LabReflex

Christopher Zahner, MD

4.9 (7)
Medicine

A conversational podcast about more innovative diagnostics, lab insights, and the future of clinical testing. Hosted by Dr. Christopher Zahner, LabReflex brings expert voices, industry trends, and practical conversations straight from the laboratory bench to your brain.

20h ago

Pulse: ADLM Signals, Cyclospora Surges, and CLIA Redraws the Lab

# ADLM Signals, Cyclospora Surges, and CLIA Redraws the Lab ## SummaryChris and Aakash break down three major signals from ADLM 2026: simultaneous centralization of high-throughput systems and decentralization of point-of-care testing, AI shifting from demos to implementation reality with data infrastructure as the real bottleneck, and the Cyclospora outbreak exploding to 4,200 confirmed cases across 41 states. They also unpack CMS and CDC's request for information on modernizing CLIA (last updated 1988) for AI-assisted diagnosis, remote oversight, and data-only facilities, plus the FDA-approved Elecsys sFlt-1/PlGF ratio test for pre-eclampsia risk stratification in hospitalized pregnant patients 23–34 weeks gestation. ## Key takeaways- **Centralization and decentralization are happening simultaneously** — vendors are building massive high-throughput analyzers for core labs while pushing smaller instruments closer to patients, creating a "mismatch" in product strategy.- **AI implementation has moved past the demo phase** — the real questions now: who monitors data drift, who owns the data, who tracks software versions, and how performance varies across patient populations.- **Data infrastructure is the bottleneck, not algorithms** — companies gaining traction (like Bunker Hill) function as infrastructure plays, solving how to connect to health system databases; CAP is prioritizing AI inspection frameworks.- **Cyclospora outbreak is not slowing** — 4,200 lab-confirmed domestic cases, 7,400+ probable cases under investigation, 300+ hospitalizations, 41 states; CDC warns reporting lags illness onset by ~6 weeks; linked to Taylor Farms iceberg lettuce recall (~2,000 confirmed cases).- **Testing gaps obscure true outbreak scope** — Cyclospora isn't on routine GI panels; genetic tracking is harder than Salmonella (no whole-genome sequencing); labs may need to spin up testing if this becomes seasonal.- **CLIA modernization RFI is a critical inflection point** — CMS/CDC asking whether AI-assisted interpretation constitutes a "test," whether data-only facilities count as labs, and how remote competency assessment works for 90+ clinic locations.- **FDA-approved pre-eclampsia test (Elecsys sFlt-1/PlGF ratio)** — 91% sensitivity, 77% specificity for predicting severe pre-eclampsia within 2 weeks in singleton pregnancies 23–34 weeks; intended to support (not replace) clinical judgment; education needed to prevent off-label confirmatory use. ## Chapters- Opening and Personal Catch Up- ADLM Signals Overview- Centralization and Decentralization Trends- AI Implementation Beyond Demos- Data Infrastructure as AI Bottleneck- Cyclospora Outbreak Escalation- CLIA Modernization Request for Information- FDA Approved Pre-eclampsia Test ## Mentioned- **ADLM** (Association for Diagnostics & Laboratory Medicine) — 2026 conference- **Bunker Hill** — health AI company focused on data infrastructure/connectivity- **CAP** (College of American Pathologists) — prioritizing AI inspection frameworks- **CDC** — Cyclospora outbreak tracking, reporting lag warnings- **CMS** — CLIA modernization request for information (with CDC)- **Taylor Farms** — iceberg lettuce recall linked to Cyclospora- **Elecsys sFlt-1/PlGF ratio** (Roche) — FDA-approved pre-eclampsia risk test- **CLIA** (Clinical Laboratory Improvement Amendments) — 1988 legislation under review ## Quotes> "The data infrastructure they call it a layer... is going to be as important as anything else as far as AI implementation... a lot of the health AI companies that are actually gaining real traction are almost like infrastructure companies." > "The core important questions still need to be asked as, like who monitors the data drift, who owns that data? Who tracks these software versions and like who checks the performances differences amongst different patient populations." > "It used to just be build the walls higher and thicker, right? That's pretty much what we thought of and now it's a very different world." (00:00) - Opening and Personal Catch Up (01:32) - ADLM Signals Overview (02:41) - Centralization and Decentralization Trends (04:00) - AI Implementation Beyond Demos (05:44) - Data Infrastructure as AI Bottleneck (08:07) - Cyclospora Outbreak Escalation (13:59) - CLIA Modernization Request for Information (19:29) - FDA Approved Pre-eclampsia Test
Jul 24

