Promus Diagnostics Review – A Real Lab With a Storefront That Does Not Exist

This review went in one direction and came out in another. We expected to check a marketing claim about accreditation and find it soft. Instead the accreditation claim is completely true, verifiable in federal records to the certificate number, and it is the shop that turns out not to exist. What follows is both halves of that: how to verify a laboratory yourself, and why the most important number in this whole category is 60%.

What Promus Diagnostics Is, and What It Is Not

Promus Diagnostics, LLC is a molecular diagnostics laboratory at 525 S Kimball Avenue, Southlake, Texas 76092. It runs PCR-based panels that look for bacterial, viral, fungal and parasitic targets along with antibiotic resistance genes, across six areas: gastrointestinal, genitourinary and UTI, integumentary (hair, nail, skin), respiratory, sexual health and STI, and women’s health. The branded workflow is called Promus Pathway.

Three things it is not, and each one matters to a reader of this site.

It is not a consumer product any more. It was one: a division called Promus Direct sold at-home kits from 2023 until some point after September 2025. That arm is gone, and today the website is built for clinicians — a provider portal, a “Providers” section, a test menu written for people who order tests. The only two patient-facing links are a portal login and a “Pay My Bill” page. Nothing on the current site offers a patient a way to buy a test, and the current site never mentions the at-home business at all.

It is not yet the broad testing company the homepage implies. Pharmacogenomics, blood and wellness testing, and pathology each appear on the front page under the heading “COMING SOON”. That is honest labelling, and we credit it — but if you arrived expecting PGx or blood work, it is not available.

It is not a weight-loss service, and we are reviewing it on a weight-loss site because it was offered to us as an affiliate programme. The one place it genuinely intersects with our readers is gut symptoms on GLP-1 medication, which we cover below — and there the useful advice runs slightly against ordering the test.

The Storefront Does Not Exist

The affiliate programme sends traffic to https://promusdirect.shop/. We followed the link, got a 302 redirect to that address, and then got nothing at all — not a 404, not a parked page, not a holding notice. The connection could not be made because the name does not resolve.

Because “the site did not load” is the kind of thing that is usually a fault at the tester’s end, we checked it five ways and controlled every one:

The authoritative answer came back from the .shop registry itself — the SOA record for shop. at a.gmoregistry.net. When the registry that runs the entire top-level domain tells you there is no such name, there is no such name. The domain registration itself still shows as active, which usually means it is registered but has no nameservers pointing anywhere.

It did exist, though, and the Internet Archive shows exactly what it was. Under the name Promus Direct, it ran as a Shopify store, and the working capture from 27 March 2024 describes itself as “a specialized medical laboratory offering a variety of at-home test kits to help you better understand your health”. It sold two products, and only two:

The store had a “Register Test Kit” step in its navigation, so the model was mail-order at-home collection: buy the kit, register it, send the sample back. It took the full range of ordinary payment methods — a real checkout, not a lead-capture form.

The .shop address turns out to have been a secondary front for a larger site at promusdirect.com, which its own pages linked to for contact and kit registration. Following that thread gives the real timeline, and it is a longer and sadder one:

So this was not a business that never launched. It ran for roughly two years, sold a real product through a real checkout, and wound down — with both items showing “Currently Out of Stock” in the last capture before the lights went off. The laboratory behind it is still very much operating. What closed was the direct-to-consumer arm.

Which leaves the awkward part. The affiliate listing for this offer records a last update of 19 March 2025 and a status of active — for a .shop storefront that had been parked since December 2024, and a domain that has since stopped existing entirely. Somewhere between the laboratory, the network and the publisher, nobody checked. We are the ones who were supposed to check, so we did, and this is the result. No link, no commission, and if all you wanted was whether you can buy it: you cannot.

We Checked the Credentials. They Are Real.

The advertiser describes Promus as “a high-complexity CAP-Accredited & CLIA-Certified laboratory”. In this business those words get used loosely often enough that we treat them as claims to verify rather than facts. Here they hold up completely.

CMS publishes the entire national register of certified laboratories as the Provider of Services File – Clinical Laboratories, free to download and updated quarterly. We pulled the Q2 2026 file — 681,059 laboratories — and found exactly one match:

We did not simply trust our own reading of the certificate-type code, either. We validated it against the same file: of 29,971 type-3 laboratories, 8,012 carry a CAP flag — against 1 out of 520,719 type-2 laboratories and 0 out of 76,232 type-4. And Mayo Clinic Laboratories appears as type 3 with a CAP flag and 13.7 million tests. The coding means what we think it means.

