whatever method someone has chosen to use to attempt suicide. One of the reasons someone is suicidal is an impaired ability to solve problems. Consequently, once someone has a suicide plan, he or she is unlikely to change it when faced with a barrier. Another intervention is safety planning, which involves talking through and writing down what actions to take when suicidal thoughts take over.

“Suicidal thinking does not last forever,” Labouliere notes. In the moment, people might find their pain unbearable, but if they can get through that crisis, then they likely can engage in therapy, consider their options, and work to address whatever factors led to the situation. She adds that it’s important to tell people in crisis that effective, research-supported treatments are available. “A lot of folks that feel suicidal are so hopeless and feel that their problems are so intractable that the message that there are treatments can be very powerful,” she says.

If someone does die by suicide, it’s important for the department and school to step up actions to pay attention to their community. Cerel’s research suggests that after someone dies by suicide, as many as 30 people might have their lives majorly disrupted. “It’s not just a handful of close family members whose lives are changed,” she says. Also, “I think that suicides can really hit science departments hard because they’re not used to talking about feelings,” she says.

Since Buhlmann has started bringing up the topic of mental health in his department, “there’s been tremendously positive feedback from other faculty,” he says. “Anyone who’s been a chemistry professor long enough knows students in their group or department who have severely struggled,” he adds. “There are so many good people out there who want to help but don’t know how. When you start to talk about it, everyone engages.” 

CORRECTION: This story was updated on Aug. 23, 2017, to correct the date of Jason Altom's death. He died in 1998, not 1988.