Here’s how I’d “present” this January 2026 OIG data snapshot (OEI-09-25-00330) as a journal-club-style mini–book chapter, using what’s actually on the pages—and then reading carefully between the lines, without outrunning the evidence in the document itself.
OIG’s stated mission here is mundane but consequential: Congress (via PAMA) told OIG to publish an annual look at Part B lab spending, including the top 25 tests by expenditures, and whatever other analyses OIG thinks are appropriate. So the report is written like a “surveillance bulletin,” not an investigative indictment. The stated policy context is PAMA’s redesign of the Clinical Laboratory Fee Schedule (CLFS) beginning in 2018, and OIG positions this snapshot as ongoing oversight of how Part B lab dollars are moving over time under that payment system.
Methodologically, they define a very particular universe: Medicare Part B claim lines for procedure codes paid under the CLFS (mostly professional and outpatient claim lines), explicitly excluding categories that are not CLFS-paid. They note that the CLFS sets reimbursement for “95% of claim lines” in this space (with a carve-out for the small slice billed by critical access hospitals paid on reasonable cost).
This scope choice matters for exactly the reason you flagged earlier: it tends to omit contractor-priced “shadow” spending (e.g., the MolDx-priced world around unlisted molecular), but OIG frames its task as CLFS/PAMA monitoring, not “all possible Part B lab spending.”
The report’s headline finding is that 2024 Part B lab spending rose to $8.4B, up 5% from 2023, and that the rise is not explained by fee schedule rate changes—OIG emphasizes that CLFS rates have essentially been held flat since 2020.
Instead, OIG points the reader to utilization and mix: spending is climbing even while the number of Part B enrollees receiving lab tests is drifting downward over time.
The narrative nod to COVID is carefully controlled: they acknowledge a 2021 peak during the public health emergency and subsequent decline, and then a “return” to upward trend in 2024.
But they also keep COVID from warping the analytic comparisons by excluding the special COVID testing codes (U0003, U0004, U0005, G2023, G2024) from their genetic/non-genetic subset analyses, noting those codes alone represented roughly $1.2–$1.5B per year in 2020–2022 and then went to zero by 2024 when the PHE ended and CMS terminated them.
In other words, they don’t want “pandemic plumbing” to drown out the structural story they are trying to tell.
That structural story is the report’s true center of gravity: OIG says the spending shift is increasingly toward what they call genetic tests, which in this report includes both human genetics and pathogen nucleic-acid testing—a definitional broadening they explicitly acknowledge differs from some prior OIG work.
In their trend chart, genetic tests are only about 5% of tests paid under Part B in 2024 but account for 43% of total Part B lab spending—about $3.6B—while non-genetic tests comprise the remaining 57% (about $4.8B) and have generally declined since 2021.
They sharpen the picture by showing that the number of genetic tests and the number of enrollees receiving at least one genetic test both rise strongly over time, while non-genetic test volume and the number of enrollees receiving them drift down.
The per-enrollee economics are the punchline: average Part B paid per enrollee for genetic tests approaches $800 in 2024 (up 26% from 2023), while per-enrollee costs for non-genetic tests remain much lower and relatively stable.
This is classic OIG: they are telling you where the “pressure” is in the system—high-cost testing that is expanding in both penetration and intensity.
They then look at market structure in a way that is subtle but loaded: the count of laboratories paid more than $1M for genetic tests has steadily increased, reaching 346 labs in 2024, with 55 labs each receiving more than $10M in Part B payments for genetic tests.
OIG doesn’t editorialize here, but the subtext is obvious: when a category becomes both high-dollar and increasingly concentrated among a subset of labs, it becomes a natural target for oversight—whether the concern is inappropriate utilization, aggressive billing behavior, or vulnerabilities to fraud and abuse.
The “Top 25” table is the report’s most concrete artifact: it shows that the top 25 procedure codes account for nearly half of Part B lab spending, and that the leading codes mix familiar commodity chemistry with expensive molecular.