A second, independent federal source agrees. In the national NPI registry, Promus Diagnostics, LLC is NPI 1033680681, enumerated 7 December 2018, status active, primary taxonomy 291U00000X — Clinical Medical Laboratory, at the same address and the same phone number published on the company’s website.

One small precision on the word “high-complexity”. A Certificate of Accreditation is issued to laboratories performing non-waived testing — moderate and/or high complexity — as opposed to a Certificate of Waiver, which covers only simple, low-risk tests. So the certificate is consistent with high-complexity work, and molecular pathogen panels are high-complexity testing. What we verified precisely is the certificate type and the accreditor; we did not separately obtain a complexity classification, so we are reporting the first two rather than restating the third.

A nice corroboration turned up in the archive. The defunct consumer site published both numbers on its “Our Company” page — “#8412352 CAP-accredited… #45D2156626 CLIA-certified” — and that CLIA number is character-for-character the one we independently pulled from the CMS file. Two sources that never spoke to each other agree. Note, though, that the current laboratory site does not publish either number on its home, about or contact pages; we had to reach into an archived consumer site to find them, which is a small missed opportunity for a company whose credentials are its strongest card.

On people: the about page names Pejman Rahimian, PhD as Chief Scientific Officer and CLIA Laboratory Director, with Victor J. Sansone III as Executive Chairman and CEO — and Sansone also appears as the authorised official on the NPI record. A doctoral-level scientist rather than a physician is entirely permissible as a CLIA director for high-complexity testing, and we note the credential only so you know what it is: a PhD, not an MD.

What the Independent Record Says

Very little, and in this instance that is mildly good news rather than a gap we are papering over. We went looking for complaints and litigation and found essentially none.

• Better Business Bureau: no profile at all. A search for “Promus Diagnostics” returns no results. No rating, no accreditation, no complaints — which for a business-to-business laboratory with no consumer arm is unsurprising rather than suspicious.
• Litigation: none against the company. A phrase search of CourtListener’s federal archive found no lawsuits naming Promus Diagnostics as a party. Two incidental string matches exist and neither is an action against the company: its name appears inside an exhibit list filed in an unrelated securities case against a different PCR company, and it appears as a listed creditor in an individual’s consumer bankruptcy schedule — that is, someone owed the lab money, not the reverse. We validated the search with a deliberately nonsense control phrase, which returned zero, so the absence is meaningful rather than a broken query.
• Trustpilot: we could not check. Both relevant Trustpilot URLs returned HTTP 403 to every method we tried. We are reporting this as unchecked, not as clean.

We also did not check HHS-OIG exclusions, Department of Justice actions or Texas state filings, so we make no claim in either direction about those.

How Big Is This Lab?

Smaller than the marketing language suggests, though not unusually so. The same CMS record carries a reported accredited test volume of 14,140. To put that in context, we took every type-3 laboratory in the file reporting a non-zero volume — 18,345 of them — and ranked it:

So Promus sits just below the middle of accredited American laboratories — roughly 55% of the median volume, and around 272 tests a week. That is a real working laboratory, not a shell, and not a giant. It is worth holding beside the phrase “trusted by providers nationwide”, which is true in the sense that providers anywhere can send samples, and which describes a lab running about fifty tests on a business day.

Two caveats on that number, because we would rather under-claim it. The volume figure is self-reported on the CMS certification form and updated at certification or recertification, so it may lag actual throughput. And a low-volume specialist laboratory is not inherently worse than a high-volume one — volume is a size measure, not a quality measure.

How the Testing Actually Works

The brief we were given asked us to describe the path “from sign-up to delivery or first use”. For this product there is no sign-up, so here is the path as it actually runs:

1. A clinician decides a test is warranted and orders a specific panel. You do not initiate this; that is the design, and for a high-complexity infectious disease panel it is the right design.
2. A specimen is collected — urine, stool, a swab — in the clinic or, for some panels in this category, at home under instruction.
3. The specimen goes to the laboratory in Southlake and is run on a multiplex PCR platform that amplifies DNA from many organism targets at once, alongside a separate set of antibiotic resistance gene targets.
4. A report goes back to the ordering clinician, not to you, listing which targets were detected and which resistance genes were found.
5. Your clinician interprets it against your actual symptoms. This step is not a formality. It is where nearly all of the value and nearly all of the risk of this category sits, for reasons the next two sections explain.

