Theranos blood tests were not shown to be more accurate than conventional laboratory testing. The strongest independent comparison, involving 60 healthy adults and 22 measurements, found clinically important disagreement between Theranos and established laboratory services, more out-of-range results, and dramatically more rejected samples. Later federal administrative and appellate findings describe reliability and quality-control failures, a limited assay menu, slower or inadequate throughput, and the use of conventional third-party analyzers for patient testing.
A finger prick and a small specimen can be useful collection methods, but they do not prove that an analyzer is analytically valid. Accuracy depends on the entire chain: collection, specimen preparation, calibration, quality control, validated assays, proficiency testing, reporting, and clinical interpretation.
What Theranos promised
Theranos promoted a system built around a proprietary small-volume collection device, called a “nanotainer,” and analyzers identified in regulatory records as the TSPU, Edison, and miniLab. Public messaging said patients could obtain many tests from a few drops of blood, avoid larger needles and multiple vials, receive results faster, and obtain results that were more accurate and reliable than conventional venous blood testing.
Walgreens Wellness Centers marketed clinician-directed testing using finger-stick collection. The appeal was straightforward: less invasive collection, a smaller specimen, potentially lower logistical burden, and a broad menu of tests from one visit. Those are claims about convenience and system design, not proof that each assay produces clinically dependable numbers.
Recommended Free Tools
#1 Best Overall
The company’s public description also implied that proprietary automation would reduce human error. In practice, the evidence later presented to regulators and courts described a system with accuracy and reliability problems, limited testing capability, recurring quality-control failures, and use of commercially available third-party analyzers for patient testing.
How the methods differed
| Comparison point | Theranos model as described in records | Conventional laboratory model |
|---|---|---|
| Collection | Finger prick using a nanotainer or other low-volume collection process; typical volume was no more than 150 microliters. | Venipuncture into standardized tubes; the 2016 comparison describes typical volumes of at least 1.5 milliliters. |
| Specimen preparation | Small samples required specialized handling and, according to reporting summarized by The Pew Charitable Trusts, many specimens were diluted so they could be run on modified conventional analyzers. | Larger specimens are collected in validated tubes and processed under established laboratory procedures for the relevant assay. |
| Analyzers | Theranos presented the Edison and miniLab as proprietary platforms, while later findings documented use of third-party analyzers for patient testing and problems with the proprietary system. | Laboratories use commercially validated analyzers, calibration procedures, internal controls, and assay-specific operating requirements. |
| Assay menu and throughput | Marketing suggested a broad menu from a few drops. Administrative findings concluded that the proprietary analyzer performed only a limited number of tests and could not match conventional machines for high-throughput work. | Large laboratories run extensive menus on platforms designed for standardized, high-volume processing. |
| Oversight | Testing occurred in a CLIA-certified laboratory under a laboratory-developed-test regulatory position before later FDA and CMS actions. | Laboratories operate under CLIA requirements, including laboratory-specific analytical-performance standards and proficiency testing; devices may also be subject to FDA pathways depending on their status and use. |
The key distinction is that specimen volume is a pre-analytic variable. Reducing volume can increase sensitivity to contamination, clotting, evaporation, cell damage, dilution, and collection variability. A low-volume method can be valid when each step is validated for that specimen type and volume, but the collection format alone cannot establish accuracy.
What the independent comparison found
Study design
The peer-reviewed study Evaluation of direct-to-consumer low-volume lab tests in healthy adults was published in the Journal of Clinical Investigation on 2 May 2016. It enrolled 60 healthy adults in Phoenix, Arizona, and compared 22 common measurements obtained through Theranos finger-prick services with results from venipuncture services at Quest Diagnostics and Labcorp.
Each service was treated as a practical testing system rather than as a laboratory component that investigators could fully disassemble. The comparison therefore captured what a consumer-facing service delivered, including collection and reporting, but it did not constitute a technical audit of every instrument, reagent lot, operator, or Theranos assay.
Measured differences
| Finding | Result | Why it matters |
|---|---|---|
| Out-of-range flags | Theranos produced 1.6 times more out-of-range flags than the other services. | More flagged values can trigger repeat testing, referrals, anxiety, or treatment decisions even when a patient is healthy. |
| Interservice variability | 15 of 22 measurements, or 68%, showed significant variability between services. | Results were not interchangeable across the tested services for most measurements examined. |
| Missing data | Theranos: 2.2%; Labcorp: 0.2%; Quest: 0%. | A missing result can delay a clinical decision and may require another collection. |
| Sample rejection | The odds of a Theranos sample rejection were 12.5 times those of the other services, with a 95% confidence interval of 6.9–22.4. | Collection and pre-analytic handling were substantially less reliable in this comparison. |
| Lipid panel | Results were nonequivalent, with systematic bias in total cholesterol, HDL-C, and LDL-C. | Different numbers could change cardiovascular-risk interpretation or lead to unnecessary follow-up. |
The authors concluded that the disparities could alter clinical interpretation and health-care utilization. The issue was not merely whether a result looked plausible; clinically relevant disagreement can change whether a clinician orders another test, labels a condition, starts treatment, or reassures a patient.
