FICTION · An Unfair Advantage Intervention
Fiction. Real history. Sourced below.
People were harmed in this case. Patients received blood-test results that were unreliable and made medical decisions based on them, and a federal regulator found the laboratory’s practices posed immediate jeopardy to patient health and safety. That is the first fact here. This article does not offer a checklist drawn from it, and there is no framework applied to it, because the material does not belong to that use.
Safeway ran its own pilot clinic and got questionable, delayed results. It proceeded anyway. Walgreens was never permitted to examine the device at all. It proceeded anyway. The two partners failed in opposite directions, and the standard lesson drawn from this case does not explain either one.
Deborah Johnson and Claire Anderson took this one and did not agree, and I am not going to resolve it for them. Deborah’s working life has been spent issuing supplier warnings that arrived on time and changed nothing, and she reads this as a case about signals that were available and were absorbed. Claire thinks that reading is comfortable and probably wrong, because no review process survives a counterparty that manufactures its own evidence. What they agree on is that the usual telling, in which two careless companies missed an obvious fraud, is not what the record shows.
Full fiction disclosure and sourcing note
Deborah Johnson and Claire Anderson are fictional characters from The Unfair Advantage by Todd Hagopian, analyzing the documented history of the Theranos partnerships from the present day. Neither has met, advised, or corresponded with anyone named here, and no dialogue, thought, or motive is invented for any real person. Quoted words come from filed documents. Facts are drawn from the public record, including regulatory correspondence and court filings, and are accurate as of 2026; the analysis is invented. Legal outcomes are stated precisely because they differ between the two defendants.
- The Promise
- The Report
- The Pilot
- The Disagreement
- The Patients
- The Record
- About the Stagnation Assassin
The Promise
Theranos claimed it could run a wide range of diagnostic tests from a fingerprick rather than vials drawn from the arm. It raised on the order of $700 million against a paper valuation near $9 billion, assembled a board heavy with former cabinet officers, and signed two retail partners: Walgreens, which committed roughly $140 million and opened wellness centres in more than forty stores, and Safeway, which spent around $350 million retrofitting hundreds of locations with clinics.
The device did not work reliably. Theranos ran most tests on conventional commercial analysers, the same equipment its pitch described as obsolete, while presenting results as though they came from its own technology. Every other element of the company, the valuation, the board, the partnerships, rested on a capability that was not there.
“And I want to be careful about the sneering version of this,” Deborah said. “This fooled sophisticated investors, experienced directors, and the business press for years. The interesting question is not how anyone believed it. It is what specifically the partners examined, because they did examine things.”
The Report
The single most revealing object in this case is a document. Theranos produced a validation report carrying the logos of Pfizer, Schering-Plough, and GlaxoSmithKline, presented as though those pharmaceutical companies had reviewed and endorsed the technology. They had not. The report was written inside Theranos and the logos were added to it. A Pfizer scientist had in fact assessed the claims and found them unsupported, advising against collaboration, and the fabricated endorsement circulated regardless.
On the Schering-Plough version, the company’s own name was misspelled, and nobody who received it noticed.
“That detail is the whole of it for me,” Deborah said. “Not because catching a typo would have saved anyone, but because of what it tells you about the mode of reading. Somebody looked at that page and processed it as reassurance rather than as evidence. When your verification consists of examining paperwork the subject supplied, you are not testing the subject. You are testing whether they can produce paperwork, and any competent fraud passes that test on the first attempt.”
The Pilot
The two partners did not fail the same way, which is why treating them as one cautionary tale loses the information.
Safeway went further than it is usually credited for. It ran a pilot clinic at its own corporate offices, staffed and observed on its own premises, and that pilot produced results that were questionable and slow. It had, in other words, generated its own independent signal, without needing to trust anything Theranos handed it. It continued with the partnership.
Walgreens never obtained meaningful access to the device. It was not permitted to examine the machine in the way a purchaser of a technology normally would, and it committed regardless.
“So one partner got a warning and did not act on it, and the other never got the information and did not treat that as disqualifying,” Deborah said. “Both are failures of the same kind in my experience, which is that a warning is not a thing that arrives, it is a thing that has to survive contact with everyone who wants the deal to happen. The documented pressure in both cases was competitive: the fear that a rival would take the partnership if they slowed down. That fear is what converted a red flag into a reason to hurry.”
The Disagreement
Claire does not accept the framing, and her objection is the reason this article does not end with a rule.
“Deborah’s account requires that a better process would have produced a different outcome, and I do not think the record supports it,” she said. “Consider what the partners were up against. A counterparty that fabricates validation reports from three pharmaceutical companies will also stage a device demonstration. It will also produce clean data on request. Walgreens asking harder questions gets better-quality answers, not truer ones, because the constraint was never the quality of the questions.”
“The one thing that would have worked was physical access on the partner’s own terms, with its own scientists, on samples it controlled. And Theranos would not grant that. So the decision Walgreens actually faced was not whether to verify. It was whether to proceed without verification, knowing it had been refused. That is a much simpler failure than a diligence failure, and it is also a much more common one, because refusing to proceed means losing the deal, and losing the deal is a visible cost while an unverified claim is an invisible one.”
“Then say what Safeway’s excuse is,” Deborah said. “It had the access. It ran the pilot. It got the signal and continued.”
“I do not have one,” Claire said. “That is the case that troubles my argument, and I hold it anyway.” They left it there, and so do I. Whether a disciplined process catches this, or whether the only real decision point was much earlier and much cruder, is not settled by the evidence available.
The Patients
On January 25, 2016, following a survey of the company’s Newark laboratory, the Centers for Medicare and Medicaid Services notified Theranos of deficiencies including one that, under the hematology condition of the federal laboratory rules, constituted immediate jeopardy to patient health and safety. The company’s response the following month was found not to demonstrate compliance.
Theranos subsequently voided or corrected two years of results from its own devices, issuing tens of thousands of amended reports to physicians and patients. The company characterised this as approximately one percent of the tests it had performed. Separately, a study published in March 2016 found the company’s methods more likely to return abnormal results than standard venous draws at established laboratories.
What those figures describe is people who received a frightening result that was wrong, or a reassuring one that was, and who made decisions with a physician on that basis. The proportion does not change what happened to any individual within it. Walgreens marketed these locations as convenient, trustworthy healthcare rather than as an unproven technology a patient might reasonably decline, and that gap between the marketing and the thing is where the harm sits.
The Record
Safeway walked away in 2015. Walgreens terminated the partnership in June 2016 and sued for the roughly $140 million it had committed, settling for under $30 million. Theranos dissolved in September 2018.
The legal outcomes differ between the two defendants and the difference matters. In January 2022 Elizabeth Holmes was convicted of one count of conspiracy to defraud investors and three counts of wire fraud relating to investors. She was acquitted of the patient-related fraud counts, and the jury did not reach a verdict on three further investor counts. She was sentenced to 135 months and surrendered in April 2023. Ramesh Balwani was convicted in July 2022 on all twelve counts, covering both investor and patient fraud, and was sentenced to 155 months, a longer term than Holmes received.
That split is worth stating plainly in an article about patients, because it is frequently collapsed. A jury found the investor fraud proven against both and the patient fraud proven against one. Whatever conclusions anyone draws from this case, they should start from what was actually decided rather than from the summary version.
About the Stagnation Assassin
Todd Hagopian is VP of Product Strategy and Marketing at JBT Marel and the author of the Stagnation Assassin series.