We could not verify turnaround times, specimen stability requirements, whether collection kits are shipped to patients, or how results are delivered, so we are not stating any of those.

Pricing: What We Could and Could Not Find

Nothing is on sale today at any price, and the laboratory’s own site publishes no rates at all. But the consumer business did have prices while it ran, and they are worth putting on the record — both because they tell you what this cost and because they are the only figures anyone can honestly quote.

An archived product page lists what came in the box: an instruction pamphlet, a urine collection cup, a vacutainer tube, an antiseptic towelette, a biohazard bag and a pre-paid return shipping bag. Shipping was advertised as free, and the site ran a 10% discount code — fittingly, STEWARDSHIP — in its final months, which would have brought a kit to about $251.

$279 is a lot for a urine test, and the payment terms were the part to read closely. The site advertised “HSA/FSA eligible” prominently. Its own FAQ was considerably more candid about ordinary insurance: “we cannot guarantee purchase coverage of our at-home test kits using health insurance, as our test kits are not specifically covered under health insurance plans”. In other words this was a cash purchase, spendable from a tax-advantaged account but not something to expect a plan to reimburse.

For anything ordered through a clinician today, no price is published, and what you pay would depend on network status, deductible and how the panel is coded. There is no monthly cost, because there was never a subscription.

That is not unusual for clinician-ordered laboratory work, and it is worth explaining why, because the explanation is the actionable part. What you pay for a lab test typically depends on whether the laboratory is in network for your insurance, where you are against your deductible, and how the specific panel is coded and billed— and large multiplex panels can be billed very differently from a simple culture. The same test can cost one patient nothing and another a great deal.

The 60% Problem: Resistance Genes Versus Reality

This is the part of the review that matters most, and it applies to the whole category rather than to this company specifically.

The pitch for resistance gene testing is intuitive: find the genes that make bacteria resistant, and you can pick an antibiotic that will work. Promus frames its own offering around exactly this, describing a commitment to “antimicrobial stewardship” and to “prescribing antimicrobial treatment with utmost confidence”.

The best evidence on whether that works is a study published in Infection and Drug Resistance in 2021. Researchers collected urine from 1,155 patients at 37 urology clinics across the United States, used multiplex PCR to detect 27 resistance genes, and compared the gene results against actual laboratory susceptibility testing on the same samples across 14 antibiotics.

Read that conclusion carefully, because it is not a dismissal — it is a condition. Resistance gene data is useful in conjunction with susceptibility testing. On its own, at 60% agreement, it is not a reliable basis for choosing a drug.

The study also found why it goes wrong, and the mechanism is easy to understand. Concordance was significantly worse for vancomycin, meropenem and piperacillin/tazobactam in polymicrobial infections than in monomicrobial ones. When a sample contains several organisms, a resistance gene shows up as present — but the test cannot tell you which organism is carrying it. A gene sitting in a harmless bystander organism reads identically to the same gene sitting in the bug that is actually making you ill.

There is a second, related issue: PCR finds a great deal more than culture does. A 2025 study in Therapeutic Advances in Urology compared an expanded panel against standard culture in 56 specimens from patients over 50 who had already failed treatment. Culture identified an organism in 50% of samples; PCR found at least one uropathogen in 92.9%, and found additional co-infecting organisms in 71.2% of its positives.

Whether that is a triumph or a trap depends on a question that study does not answer: does treating the extra findings make patients better? PCR amplifies DNA. DNA is present in organisms that are dead, and in organisms that live in you quite happily without causing disease. Asymptomatic bacteriuria — bacteria in the urine of someone with no symptoms — is common, particularly in older adults, and the long-standing clinical consensus is that in most patients it should not be treated with antibiotics. A more sensitive test finds more of it.

And here Promus deserves real credit, because it has visibly thought about exactly this problem. Its Promus Pathway pages state that results are reported in CFU/mL rather than DNA copies/mL — that is, as an estimate of how much viable organism is present rather than merely whether its DNA was detected. That distinction is precisely the one the colonisation objection turns on. A presence/absence result cannot tell a trivial amount of a bystander organism from a heavy growth of the thing making you ill; a quantitative one gives a clinician something to weigh. It is a meaningful design choice and it is the strongest answer to the criticism above that a lab in this category can give.