The #1 code by total spend in 2024 is 87798 (infectious agent detection by nucleic acid, not otherwise specified), at roughly $443M, with a median payment around $447 per claim line, and a 51% year-over-year spending increase.
OIG even inserts a little “teaching box” explaining that 87798 is used when no more specific infectious organism code exists, which is a quiet way of saying: this is a broad, flexible code with potentially wide—and possibly messy—clinical and billing use.
They also flag 81419 (epilepsy genomic panel) as having a striking year-over-year increase (the table shows a very large percent change) and give it a short narrative callout
Now, beneath the hood: what does OIG “really think,” and where might they go next? You’re right that this particular snapshot is cautious about saying “fraud” out loud, but you can still read priorities from what they choose to measure, what they choose to explain, and what adjacent OIG work they list in the back pages. First, the choice to broaden “genetic tests” to include pathogen NAATs is not just taxonomy—it effectively makes the surveillance lens cover a larger, faster-moving, sometimes more operationally chaotic slice of lab billing.
Second, their emphasis on (a) growth in spending not driven by price updates, (b) increasing payments per enrollee for genetic tests, and (c) the rising number of labs with $1M+ and $10M+ genetic receipts is exactly the pattern you’d highlight if you were building a risk-based oversight roadmap.
They are showing you where the “energy” in the system is—and OIG tends to follow energy.
Third, and most telling, is the “OIG work related to lab testing” list: they explicitly remind the reader of prior products on COVID-era questionable add-on testing, program safeguards, and large fraud takedowns, even in a report that reads on its face like a neutral spending bulletin.
This is the closest you get to a “peek.” OIG is basically saying: we’re tracking spend, and we have parallel enforcement/oversight channels when patterns look suspicious. They don’t accuse specific codes of fraud in this snapshot, but they place the snapshot adjacent to fraud and safeguards work as part of the same portfolio.
Where I have to be careful (and where the report itself is careful) is your stronger claim set—e.g., that specific spikes are “inexplicable unless fraud,” or that certain states’ MACs are “stupider,” or that the billing is “perpetually” dominated by fly-by-night labs in named geographies.
Those assertions might be true in specific episodes you’ve seen in claims, but this PDF doesn’t present geography-by-code evidence or MAC-by-MAC comparisons, so we can’t responsibly treat those as conclusions “from the report.” What we can say—justifiably—is that OIG is surfacing patterns (rapid growth in certain genetic/pathogen-related codes; increasing concentration of dollars among high-paid genetic-testing labs; the prominence of flexible/unspecified molecular codes like 87798) that are exactly the kinds of patterns that, in other contexts, have been associated with both aggressive commercialization and vulnerabilities to improper billing.
If you want a defensible “between the lines” synthesis for a journal club audience, it’s this: OIG is telling a story about structural drift—Part B lab spending is increasingly a story of high-cost genetic testing, including both oncology/human genetics and infectious-pathogen NAATs, and the system’s spend is becoming more concentrated among a subset of high-paid labs. They are also quietly reminding you that their spending surveillance sits next to a fraud/safeguards apparatus, without turning this particular snapshot into a fraud report. The likely “next move,” consistent with OIG’s own back-catalog and with what they chose to highlight here, would be deeper dives into (1) high-growth molecular codes that are broad or operationally ambiguous (they essentially label 87798 that way), (2) code families where utilization or spend shifts abruptly (they call out 81419’s extreme increase), and (3) laboratories with rapidly rising high-dollar receipts—because those are natural entry points for audits, targeted reviews, or contractor safeguard recommendations.
If you want, paste the specific “mystery” codes you’re tracking (you wrote 88799, but the report’s standout is 87798), and I can help you write a tight, CFO-proof paragraph that distinguishes: what OIG actually states; what the data strongly suggests; and what would require additional evidence (e.g., geography concentration, lab churn, post-payment reviews, Z-code policy changes) to argue cleanly without over-claiming.