Deep Dive: Disruption in the Laboratory

What is Disruption in the LaboratoryDefining disruption requires looking at workflow and ROI rather than just technical novelty. SummaryMarket disruption in the lab is not automatically signaled by new technologies or game-changing results. True disruption addresses underlying economic, workflow, and staffing bottlenecks. The focus should be on whether a technology simplifies the lab's work or shifts complexity elsewhere, ultimately impacting the return on investment for the institution. Key takeaways* Disruption in the lab must address underlying economics, workflow, and utility of results, not just technical novelty.* Novelty alone does not equal disruption; it must solve a specific operational problem like staffing shortages or diagnostic delays.* The key question is whether a technology removes more complexity than it creates for laboratory personnel.* Disruption often involves shifting value capture: the lab may implement a change that benefits the clinician, making the sale harder to justify internally.* Physical AI solutions, like robotics, represent disruption by automating manual physical processes, which can be a path forward.* New testing methods must simplify bench work for fungal or microbiology teams rather than just expanding the scope of testing. MentionedABB, Roche, Physical AI, Mass Spec, Spine Stat, AI tool for coronary artery calcifications on CT scans. (00:00) - Intro (00:08) - Start (01:22) - Disruption In The Laboratory Defined (02:34) - Optimizing Ancillary Tests Versus New Tools (06:56) - Physical AI And Robotics Change (11:02) - Disruption In Fungal Testing Speed (15:25) - Testing Expands Footprint Not Workflow (23:16) - The Real Measure Of Disruption (25:54) - Outro
Jul 13

Pulse: After the Cookout Parasites, Pneumonia, and the Lab’s Summer Surge

# After the Cookout: Parasites, Pneumonia, and the Lab’s Summer Surge ## SummaryThe conversation covered outbreaks like Cyclosporiasis and Legionnaires Disease, showing how laboratory testing connects to public health investigations. It also discussed the need for better data integration within labs and the potential for diagnostic informatics to drive organic growth in the lab industry. ## Key takeaways* Outbreaks like Cyclosporiasis and Legionnaires Disease highlight the need for laboratories to act as central hubs connecting patient diagnostics with environmental and public health investigations.* Connecting local testing data with broader signals, such as social media searches, could alert labs to potential outbreaks faster.* The laboratory needs to move upstream by driving decisions based on aggregated data rather than just responding to requests from health systems.* Integration of genomic profiling and molecular assays into platforms like EHRs can improve availability and utilization of testing.* The focus for the lab industry should shift toward data informatics to facilitate organic growth, rather than solely focusing on vendor sales pitches. ## Chapters* Welcome To LabReflex* Summer Outbreak Effects* Legionnaires Disease Investigation* Connecting Data And Outbreaks* Oncology Integration Story* Lab As Central Hub ## MentionedQuest Diagnostics, Flatiron Health, ADLM 2026 (00:00) - Intro (00:08) - Welcome To LabReflex (03:54) - Summer Outbreak Effects (08:20) - Legionnaires Disease Investigation (13:10) - Connecting Data And Outbreaks (15:08) - Oncology Integration Story (24:43) - Lab As Central Hub (26:00) - Outro
Jul 9

Deep Dive: The DMT Defined (Part 2 of 6)

DMT Defined Defining what a DMT is and what it is not. Summary The discussion focuses on defining Diagnostic Management Teams (DMTs) by emphasizing that a true DMT must be expert-driven, patient-specific, and aim to establish a single diagnosis in the context of the patient's illness. It contrasts this with simpler processes like reflex testing or utilization management, highlighting the need for comprehensive interpretation rather than just providing test results. Key takeaways A DMT must recommend necessary tests and provide a patient-specific diagnosis within the clinical context.The value of a finding depends on whether it is provided at a time when clinical decisions are being made.Reflex testing is a tool, not the DMT itself; the DMT establishes the diagnosis based on that information.A true DMT requires expert-driven input and aims for a single diagnosis, unlike utilization management which focuses on groupthink.The process needs to be patient-specific, requiring time, which is why it is often avoided in favor of case conferences for less complex issues.Chapters 0:00 Defining What A DMT Is 1:10 The Evolution Of DMTs 3:29 Critical Ingredients Of A DMT 12:45 DMT Versus Utilization Management 14:04 Reflex Testing Is Not A DMT 37:10 Checklist For Real DMTs Mentioned Dr. Michael Appasada, Penn, Mass General, Spalding Rehabilitation Hospital, Amazon One, Amazon Health, Open Evidence AI, Plavix. Quotes > You have to say what tests are necessary and you have to provide a diagnosis for that patient, patient specific in the clinical context of that patient. > It is a consult that is expert driven and patient specific in the context of the patient's illness. And its goal is to conclusively establish a single diagnosis. > Reflex testing is not the DMT because reflex testing says to figure out the diagnosis, you would want to do this test first based upon that you do this test or that test. (00:00) - Intro (00:08) - Defining What A DMT Is (01:19) - The Evolution Of DMTs (03:38) - Critical Ingredients Of A DMT (12:53) - DMT Versus Utilization Management (14:13) - Reflex Testing Is Not A DMT (37:18) - Checklist For Real DMTs
Jun 30