What those figures do—and do not—prove
The study provides evidence of substantial disagreement in the tested panel and population. It does not establish a single error percentage for every Theranos test or every patient who used the service.
- The sample contained 60 healthy adults, not a representative cross-section of all patients or disease states.
- Only 22 measurements were compared.
- Timing, subject biology, collection method, and service variability could all influence results.
- The investigators reported no technical replicates for the Theranos data, limiting conclusions about repeatability within the same sample.
- The services were compared as black boxes, so the study could identify clinically important disagreement without assigning every discrepancy to a specific instrument component.
No comprehensive, independently audited all-assay error rate covering every Theranos patient result has been established. It is therefore inaccurate to convert this study into a universal claim that a particular percentage of all Theranos results was wrong. The defensible conclusion is narrower and more useful: in this controlled comparison, many measurements disagreed significantly, lipid results were nonequivalent, and Theranos samples were rejected far more often.
Why a finger prick cannot substitute automatically for venipuncture
Collection volume changes the technical problem
Venipuncture generally provides a larger, more uniform specimen. A finger prick produces a small capillary sample that can contain a mixture of blood and tissue fluid, especially if the finger is squeezed aggressively. The smaller volume leaves less margin for dead space, transfer loss, clot formation, bubbles, evaporation, or an insufficient fill. Every one of those factors can affect whether an analyzer accepts a specimen and whether the measured concentration represents the patient’s circulating blood.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Scan for outdated or missing drivers - takes under a minuteDriver Scan →Different specimens require validation
Assays are validated for a defined specimen type, collection device, volume, anticoagulant, transport condition, and analytic platform. A method validated on venous plasma cannot simply be assumed to work on a diluted capillary specimen. The laboratory must demonstrate precision, analytical accuracy, reportable range, interference performance, reference intervals, and acceptable error under the exact conditions used for patient testing.
Pre-analytic failures can look like analytic failures
A rejected or distorted result may originate before the instrument measures anything. Inadequate filling, hemolysis, clotting, contamination, delayed processing, or an incorrect dilution can produce missing data or implausible values. The 2016 comparison’s rejection and missing-data findings show why collection and specimen handling belong in any accuracy discussion.
How conventional laboratory testing is controlled
Conventional testing is not automatically perfect, but it is built around layered controls. A typical workflow includes patient identification, standardized venipuncture, correct tube selection, transport and accessioning, centrifugation or other preparation, instrument calibration, internal quality-control checks, result review, and reporting through a laboratory operating under CLIA requirements.
CLIA requires laboratories to establish analytical performance and participate in proficiency testing, with total allowable error criteria applied to relevant measurements. These controls address precision and accuracy over time, not just whether one result appears reasonable. A laboratory can still produce an error, but a mature quality system is designed to detect, investigate, document, and correct recurring problems.
Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchPC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Large analyzers also provide throughput advantages. They can process many specimens with standardized reagents, automated checks, barcoding, and maintenance schedules. The 2026 HHS Departmental Appeals Board record summarizes findings that the proprietary Theranos analyzer performed a limited number of tests, was slower than some competitors, and could not compete with conventional machines for high-throughput work.
FDA and CMS/CLIA: different responsibilities
FDA and CMS did not regulate the same part of the testing system. FDA oversees medical devices and in-vitro diagnostic products through applicable market pathways. CMS administers CLIA, which governs laboratory operations and certification for human testing.
Historically, some laboratory-developed tests operated under FDA enforcement discretion. That position did not remove the laboratory’s obligation to establish analytical performance before reporting patient results. A CLIA-certified laboratory still had to meet laboratory-specific standards for the tests it performed.
The Pew Charitable Trusts reported that Theranos opened a CLIA-certified laboratory in Newark in 2012 and ran patient testing for about two years without premarket FDA review under its laboratory-developed-test position. FDA later cited problems involving the nanotainer’s classification and record-keeping. Theranos received FDA clearance for an HSV-1 test in July 2015, while CMS issued a warning in January 2016 and revoked the laboratory certificate later that year. Theranos subsequently invalidated patient results.
The Tool Desk
Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Rank #3
- Brand New in box. The product ships with all relevant accessories
These events illustrate why “FDA-cleared” and “CLIA-certified” are not interchangeable descriptions. One concerns a device or diagnostic product’s regulatory pathway; the other concerns the operation of a human-testing laboratory. The applicable requirements depend on what is being tested, how the test is developed and marketed, and how the laboratory performs and verifies it.
What later administrative and appellate findings established
The 2026 HHS Departmental Appeals Board record summarizes allegations and findings that Theranos represented its analyzer as capable of performing the full range of finger-stick tests with faster, more accurate, and more reliable results. The findings instead describe accuracy and reliability problems, a limited test capability, inadequate throughput compared with conventional systems, and use of commercially available third-party analyzers for patient testing.
