Here’s a hard truth no one wants to say out loud: we are failing our clinicians when it comes to molecular diagnostics.
I coordinate emergency diagnostics for a busy hospital network. In my role, I triage urgent PCR requests from the ED and ICU. And over the last five years—especially since the pandemic normalized PCR as a household word—I’ve seen a dangerous trend: clinicians treat PCR results like a weather forecast. Fast, binary, and final.
It’s not their fault. We in the lab haven’t done the one thing that would change everything.
We haven’t educated them on how PCR actually works.
Opinion: An informed user is a safer user
This is my hill. I believe that helping clinicians understand the mechanics of real-time PCR—not just the acronym—makes them better diagnosticians. And an educated clinician asks better questions, orders fewer unnecessary tests, and trusts the result appropriately.
Here’s why I think most labs get this backwards.
Argument 1: The “black box” problem is a patient safety issue
When a clinician orders a “respiratory panel” or a “meningitis panel,” they often have no idea what’s happening inside the black box (which, today, might be a Roche cobas 6800 or a GeneXpert). I don’t blame them. Until three years ago, I didn’t fully understand the amplification curve either.
The problem? They interpret a positive PCR as “infection present.” That’s true—mostly. But PCR detects nucleic acid. Not necessarily viable, live, replicating pathogen. For a host of respiratory viruses—especially in immunocompromised patients—PCR can stay positive for weeks after the infectious period ends.
I’ve seen patients kept in isolation for 12 days because the PCR was still positive. Their clinical symptoms had resolved. The infection control team had to intervene. The cost? A bed block, a disgruntled patient, and wasted PPE.
“The test says ‘detected.’ The patient looks fine. Which do you trust? The answer is: you need to understand what ‘detected’ means in that context.”
Educating clinicians on the difference between a Ct value (cycle threshold) and a simple positive/negative output would change this. A high Ct value in a recovering patient is a clue, not a crisis. A low Ct value in a symptomatic patient is actionable. But if all they get is a binary result, they’ll treat both the same way—and that’s a failure of the diagnostic ecosystem.
Argument 2: Speed is seductive, but understanding is sustainable
Roche’s point-of-care PCR solutions, like the cobas Liat, are fast. We’re talking 20 minutes for influenza/RSV. That’s incredible. And I use it every flu season for our ED.
But here’s the trap: “fast” convinces clinicians that PCR is just another STAT lab, like a troponin or a basic metabolic panel. It’s not.
I once had a surgeon waiting for a cobas Liat result before deciding on an emergency appendectomy in a child. The test was negative. He was relieved—he could avoid surgery. But when I asked him what a “negative” PCR meant in that context, he said, “It means she doesn’t have an infection I need to worry about.”
Wrong answer.
A negative PCR means the nucleic acid target was below the limit of detection. It doesn't rule out an early infection, a sample quality issue, or a pathogen not on the panel. The surgeon ended up consulting with infectious disease, who ordered a second test 24 hours later. It was positive for adenovirus. The child did not need surgery, but the clinical decision was delayed because the first result was interpreted as a definitive “all-clear.”
The lesson? Speed without understanding creates risk.
Argument 3: You don’t need to be a scientist—but you need to care about the manual
I’m not suggesting every ED physician become a molecular biologist. That would be unreasonable. But I’ve seen entire departments treat manufacturer-provided training as a checkbox exercise.
I wish I had tracked how many times clinicians skimmed the quick-reference card for our Roche cobas 4800 system and assumed they knew the workflow. (I really should have kept a log.)
The reality: PCR instruments have specific sample type requirements, stability conditions, and contamination risks. If a clinician doesn’t understand that a nasopharyngeal swab in the wrong transport medium can give a false negative—they’ll blame the kit, not the collection method.
(Note to self: actually, this is a training gap we can fix with a 15-minute huddle. Why haven’t we done it yet?)
Educating our clinical colleagues on these nuances doesn’t just reduce frustration. It reduces real error. In March 2024, I had a resident ask why a CT value was “high” on a patient with no symptoms. They had 10 minutes before discharge. I explained: “It might be residual RNA. Clinical correlation is needed—not a repeat test.” That saved the patient an unnecessary hour of waiting and a $200 repeat test. Not huge. But multiply that by every shift.
But wait—is it really my job to educate everyone?
I get this pushback a lot. “I’m a lab professional, not a teacher.” Or: “Roche provides all the materials—that’s their job.”
Fair point. But here’s the thing: the materials are great, but they’re not delivered in the moment. When a clinician is standing in your lab doorway asking, “What does this Ct number mean again?” you either help them understand—or they walk away with the same confusion. Roche can’t be there at 3 AM. You can.
If you want to argue that this is “scope creep,” fine. I’d argue the opposite: it’s scope deepening. The lab’s value isn’t just in producing the result—it’s in making the result usable. And that requires education.
(Full disclosure: I’m aware that not every lab has the staffing to do this. If you’re running 500 samples a day with a skeleton crew, I’m not judging. This is my view based on our mid-volume setup where we run about 200 PCRs a day.)
Conclusion: The goal isn’t to make everyone an expert. It’s to make everyone safe.
I still believe that PCR is one of the most powerful tools we have. Roche’s diagnostics portfolio—from the high-throughput cobas 6800 to the compact cobas Liat—has changed how fast and how accurately we respond to infectious threats. But a tool is only as good as the person using it.
An informed clinician doesn’t just use PCR; they respect it. They know when to question a result, when to wait, and when to act. And that’s the kind of care I want to be part of.
So yes—I spend time educating my clinicians. Not because I have to. Because I’ve seen what happens when they don’t understand. And the cost of that ignorance is a lot more than a reprint fee.
