Theranos red flags and timeline, read as decision signals

Which Theranos red flags could a decision model read from public data?

The EQOS Asia study lists 6 signals that anyone could find in about 10 minutes, using sources open in 2013: PubMed, the FDA, the SEC, LinkedIn, Glassdoor, and the company website (EQOS Asia Theranos case page, as of October 2026). The page states that every input was public. That rule gives the exercise its value, because a reader can repeat each check today on a new supplier or partner.

A decision model sorts these red flags into 4 groups: scientific evidence, board expertise, secrecy about the technology, and patient safety. Each group maps to named dimensions of the 9-dimension method, so a red flag becomes a number that can be tracked over time. The next sections place those numbers on a calendar, which turns a list of worries into a sequence of decisions.

What did the PubMed check show among the Theranos red flags?

The first red flag is a literature check. The study reports 0 PubMed results for the company name, 0 results for the founder’s name, and no FDA approval, after 10 years of research and $400M raised (EQOS Asia Theranos case page, as of October 2026). Three small searches, each with a yes or no answer, produce the strongest signal on the page.

The reading is simple. A claim of scientific novelty leaves peer-reviewed traces, so an empty result lowers the transparency score and opens a request for independent verification. The check works in any sector: search the literature, the patent register, or the regulator’s database for the claim at hand. The figure sets out the three checks and their results.

How does the Theranos timeline run from 2013 to 2022?

The study sets 4 dates on its timeline. In 2013 the back-test returns φ(S) = 0.085 and issues 4 conditions. In 2014 a pharmacy chain deploys the tests in 40 pharmacies. In October 2015 a newspaper investigation exposes the fraud. In January 2022 a verdict finds the founder guilty on 4 fraud counts (EQOS Asia Theranos case page, as of October 2026).

Read as decision signals, the dates fall into three phases: a quiet phase where only public absences exist, a deployment phase where harm to patients becomes possible, and an exposure phase where outside parties confirm the pattern. The model’s contribution sits in the first phase, about 2 years before the exposure, when a low-cost check could still change the outcome.

Why does the 2014 pharmacy rollout matter among the Theranos red flags?

The rollout in 40 pharmacies is the point where a business risk becomes a patient risk. The study records that the deployment took place before the device approval that its fourth condition requires (EQOS Asia Theranos case page, as of October 2026). Deployment before validation is a red flag in any regulated field, because users carry the cost of an unproven claim.

The study documents three patient cases: an HIV-positive patient reported as negative, a healthy patient flagged with suspected cancer, and a coagulation disorder reported as normal. Its summary panel lists 890 000 patients with false results. These cases explain why the ethical safeguard blocks both investment and deployment, and why the block overrides every growth figure.

How do the Theranos red flags score across the 9 decision dimensions?

The back-test places 8 of 9 dimensions in the critical zone, below 0.3. In Domain A, coherence scores 0.05 and integrity 0.03, with resonance at 0.38. In Domain B, care scores 0.04, trust 0.06, and communication 0.02 (EQOS Asia Theranos case page, as of October 2026). Each red flag lands on a named dimension, so the reader sees where the weakness sits.

Domain C shows clarity at 0.05, stability at 0.08, and energy at 0.25. The score is a weighted geometric product of the 9 dimensions, so a single collapsed dimension lowers the whole result, and the study reports φ(S) = 0.085 (EQOS Asia method page, as of October 2026). The weighting itself stays proprietary to the engine.

Which checks would a decision model request at each Theranos date?

In 2013 the model requests 4 conditions: a peer-reviewed article in a leading journal, a technical review by an independent certified pathologist, at least 2 independent medical experts on the board, and device approval from the regulator before patient use. The study ties each condition to a dimension, such as transparency rising from 0.02 to 0.40 (EQOS Asia Theranos case page, as of October 2026).

In 2014 the relevant request is narrower: hold the pharmacy rollout until the approval exists. In 2015 the question changes to how much exposure is already taken. The same list of conditions guides each date, and a refusal to meet a condition counts as a signal of its own, recorded in the file for the next review.

How do you build a red-flag timeline for your own decision?

List the claims that carry the decision, then date each public check that can test them. Put the cheap checks first: a literature search, a registry lookup, a review of the board. Record the result, the date, and the dimension affected. A one-page timeline of this kind lets a team see when the evidence first became available and who looked.

Then score the situation. The free simulator gives a first φ(S) estimate from ratings on 9 dimensions (EQOS Asia simulator page, as of October 2026), and the method page describes the Audit pillar that lists the 3 priority risks. The Theranos study is a retrospective exercise on public data. This page is general information, not legal, tax or investment advice.

Frequently asked questions

About the author

Stéphane Jambu

President, EQOS Forecast Asia

Stéphane Jambu leads EQOS Forecast Asia and works as a consultant in e-commerce, search visibility, and AI. LinkedIn profile