The record quotes a Ninth Circuit summary stating: “Theranos’s blood-testing device failed to deliver faster and more accurate testing results than conventional technology.” It also quotes the finding that the device “consistently failed quality control checks and was unable to provide accurate results.” Those statements concern the device and quality-control record described in the proceedings; they should not be expanded into an invented numerical error rate for every patient or assay.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Theranos versus conventional laboratories by decision factor
| Decision factor | Theranos evidence | Conventional laboratory implication |
|---|---|---|
| Specimen volume | Typically no more than 150 microliters in the 2016 description. | Venipuncture typically supplied at least 1.5 milliliters, providing more material for repeat or reflex testing. |
| Analytical validity | Significant disagreement across 15 of 22 measurements and nonequivalent lipid results in the independent comparison. | Assays are expected to meet documented performance requirements for the specimen and platform used. |
| Precision and repeatability | The study lacked technical replicates for Theranos data, and later findings cite quality-control failures. | Internal controls, calibration, maintenance, and proficiency testing are designed to monitor ongoing performance. |
| Sample rejection | 12.5-fold higher rejection odds than the comparison services. | Standard tubes and established collection procedures generally reduce, but do not eliminate, pre-analytic rejection. |
| Assay breadth | Marketing implied broad capability; findings describe a limited proprietary menu and third-party analyzer use. | High-volume laboratories commonly combine multiple specialized platforms to cover broad menus. |
| Throughput | Later findings say the proprietary system could not match conventional machines for high-throughput work. | Automation and batch processing support large numbers of routine specimens. |
| Transparency | Proprietary claims were difficult for outsiders to evaluate, and the independent study treated the service as a black box. | Established laboratories publish test methods, reference information, specimen requirements, and quality policies, although transparency varies by laboratory. |
| Clinical risk | More out-of-range flags and biased lipid results could lead to unnecessary follow-up or incorrect interpretation. | Any abnormal or unexpected result still requires clinical context and, when appropriate, confirmation. |
Why Theranos results were invalidated
Result invalidation followed the combination of regulatory action and evidence that the testing system could not reliably support reported patient values. CMS revoked the laboratory’s certificate after its warning, and Theranos later invalidated patient results. Invalidation is not a claim that every individual number was wrong; it means the laboratory could no longer assure that the affected results met the required standards for clinical reporting.
For patients, the practical consequence was that a previously reported value could not safely be treated as a dependable basis for diagnosis or treatment. A clinician needing that information would have to order a new test through an appropriately licensed laboratory and interpret it alongside the patient’s symptoms, history, medications, and other results.
Can a finger-prick test replace a conventional blood draw?
Sometimes a validated capillary test can answer a specific clinical question. The deciding factors are the exact assay, specimen type, collection device, analyzer, quality system, and intended use—not the presence of a finger prick by itself.
A capillary method should be considered a substitute only when the laboratory or test manufacturer has demonstrated performance for that method and the ordering clinician accepts it for the decision at hand. A broad promise to run many unrelated tests from a few drops is not equivalent to assay-by-assay validation.
- Ask which laboratory performs the test and whether it is currently licensed and CLIA-certified for the testing offered.
- Confirm the specimen type and collection requirements for the specific assay.
- Check whether the method has published analytical-performance data and appropriate reference intervals.
- Ask how rejected, insufficient, hemolyzed, or otherwise compromised specimens are handled.
- For a result that would change diagnosis or treatment, discuss whether venous confirmation is appropriate.
What to do if you have an old Theranos result
- Do not assume the result is reliable merely because it was printed, displayed in a portal, or used in a previous appointment.
- Tell your current clinician the date, test name, value, reference interval, and the fact that it came from Theranos.
- Request repeat testing through an appropriately licensed laboratory, using the specimen type and method your clinician considers suitable.
- Bring related records, including symptoms, medications, diagnoses, and any treatment that began because of the original value.
- Do not stop prescribed treatment or make a new treatment decision solely from an old Theranos report; use clinical advice and confirmed testing.
How to evaluate any low-volume testing service
The Theranos episode offers a general checklist for judging future claims about “a few drops” of blood.
- Separate collection from measurement. A less invasive sample does not establish analytical accuracy.
- Look for assay-specific validation. Broad claims about a platform are weaker than published data for each test and specimen type.
- Check quality controls. Ask whether the laboratory performs calibration, internal controls, proficiency testing, and documented corrective actions.
- Inspect the rejection policy. A service should explain what happens when a sample is insufficient, hemolyzed, clotted, contaminated, or outside validated conditions.
- Demand clear regulatory descriptions. “FDA-cleared,” “FDA-approved,” “laboratory-developed test,” and “CLIA-certified laboratory” describe different things.
- Consider clinical consequences. A false positive can prompt unnecessary testing or treatment; a false negative can delay diagnosis; a biased value can distort risk calculations.
- Prefer independent evidence. Peer-reviewed comparisons and transparent performance data are more informative than marketing claims alone.
The practical verdict
Theranos did not demonstrate that finger-stick testing from a few drops was faster, more accurate, or more reliable than conventional laboratory testing. The independent comparison found substantial disagreement, higher rejection and missing-data rates, and nonequivalent lipid results, while later official findings documented quality-control, reliability, scope, throughput, and equipment-disclosure problems. The evidence supports using validated, appropriately regulated testing and confirming any result that could affect diagnosis or treatment.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
Say something about this film