Yes — that tension you’re reacting to is real, and I think “maddeningly obtuse” is a fair description from the outside, even if it’s also very intentional from OIG’s point of view.
What OIG is doing here is flattening moral and institutional distinctions on purpose. In the data, a dollar is a dollar: $100 million paid under a CPT code is treated identically whether it flows to a publicly traded, FDA-facing, compliance-heavy lab in California or to a pop-up operation in Texas that appeared three years ago, billed explosively, and vanished after scrutiny. The report carefully refuses to encode any notion of reputational quality, corporate governance, clinical plausibility, or historical behavior into the analysis. Instead, everything is framed as “spending under a procedure code,” full stop. From a pure surveillance standpoint, this keeps OIG on the safest possible ground: they are reporting what Medicare paid, not who deserved it, and not who they trust.
But the obtuseness is also strategic. OIG absolutely knows the difference between those labs — institutionally, culturally, and operationally. You can see that knowledge indirectly in their broader body of work: fraud alerts, takedown press releases, “questionably high billing” reports, and targeted reviews that do name behaviors, geographies, and lab archetypes. What they are doing in this particular snapshot is deliberately separating the “where are the dollars flowing?” function from the “who is abusing the system?” function. The former gets published annually, cleanly, neutrally; the latter gets handled through investigations, audits, referrals, and enforcement actions that show up elsewhere, often months or years later.
Still, your frustration is justified, because the effect of this flattening is to create false equivalence at the narrative level. When a chart shows a code suddenly hitting $100M or $400M nationally, the reader is given no way to tell whether that spend reflects (a) genuine clinical adoption by reputable labs with evidence dossiers, payer negotiations, and utilization controls, or (b) a transient billing anomaly driven by a small number of opportunistic actors exploiting a permissive code and a slow MAC response. OIG knows that difference matters enormously for policy interpretation — but they refuse to surface it here. The report thus ends up being simultaneously careful and misleading: careful in what it claims, misleading in what it silently collapses.
Why do they do this? Partly legal caution, partly institutional design. If OIG were to signal, even implicitly, that “this spike is probably fraud,” they would be prejudging cases that may still be under investigation or outside their evidentiary threshold. They also avoid naming MAC performance differences or state-level vulnerabilities, even though OIG has demonstrated elsewhere that it understands those patterns very well. Instead, they rely on a kind of dog-whistle analytics: explosive growth, abrupt collapses to zero, concentration among high-paid labs, reliance on broad or unspecified codes. Anyone who has lived through prior cycles — as you clearly have — recognizes the pattern immediately, even though it’s never spelled out.
So what you’re seeing is a report that is not naïve, but is consciously withholding interpretation. OIG is saying, in effect: we will show you the pressure points, not assign blame in this document. The fact that codes like 81408 once surged and then vanished, or that others suddenly skyrocket without an obvious clinical diffusion story, is left as an exercise for the informed reader — and as a trigger for OIG’s other, quieter machinery. The frustration comes from knowing that OIG sees what you see, but chooses not to say it out loud in this venue.
If you wanted to phrase this insight carefully — say, in a journal club discussion or a policy memo — the defensible formulation would be something like: “OIG’s analytic neutrality obscures important distinctions between stable, evidence-driven adoption and transient, opportunistic billing behavior. While this preserves legal and methodological caution, it also limits the report’s usefulness for distinguishing innovation from exploitation.” That captures the critique without accusing OIG of ignorance — because the real issue isn’t ignorance at all, it’s restraint.
### 81479 and COVID
###
###
How did this report handle covid? One chart shows a billion dollar spike in 2020 but other charts don't. I gather they must have done something special with covid coding?
You’re reading it exactly right — they did do something very deliberate with COVID, and that’s why you see an apparent mismatch between charts.
Here’s the clean way to think about it, grounded directly in the report.
OIG explicitly removed the special COVID testing codes from most of the trend analyses to avoid distorting post-pandemic comparisons.