For scale, the company describes panels of “up to 94 targets on a single panel” and a menu spanning “over 100 pathogen targets and 30+ resistance gene targets”, with the urinary panel listing 34 pathogen targets across 12 antimicrobial classes and a claimed “95% sensitivity in CFUs/mL within just 24 hours”. We have not verified that sensitivity figure, and we would want to see the population it was measured in before treating it as a performance claim. One presentational note: the site describes resistance by antimicrobial class rather than naming individual resistance genes, which is arguably more clinically digestible — but it means you cannot tell from the website which specific genes are on the panel.

So the honest framing of this category is: a more sensitive test is not automatically a more useful one, and a test marketed as a tool against antibiotic resistance can, used without judgement, produce more antibiotic prescribing rather than less. That is not an accusation against Promus, whose stated commitment to stewardship we take at face value — the company’s name is reportedly Latin for exactly that, and the quantitative reporting backs the sentiment with a design decision. It is simply the question to ask them, or any lab like them: does this panel pair gene detection with phenotypic susceptibility testing? The 1,155-patient study says that pairing is what turns the data from unreliable into valuable.

If You Are on a GLP-1 and Have Gut Symptoms

This is the one place this laboratory genuinely intersects with what most readers come to this site for, and the advice runs slightly against ordering the test.

Gastrointestinal effects are the dominant reported adverse events for GLP-1 medications. In the FDA’s adverse event reporting system, across 73,001 reports naming semaglutide, the most frequently reported reactions include nausea (10,674), vomiting (6,964), diarrhoea (6,462) and constipation (4,724). Across 160,663 reports naming tirzepatide: nausea (17,092), diarrhoea (10,006), vomiting (8,267), constipation (6,502).

Those are spontaneous reports with no denominator — they cannot tell you how likely any effect is, and we are not presenting them as incidence figures. What they show clearly is the shape of the problem: when these drugs cause trouble, it is usually the gut, and that matches both the trial literature and the product labelling.

Now put that next to a sensitive multiplex GI panel. You are on a drug whose commonest adverse effect is diarrhoea. You develop diarrhoea. A panel sensitive enough to detect organisms that culture misses is very likely to find something, because most people are carrying something. If that finding is read as the cause, you may end up taking an antibiotic for an organism that was not making you ill, while the actual cause — the dose in your pen — goes unaddressed.

Eligibility, Availability and Who Orders It

Eligibility is decided by a clinician, not by a questionnaire. There is no consumer sign-up to be eligible for. The affiliate programme is listed as United States only, and we make no claim about availability anywhere else.

Because this is a diagnostic test rather than a treatment, the usual review headings of contraindications and side effects do not map cleanly. Collecting urine, stool or a swab carries essentially no physical risk. The risks in diagnostics are informational, and they are real ones: a false positive that leads to an unnecessary antibiotic; a true positive for an organism that was not causing your symptoms, with the same result; a resistance gene finding that points at the wrong drug because it belonged to a different organism in the sample; a negative that falsely reassures; and a bill you did not expect.

One regulatory point worth being precise about, since the brief rightly insists on it. Panels of this kind run by an individual laboratory are typically laboratory-developed tests, validated in-house under CLIA rather than cleared or approved by the FDA as commercial kits. CLIA certification and CAP accreditation are not FDA approval. They govern how the laboratory operates — personnel, procedures, quality control, proficiency testing — and they are genuinely meaningful. They say nothing about FDA review of any particular assay. If that distinction matters to you, ask the lab directly which of its panels are FDA-cleared and which are laboratory-developed.

How It Compares

We are not able to rank Promus against named competitors on price, turnaround or panel breadth, because we could not obtain any of those figures for Promus. Anyone claiming to rank it for you on those grounds should be asked where they got the numbers.

Who This Is For, and Who Should Look Elsewhere

Pros and Cons

Limitations of This Review

We did not use this laboratory, and we have seen no report it produced. We cannot tell you about accuracy, turnaround, report clarity or customer service. Nothing here is an assessment of the quality of the science done in Southlake, and the low rating should not be read as one.

What we verified, we verified properly; what we did not, we have said. The CLIA and NPI findings come from federal records we downloaded and parsed ourselves, with the certificate-type coding validated against the same dataset. The DNS finding was checked against five resolvers plus the registry, with controls each time. What we could not establish for provider-ordered testing: turnaround times, specimen stability requirements, insurance or in-network arrangements, and current pricing of any kind.

