Bad Blood, Worse Ethics
By Sidney PerkowitzSeptember 7, 2018
Bad Blood by John Carreyrou
No one enjoys the process and some truly fear it. It is worth the pain only because blood tests can reveal conditions you and your doctor need to know about, from your body’s chemical balance to signs of disease. Besides the distress, there is the cost. One recent US study found a median cost of $100 for a basic blood test, with much higher costs for more sophisticated analyses. This adds up to a global blood-testing market in the tens of billions of dollars.
So when 19-year-old Elizabeth Holmes dropped out of Stanford in 2003 to realize her vision of less painful, faster, and cheaper blood testing, she quickly found eager investors. An admirer of Steve Jobs, she pitched her startup company Theranos (a portmanteau of “therapy” and “diagnosis”) as the “iPod of health care.” But instead it became the Enron of health care, a fount of corporate deceit that finally led to a federal criminal indictment of Holmes for fraud. She now awaits trial.
Bad Blood: Secrets and Lies in a Silicon Valley Startup is John Carreyrou’s gripping story of how Holmes’s great idea led to Silicon Valley stardom and then into an ethical quagmire. Carreyrou is the Wall Street Journal reporter who first revealed that Theranos was not actually achieving what Holmes claimed, though her company had been valued at $9 billion. Her net worth neared $5 billion, and her deals with Walgreens and Safeway could have put her technology into thousands of stores, thus measuring the health of millions, until Carreyrou showed that the whole impressive edifice rested on lies.
It didn’t start that way. Impressed by her creativity and drive, Holmes’s faculty mentor at Stanford, writes Carreyrou, told her to “go out and pursue her dream.” That required advanced technology, whereas Holmes’s scientific background consisted of a year at Stanford and an internship in a medical testing lab. Nevertheless, she conceived and patented the TheraPatch. Affixed to a patient’s arm, it would take blood painlessly through tiny needles, analyze the sample, and deliver an appropriate drug dosage. Her idea was good enough to raise $6 million from investors by the end of 2004, but it soon became clear that developing the patch was not feasible.
Holmes didn’t quit. Her next idea was to have a patient prick a finger and put a drop of blood into a cartridge the size of a credit card but thicker. This would go into a “reader,” where pumps propelled the blood through a filter to hold back the red and white cells; the pumps then pushed the remaining liquid plasma into wells, where chemical reactions would provide the data to evaluate the sample. The results would quickly be sent wirelessly to the patient’s doctor. Compact and easy to use, the device could be kept in a person’s home.
In 2006, Holmes hired Edmond Ku, a Silicon Valley engineer known for solving hard problems, to turn a sketchy prototype of a Theranos 1.0 card and reader into a real product. Running a tiny volume of fluid through minute channels and into wells containing test reagents was a huge challenge in microfluidics, hardly a Silicon Valley field of expertise. As Carreyrou describes it:
All these fluids needed to flow through the cartridge in a meticulously choreographed sequence, so the cartridge contained little valves that opened and shut at precise intervals. Ed and his engineers tinkered with the design and the timing of the valves and the speed at which the various fluids were pumped through the cartridge.
But Ku never did get the system to perform reliably. Holmes was unhappy with his progress and insisted that his engineers work around the clock. Ku protested that this would only burn them out. According to Carreyrou, Holmes retorted, “I don’t care. We can change people in and out. The company is all that matters.” Finally, she hired a second competing engineering team, sidelining Ku. (Later, she fired him.) She also pushed the unproven Theranos 1.0 into clinical testing before it was ready. In 2007, she persuaded the Pfizer drug company to try it at an oncology clinic in Tennessee. Ku fiddled with the device to get it working well enough to draw blood from two patients, but he was troubled by the use of this imperfect machine on actual cancer patients.
Meanwhile the second team jettisoned microfluidics, instead building a robotic arm that replicated what a human lab tech would do by taking a blood sample from a cartridge, processing it, and mixing it with test reagents. Holmes dubbed this relatively clunky device the “Edison” after the great inventor and immediately started showing off a prototype. Unease about the cancer test, however, had spread, and some employees wondered whether even the new Edison was reliable enough to use on patients.