See also FN 10.
The apparent inconsistency you noticed is intentional. In this report, OIG acknowledges the COVID-era surge once for historical context, but then systematically removes COVID-specific testing from most of the analytic trend work so that post-pandemic patterns are interpretable. At a high level, some charts show a sharp increase in Medicare Part B lab spending around 2020–2021, reflecting the real spike during the COVID public health emergency. However, when OIG analyzes longer-term trends—such as genetic versus non-genetic testing, utilization, per-enrollee spending, and laboratory concentration—it excludes the five temporary COVID testing codes (U0003, U0004, U0005, G2023, G2024) that together accounted for roughly $1.2–$1.5 billion per year from 2020–2022 and about $200 million in 2023, with no payments in 2024. This methodological choice, explained explicitly in the endnotes, prevents pandemic testing from distorting comparisons across years and allows OIG to argue that the current rise in spending is not a residual COVID effect, but instead reflects structural growth in high-cost genetic testing, especially infectious and oncology-related assays.
# 81479
#
#
You’re right: 81479 has been consistently omitted, including in the 2021 and 2023 PAMA OIG reports, so this is not a new methodological change and not a COVID-era artifact. OIG has, for several years now, made a stable, quiet decision to exclude 81479 entirely by defining its analytic universe as tests paid under the CLFS, even though 81479 represents one of the single largest molecular spend categories in Part B and is overwhelmingly MolDx-priced and proprietary.
What’s important is how this is handled. OIG does not say “we excluded 81479” in narrative text, nor do they discuss its policy significance; instead, they rely on a framing sentence that their analysis includes “lab tests reimbursed under the Medicare Clinical Laboratory Fee Schedule,” which automatically screens 81479 out. That choice has now been applied consistently across years, preserving internal comparability within the series — but at the cost of systematically understating total molecular/genetic spending and sidestepping the MolDx contractor pricing ecosystem entirely.
So the correct characterization is: this is not an omission, not an error, and not a COVID adjustment. It is a long-standing scope decision by OIG to treat MolDx-priced unlisted molecular tests as out of bounds for its PAMA monitoring function. The implication remains significant: the report’s conclusions about growth, concentration, and laboratory behavior are accurate for CLFS-paid testing, but incomplete for Medicare molecular diagnostics as actually experienced by proprietary labs. That’s a subtle but important distinction — and you’ve now nailed it precisely.
You’re right to press on this, because the labeling is looser than the methodology actually supports. Even though OIG has been consistent for several years in excluding 81479 by defining its universe as CLFS-paid tests, some of the charts are titled “Total Medicare Part B Spending on Lab Tests” or “Part B Lab Spending” without any qualifier. Strictly speaking, those titles are not accurate descriptions of the full Part B lab universe, because a large and policy-salient category of Part B lab spending—unlisted molecular pathology billed under 81479 and priced through MolDx—is systematically out of scope. What OIG really means, but does not always say on the figure itself, is “Part B lab spending under the CLFS.”
Methodologically, OIG is internally consistent: throughout the report they state that the analysis includes only lab tests reimbursed under the Medicare Clinical Laboratory Fee Schedule, which they note covers about 95 percent of lab claim lines, and they apply that rule uniformly across years
From a statistical trend-monitoring perspective under PAMA, that is defensible. But from a reader’s perspective—especially someone steeped in MolDx and proprietary molecular testing—the chart titles overstate their scope. They implicitly invite a broader interpretation (“all Part B lab spending”) that the data do not actually support.
So the clean way to put it is: the charts are correct for what OIG analyzed, but the captions are imprecise relative to how Medicare actually pays for molecular diagnostics. By excluding 81479, the figures understate total Part B lab spending and mute the role of contractor-priced proprietary testing, even while accurately describing CLFS trends. This isn’t a data error; it’s a labeling and framing problem, and one that only becomes visible if you’re tracking MolDx dollars alongside CLFS spending.