The prices and kit details are historical, and we have labelled them as such. Every consumer figure in this review — the $279, the discount code, the kit contents, the HSA/FSA language, the out-of-stock status — comes from Internet Archive captures of a storefront that no longer operates, dated between March 2024 and September 2025. They tell you what this cost when you could buy it. They are not current prices, because there are no current prices. The panel target counts and the 95% sensitivity claim come from the company’s own marketing pages and we have not independently verified either; we have not seen a validation study, and no population is stated for that sensitivity figure.

The clinical evidence is about the category, not this company. The 60% concordance figure comes from a study of multiplex panels generally, not of Promus’s panels, which may perform differently — better or worse. We have no data on Promus’s own assay performance and we have not implied any. Equally, the 2025 comparison showing PCR detecting far more than culture used only 56 specimens from a highly selected group of patients who had already failed therapy.

The FAERS figures are not incidence rates. Spontaneous adverse event reports have no denominator, are subject to heavy reporting bias, and do not establish that a drug caused an event. We used them only to show which effects dominate GLP-1 reporting, which is independently well established.

Domains come back. A domain that does not resolve today may resolve next week. If you are reading this later and promusdirect.shop loads normally, the DNS section is out of date — the date of our check is 11 August 2026, and everything else in this review stands independently of it.

Frequently Asked Questions

Verdict

Read that rating carefully, because it is not the rating you might assume. We are not saying this is a bad laboratory. We have no evidence that it is, and a good deal of evidence that it is a real, properly certified, seven-year-old business doing legitimate work. The score rates the thing a reader of this site could actually buy, and the answer to that is: nothing.

Start with the good news, because it is unusual. We set out to test the phrase “high-complexity CAP-Accredited & CLIA-Certified laboratory”, fully expecting the sort of soft compliance language that falls apart on contact with a registry. It did not fall apart. CLIA 45D2156626. Certificate of Accreditation. CAP flagged as accreditor, no other. Current to 30 December 2027. Participating since October 2018. No termination, no ownership change. Cross-confirmed as an active clinical medical laboratory in the NPI registry. That claim is true, and we would rather report that plainly than manufacture a controversy.

Then we tried to buy something, and found we had arrived about a year late. There was a real consumer business here: Promus Direct, selling at-home STI and UTI kits at $279 each, with a pre-paid return mailer and a discount code called STEWARDSHIP. It ran from 2023 into 2025. In its final archived capture, on 18 September 2025, both products read “Currently Out of Stock”. Today the consumer domain is a parking page and the affiliate storefront returns NXDOMAIN from every resolver we tried and from the .shop registry itself. The laboratory carried on; the shop did not.

The affiliate network, meanwhile, still had this offer marked active with a March 2025 update — pointing at a domain that had been parked since December 2024. That is the part worth naming plainly, because it is a pattern rather than an accident: affiliate inventory does not get retired when a merchant quietly closes, so somebody eventually publishes an enthusiastic review of a shop that no longer exists. The only defence against that is for the publisher to actually click the link.

And underneath all of that sits the number worth taking away even if you never encounter this company. In 1,155 patients across 37 clinics, resistance genes found by multiplex PCR agreed with what the bacteria actually did 60% of the time — as low as 49% for some drug classes, and worse in the polymicrobial samples these panels are specifically sold to untangle. The reason is simple once you see it: a gene result cannot tell you which organism in the sample is carrying it. The study’s authors did not conclude the tests are useless. They concluded that gene detection alone is not a reliable basis for prescribing, and that paired with susceptibility testing it becomes valuable. That is the question to put to any lab selling you this: do you pair it?

There is a quiet irony in a product sold on antimicrobial stewardship. A test sensitive enough to find organisms culture misses will also find organisms that were not bothering you, and treating those is precisely the overprescribing that stewardship exists to stop. Whether it reduces antibiotic use or increases it depends entirely on the judgement of the person reading the report — which is a good argument for this being clinician-ordered, and a good argument against ever wishing it were not.

So: nothing to buy, nothing to click, and no commission earned on this page. What you can take away is the method. Ask any laboratory for its CLIA number, and spend two minutes checking it. Promus passed that test cleanly. Plenty do not, and now you know how to tell the difference.