As Carreyrou relates, Holmes’s management and her glowing revenue projections, which never seemed to materialize, were beginning to be questioned, in particular by Avie Tevanian, a retired Apple executive who sat on the Theranos board of directors. Holmes responded by threatening him with legal action. Tevanian resigned in 2007, and he warned the other board members that “by not going along 100% ‘with the program’ they risk[ed] retribution from the Company/Elizabeth.”
He was right. Holmes was ruthless about perceived threats and obsessive about company security, and marginalized or fired anyone who failed to deliver or doubted her. Her management was backed up by Theranos chief operating officer and president Ramesh “Sunny” Balwani. Much older than Holmes, he had prospered in the dotcom bubble and seemed to act as her mentor (it later emerged that they were in a secret relationship). To employees, his menacing management style made him Holmes’s “enforcer.”
Worst of all, Holmes continued to tout untested or nonexistent technology. Her lucrative deals with Safeway and Walgreens depended on her assurance that the Edison could perform over 200 different blood tests, whereas the device could really only do about a dozen. Holmes started a program in 2010 to develop the so-called “miniLab” to perform what she had already promised. She told employees: “The miniLab is the most important thing humanity has ever built.” The device ran into serious problems and in fact never worked.
Despite further whistleblowing efforts, Holmes and Balwani lied and maneuvered to keep the truth from investors, business partners, and government agencies. Eminent board members like former US Secretaries of State George Shultz and Henry Kissinger vouched for Holmes, and retired US Marine Corps General James Mattis (now President Trump’s Secretary of Defense) praised her “mature” ethical sense. What the board could not verify was the validity of the technology: Holmes had not recruited any directors with the biomedical expertise to oversee and evaluate it. But others were doing just that, as Carreyrou relates in the last part of the book.
In 2014, Carreyrou received a tip from Adam Clapper, a pathologist in Missouri who had helped Carreyrou with an earlier story. Clapper had blogged about his doubts that Theranos could run many tests on just a drop of blood. He heard back from other skeptics and passed on their names to Carreyrou. After multiple tries, Carreyrou struck gold with one Alan Beam, who had just left his job as lab director at Theranos.
After Carreyrou promised him anonymity (“Alan Beam” is a pseudonym), Beam dropped two bombshells. First, the Edisons were highly prone to error and regularly failed quality control tests. Second and more startling, most blood test results reported by Theranos in patient trials did not come from the Edisons but were secretly obtained from standard blood testing devices. Even these results were untrustworthy: the small Theranos samples had to be diluted to create the bigger liquid volumes required by conventional equipment. This changed the concentrations of the compounds the machines detected, which meant they could not be accurately measured. Beam was worried about the effects of these false results on physicians and patients who depended on them.
Carreyrou knew he had a big story if he could track down supporting evidence. In riveting detail, he recounts how he chased the evidence while Holmes and Balwani worked to derail his efforts. Theranos hired the famously effective and aggressive lawyer David Boies, who tried to stifle Carreyrou and his sources with legal threats and private investigators. Holmes also appealed directly to media magnate Rupert Murdoch, who owned the WSJ through its parent company and had invested $125 million in Theranos. Holmes told Murdoch that Carreyrou was using false information that would hurt Theranos, but Murdoch declined to intervene at the WSJ.
Carreyrou’s front page story in that newspaper, in October 2015, backed up Beam’s claims about the Edisons and the secret use of conventional testing. There was an immediate uproar, but Holmes and Balwani fought back, denying the allegations in press releases and personal appearances, and appealing to company loyalty. At one memorable meeting after the story broke, Balwani led hundreds of employees in a defiant chant: “Fuck you, Carreyrou! Fuck you, Carreyrou!”
Problems arose faster than Holmes could deflect them. When Theranos submitted poor clinical data to the FDA, the agency banned the “nanotainer,” the tiny tube used for blood samples, from further use. The Centers for Medicare and Medicaid Services, the federal agency that monitors clinical labs, ran inspections that echoed Carreyrou’s findings, and banned Theranos from all blood testing. Eventually the company had to invalidate or fix nearly a million blood tests in California and Arizona. In another blow, on March 14, 2018, the Securities and Exchange Commission charged Theranos, Holmes, and Balwani with fraud. Holmes was required to relinquish control over the company and pay a $500,000 fine, and she was barred from holding any office in a public company for 10 years.