Pulse: What’s Happening Now and What We’re Watching Before ADLM

(00:00) - Intro (00:08) - Start (00:08) - Welcome To LabReflex (02:31) - Measles Outbreak And Preparedness (10:07) - Enterprise Ready Clinical AI (22:39) - Point Of Care Testing Ecosystem (25:21) - Who Owns The AI Data (29:08) - Lab Business And Future Growth (31:33) - Outro
Jun 22

Pulse: Can Labs Keep Up? Workforce, Quality, and AI in Modern Laboratory Medicine

Workforce and AI in Laboratory Medicine Summary The medical laboratory workforce faces a shortage compounded by retirements, requiring systemic changes in education. Guidance from CAP focuses on structuring case review to reduce interpretive errors. Furthermore, the integration of AI into lab medicine is moving toward validation stages, leveraging the standardized numeric data inherent in laboratory results. Key takeaways The laboratory workforce shortage is not just about vacancies; it involves losing decades of experience through retirements, making it harder to backfill roles with equivalent expertise.MLS and MLT education must become a forefront field to address the shortage, requiring significant investment, such as proposals for establishing medical or MLS schools.Diagnostic error reduction guidance requires inter- and intra-divisional assessment of sign-outs to reinforce a structure for evaluation rather than seeking perfection.Laboratory medicine data—numeric measurements—is ideally suited for AI implementation because it is straightforward, unlike free-text physician notes.The future involves labs being involved in the validation and verification stage of AI development, ensuring models accurately reflect laboratory results.Chapters 0:54 Workforce Crisis In Labs 6:06 Diagnostic Error Reduction Guidance 8:59 AI Solutions In Healthcare 14:20 Lab Medicine And AI Future 16:46 Final Thoughts On Topics Mentioned CAP, AMA, Microsoft, ChatGPT, Anthropic. (00:00) - Intro (00:08) - Start (00:08) - Welcome To LabReflex (01:03) - Workforce Crisis In Labs (06:14) - Diagnostic Error Reduction Guidance (09:08) - AI Solutions In Healthcare (14:28) - Lab Medicine And AI Future (16:54) - Final Thoughts On Topics (18:15) - Outro
Jun 19

Field Notes 1: UAMS and MLS Training

Field Notes 1: UAMS and MLS Training The challenge of staffing Medical Laboratory Science positions, particularly in rural areas, and a scalable solution for recruitment. Summary There is a significant shortage of Medical Laboratory Science staff, especially in rural settings, which creates severe staffing challenges for hospitals. The University of Arkansas for Medical Sciences developed an approach to address this by training personnel locally rather than requiring relocation, which proved effective in attracting and retaining talent. Key takeaways Unfilled MLS positions are in the eight to ten percent range nationally.Unfilled rural MLS positions are significantly worse, at 15% or higher.Moving staff from metropolitan areas to rural communities for training is less effective than training locally.A successful model involves allowing trainees to stay in their community while completing training and continuing employment.The Arkansas program has achieved a 100 percent pass rate on board exams.This approach offers a scalable model for addressing healthcare staffing needs in rural areas.
Jun 17

Deep Dive: Beyond the Result - The Diagnostic Management Team Story (Part 1 of 6)

DMT - Origins - Ep01 The history of the Diagnostic Management Team concept and why it took decades to fully implement. Summary The evolution of diagnostic practice was driven by a need for clinicians to understand lab results, which faced significant resistance from the medical system's incentive structure. The speaker details how the shift from laboratory-focused thinking to patient-focused diagnosis required overcoming obstacles related to payment systems and professional ego. This process ultimately led to the development of the Diagnostic Management Team concept and its modern application with tools like AI. Key takeaways Early efforts to implement diagnostic changes faced resistance because the existing payment system incentivized focusing on procedures rather than comprehensive patient outcomes.The movement toward a Diagnostic Management Team required shifting focus from laboratory-minded clinicians to patient-focused experts.Obstacles included finding faculty and residents willing to commit time to multidisciplinary learning, as many were focused only on routine tests.The concept evolved by realizing that the system needed an expert to provide answers, leading to the idea of a Diagnostic Center where experts could share knowledge across specialties.The evolution of the DMT was slow because it required overcoming systemic barriers related to payment structures and professional incentives over many years. (00:00) - Intro (00:08) - Start (01:08) - Introduction And Thank You (02:57) - The Start Of The Idea (07:23) - Obstacles To Change (28:09) - Systemic Failure And Incentives (42:12) - AI And The Future Of DMT (44:19) - The Diagnostic Center Concept (48:39) - Outro

4.9

out of 5

7 Ratings

A conversational podcast about more innovative diagnostics, lab insights, and the future of clinical testing. Hosted by Dr. Christopher Zahner, LabReflex brings expert voices, industry trends, and practical conversations straight from the laboratory bench to your brain.

Creator

Christopher Zahner, MD
Years Active

2025 - 2026
Episodes

39
Rating

Clean
Copyright

© 2025 Christopher Zahner / LabReflex