This month marks 10 years since Sheharbano “Sheri” Sangji undertook her last experiment. On 29 December 2008, the 23-year-old lab technician tried to transfer a small quantity of tert-Butyllithium, which ignites on contact with air. The attempt ended in a fiery catastrophe; she died 18 agonizing days later. A subsequent 95-page state investigative report excoriated the lab’s principal investigator (PI), Patrick Harran, and the University of California, Los Angeles (UCLA), for allowing a recent college graduate to work alone on a risky and intricate task without the training, technique, equipment, or protective gear needed to perform it safely. Unprecedented criminal charges and a yearslong legal case followed.
Apologists for the university called the fire a “tragic accident.” Yet the lab safety experts I spoke with at the time were unanimous that Sheri’s death was not at all a matter of chance. It was wholly foreseeable and preventable.
They also told me that, in contrast, safety standards in industry labs were stringent. The differences with industry, the experts I spoke with agreed, were priorities, practices, and training. Far too many university lab chiefs prioritize “publishing papers and winning Nobel Prizes” over the safety of their workers, James Kaufman, founder of the Laboratory Safety Institute in Natick, Massachusetts, told me.
The more I learned, the more clearly I saw that the scandal involving safety practices in U.S. university labs stretched far beyond this single incident. As safety expert Neal Langerman wrote in the Journal of Chemical Health and Safety (JCHAS), “most academic laboratories are unsafe venues for work and study [and] only by a major change in the way we practice laboratory safety can we improve the situation.”
In the years since, multiple investigative reports have documented that lab safety incidents are not one-off mishaps but symptoms of systemic institutional failure. Multiple prestigious bodies have issued reports advising institutions how to improve their safety cultures. In what Kaufman calls “a miraculous change,” the American Chemical Society has declared lab safety a core professional value. During those same years, however, a number of major lab safety disasters—including the death of an undergraduate at Yale University in 2011, as well as the maiming of a graduate student at Texas Tech University in Lubbock in 2010 and a postdoc at the University of Hawaii in Manoa in 2016—testify that the problem is still not solved.
It’s clear that “Sheri’s death established a legacy that honors her,” Langerman told me recently by email. Though predominantly hopeful, however, the legacy is neither simple nor uniform.
A mixed legacy
“There were two things that should have happened” in the fire’s aftermath: justice and change, said Sheri’s sister Naveen Sangji in a recent interview. A medical student when Sheri died, who is now finishing her training as a trauma surgeon specializing in burns, Naveen has tirelessly advocated for strong consequences for those responsible and for better safety standards everywhere—two factors that, of course, are closely related. Serious penalties for those who permitted the conditions leading to Sheri’s death would have exerted an influence far beyond punishment of particular individuals or institutions. They would have put the scientific community on notice that lab workers’ safety needs to be an overriding priority and neglect would carry a heavy cost.
Both the university and Harran faced felony charges for violations of state occupational safety laws, believed the first ever brought in an academic lab safety case. Conviction would have carried a potential of 4.5 years in prison for Harran. The state’s investigative report found that UCLA “wholly neglected its legal obligations to provide a safe working environment” and that had Harran properly trained, supervised, and equipped her, “Victim Sangji’s death would have been prevented.”
In 2012, the university settled with the Los Angeles district attorney, accepting responsibility for the conditions leading to Sheri’s death and agreeing to a number of reforms. Harran, however, asserted his innocence for another 2 years while the university paid his legal expenses, which ultimately totaled nearly $4.5 million. He, too, then agreed to a settlement that Naveen considers a “travesty” and many in the scientific community criticized as far too lax. He accepted responsibility for the conditions leading to Sheri’s death but avoided a guilty plea, paid a $10,000 fine, and served 5 years’ probation with community service. In September of this year, his probation was ended 9 months early. The family received no notice of the hearing or any opportunity to comment.
“The justice part completely did not happen,” Naveen says. The university chose the lab chief “over the lab tech that they had hired” while the academic community failed to “close ranks around Sheri … as one of their own.”
At the time that the criminal charges were lodged, Harry Elston, editor-in-chief of JCHAS, suggested that this step would make the consequences of serious safety incidents “intolerable and potentially career ending” for PIs. Instead, the scientific world watched Harran remain at liberty and continue running his lab, his funding stream unimpeded. Clearly, the strong message Elston and other safety advocates hoped for has not arrived.
The change part, however, has nonetheless advanced, though with uneven results. Kaufman sees “very good signs of progress in some places,” with institutions “taking this increasingly seriously,” he recently told me. Nathan Watson, CEO of BioRAFT, a company that sells safety management software, also sees institutions working actively to improve safety practice in their labs and safety culture across their campuses. Many have adopted a “dual focus,” fostering shared responsibility for safety at all institutional levels while raising standards in their labs, he says. Rather than concentrating on “holding people to the line” through regulation, he continues, shared responsibility is “about people adopting ownership” of good safety awareness and practice and making these normal and integral parts of their work.
Other institutions, however, are only “trying to inch their way forward or are at an impasse,” stymied by fear that improving safety will raise costs, increase administrative burdens, or both, Watson adds. “Not all PIs buy into their responsibilities yet, but that will come,” Langerman says. Kaufman reports hearing people say, “We’re improving our lab safety program one retirement at a time.”
In a more pessimistic vein, however, Elston recently told Science Careers by email, “After 10 years and several high-profile academic research incidents after the Sangji fatality, … the academy has failed to respond to what I see as an obvious gap in the education of the next generation of scientists”—namely, training in formal hazard and risk analysis. What’s more, he continues, “I’ve already had feedback from workshop participants along the lines of ‘do we have to talk about UCLA again?’… I’m very concerned that we will quickly forget Sheri Sangji’s death and revert to ‘business as usual’ in the research enterprise.”
A chilling incident at a recent safety training workshop that Kaufman conducted underscores this danger. After seeing the U.S. Chemical Safety and Hazard Investigation Board video “Experimenting with Danger,” which highlights Sheri’s story, a young workshop participant spoke up in surprise. “Last year I was a senior in college and I was doing a research project using tert-Butyllithium,” he said. “I didn’t know it was pyrophoric.” No one responsible for this student’s safety, in short, had learned any of the lessons Sheri had so excruciatingly demonstrated years before.
There are, however, ways of changing this situation, these and other experts believe. Next month, in the second part of our series commemorating the anniversary of Sheri’s death, we will examine some of them.