Carreyrou tells this intricate story in clear prose and with a momentum worthy of a crime novel. The only flaw, an unavoidable one, is that keeping track of the many characters is not easy — Carreyrou interviewed over 150 people. But he makes sure you know who the moral heroes are of this sad tale.
Two among them are Tyler Shultz, grandson of Theranos board member George Shultz, and Erika Cheung, both recent college grads in biology. While working at Theranos, they noticed severe problems with the blood tests and the company’s claims about their accuracy. They got nowhere when they took their concerns to Holmes and Balwani, and to the elder Shultz. After resigning from the company out of conscience, they withstood Theranos’s attempts at intimidation and played crucial roles in uncovering what was really going on.
But many with a duty to ask questions did not. A board of directors supposedly exercises “due diligence,” which means ensuring that a company’s financial picture is sound and that the company’s actions do not harm others. The Theranos board seemed little interested in either function, as Carreyrou’s story of Tyler Shultz and his grandfather shows. Later on, in a 2017 deposition for an investor’s lawsuit against Theranos, the elder Shultz finally did admit his inaction. He testified under oath that, despite escalating allegations, he had believed Holmes’s claims about her technology, saying, “That’s what I assumed. I didn’t probe into it. It didn’t occur to me.”
Which brings us to the most fascinating part of the story: what power did Elizabeth Holmes have that kept people, experts or not, from simply asking, “Does the technology work?”
Much of the answer comes from Holmes herself. In her appearances and interviews, she comes across as a smart and serious young woman. We learn that her commitment arose partly from her own fear of needles, which of course adds a compelling personal note. Many observers were also gratified that her success came in the notoriously male-oriented Silicon Valley world. Her magnetism was part of what Aswath Damodaran of the NYU Stern School of Business calls the “story” of a business. Theranos’s story had the perfect protagonist — an appealing 19-year-old female Stanford dropout passionate about replacing a painful health test with a better, less painful one for the benefit of millions. “With a story this good and a heroine this likable,” asks Damodaran in his book Narrative and Numbers, “would you want to be the Grinch raising mundane questions about whether the product actually works?”
All this adds up to a combination of charisma and sincere belief in her goals. But Carreyrou has a darker and harsher view: that Holmes’s persuasive sincerity was a cover for a master manipulator. Noting her lies about the company finances and technology, her apparent lack of concern for those who might have been harmed by those lies, and her grandiose view of herself as “a modern-day Marie Curie,” he concludes his book with this:
I’ll leave it to the psychologists to decide whether Holmes fits the clinical profile [of a sociopath], but […] her moral compass was badly askew […] By all accounts, she had a vision that she genuinely believed in […] But in her all-consuming quest to be the second coming of Steve Jobs […] she stopped listening to sound advice and began to cut corners. Her ambition was voracious and it brooked no interference. If there was collateral damage on her way to riches and fame, so be it.
I would add one more thought. Holmes did not have the science to judge how hard it would be to realize her dream, then ignored the fact that the dream was failing. Instead she embraced Silicon Valley culture, which rewards at least the appearance of rapid disruptive change. That may not hurt anyone when the change is peripheral to people’s well-being, but it is dangerous when making real products that affect people’s health and lives. Facebook’s original motto, “Move fast and break things,” it seems, is a poor substitute for that old core tenet of medical ethics, “First, do no harm.”
Sidney Perkowitz is a professor emeritus of physics at Emory. He co-edited and contributed to Frankenstein: How a Monster Became an Icon(Pegasus Books, 2018), and is the author of Physics: A Very Short Introduction (Oxford University Press, 2019).
LARB Staff Recommendations
"Frankenstein: Annotated for Scientists, Engineers, and Creators of All Kinds" can remind scientists and engineers to proceed with caution.
Justin Tyler Clark reads the funny, frightening "Live Work Work Work Die."
Did you know LARB is a reader-supported nonprofit?
LARB publishes daily without a paywall as part of our mission to make rigorous, incisive, and engaging writing on every aspect of literature, culture, and the arts freely accessible to the public. Help us continue this work with your tax-deductible donation today!