This is an informal, even deeper excurision into 2007-2008 OIG documents about in office ancillary services.
###
Please provide a detailed discussion of the ins and outs of this 2007-2008 rulemaking for the detail oriented reader who is fascinated by this CMS policy history
###
The 2007–2008 CMS rulemaking embedded in the CY 2008 Physician Fee Schedule final rule occupies a peculiar and revealing place in Medicare policy history. On its face, it is a technical exercise in billing rules—anti-markup provisions, reassignment mechanics, and definitional clean-ups. In reality, it represents one of CMS’s clearest attempts to reassert control over the economic logic of diagnostic testing without banning practices outright or rewriting the Stark statute. For readers fascinated by how CMS actually governs, this episode repays close attention.
By the mid-2000s, CMS had come to believe that it was confronting not a handful of bad actors, but a structural misalignment that had emerged organically from well-intentioned statutory and regulatory changes. Stark law prohibited physician self-referral, yet preserved the in-office ancillary services exception to avoid crippling integrated practice. The Medicare Modernization Act of 2003 liberalized reassignment rules, allowing billing entities to aggregate and bill for services furnished by others under contract. Separately, Medicare had long permitted billing for purchased diagnostic tests, constrained by anti-markup limits designed to prevent pure arbitrage. Each of these rules made sense on its own. Together, however, they created a permissive environment in which physician groups could order, control, and profit from diagnostic services that were increasingly detached from the clinical office itself.
CMS’s concern was not that these arrangements were facially illegal. In fact, many of the “pod lab” models that drew attention during this period were carefully structured to satisfy Stark’s technical requirements, including the centralized building definition. The problem, as CMS framed it, was that these arrangements undermined the original policy balance. They allowed ordering physicians to capture profits from diagnostic volume in ways that resembled the very self-referral risks Stark was intended to mitigate, while exploiting reassignment and purchased test rules that were never designed for this degree of vertical integration.
At the heart of the rulemaking is CMS’s reframing of a deceptively simple question: who actually furnishes a diagnostic service for Medicare purposes? Historically, Medicare had been willing to deem a service “furnished” by a billing physician or group even when the work was done by others, provided certain safeguards were met. By 2007, CMS concluded that this legal fiction had been stretched too far. The agency began to focus less on formal labels—employee versus contractor, purchase versus reassignment—and more on functional realities: who ordered the test, where the work was performed, and whether the billing entity stood to profit from volume it controlled.
This shift is most visible in CMS’s treatment of “office” versus “centralized building.” For Stark purposes, a centralized building could qualify as an in-office ancillary location even if no patient care occurred there. CMS accepted that premise under Stark but refused to import it wholesale into payment policy. In the anti-markup context, CMS deliberately narrowed the concept of an “office” to mean a place where the physician or group regularly furnishes patient care. Space leased exclusively for laboratory operations—even if compliant with Stark—did not qualify. This distinction effectively severed the link between Stark compliance and payment privilege, and it did so without amending Stark itself.
The extension of anti-markup limits to the professional component of diagnostic tests was the most controversial aspect of the rule. Commenters argued that Congress had explicitly addressed only the technical component in section 1842(n) of the Social Security Act, and that physician services were governed by section 1848’s fee schedule rules. CMS’s response was telling. Rather than claiming explicit congressional instruction, CMS emphasized discretion. Reassignment and purchased test billing were framed as options that CMS could condition in the interest of program integrity. Physicians remained free to bill directly for services they personally performed. If they chose instead to bill for services furnished by others, CMS asserted authority to limit the resulting payment to prevent markup-driven incentives.
Employment status received similar treatment. CMS rejected arguments that full-time employees should be treated differently from part-time employees or independent contractors. From the agency’s perspective, these distinctions were formalities that obscured the economic substance of the arrangement. If the work was performed outside the billing entity’s office, or if it was purchased outright, anti-markup applied. This insistence on neutrality across employment models signaled CMS’s broader intent to avoid regulatory gamesmanship and to focus on incentive alignment rather than contractual nuance.
The rulemaking record is also notable for CMS’s candor about overutilization. CMS acknowledged that it was difficult to determine, case by case, whether increased biopsy volume or diagnostic intensity resulted from profit incentives or evolving standards of care. Yet the agency declined to resolve that clinical debate. Instead, it relied on decades of OIG, GAO, and policy research showing that, in aggregate, self-referral correlates with increased utilization. CMS’s conclusion was pragmatic: when financial incentives and ordering authority align, the risk of overuse increases, even if individual practitioners believe their decisions are clinically justified.
Quality arguments received extensive attention but little regulatory traction. Commenters described improved turnaround times, closer collaboration between clinicians and pathologists, and enhanced subspecialization under pod lab arrangements. CMS did not deny that such benefits could exist. What it refused to do was build payment policy around self-reported quality advantages tied to ownership or profit structures. The agency consistently returned to the principle that Medicare payment rules must be administrable, neutral across specialties, and resistant to incentive-driven distortions. Quality oversight, in CMS’s view, belonged elsewhere—in accreditation, CLIA enforcement, and professional standards—not in exceptions to anti-markup safeguards.
Equally revealing are the actions CMS chose not to take. The agency did not abolish the in-office ancillary services exception, did not prohibit pod labs outright, and did not dictate clinical protocols for biopsies or diagnostic intensity. Instead, it used payment policy as a moderating force. Practices could continue to structure their operations as they wished, but if they sought to bill Medicare for services furnished outside their true clinical offices, they would be limited to cost-based reimbursement rather than profit-generating markups.
Seen in retrospect, the 2007–2008 rulemaking foreshadowed much of CMS’s later regulatory posture. It reflects an enduring preference for economic guardrails over categorical bans, for functional analysis over formal compliance, and for payment policy as a lever to shape behavior without criminalizing innovation. For readers interested in how CMS actually governs complex clinical markets, this episode stands as a textbook example of administrative pragmatism—careful, contested, and quietly consequential.
###
These advisory documents covered in office ancillary services. There is a great deal of discussion of what in office means. Such as same building, etc. Is there any discussion of what ancillary services means and whether that definition should be revised? I don’t think there would be, but I don’t know.
###
Short answer first, then the explanation: no, these documents do not meaningfully revisit or redefine what “ancillary services” means, and that omission is deliberate and revealing. The discussion is overwhelmingly about where services are furnished (“in-office,” “same building,” “centralized building”) and how they are billed, not about what kinds of services qualify as ancillary in the first place
2007 CMS 1127 72FR66222 Physici…
.
What follows is the longer, historically grounded explanation.
Across the 2007–2008 rulemaking, CMS devotes enormous analytic energy to the phrase “in-office” and almost none to “ancillary services.” This asymmetry is not accidental. CMS treats the scope of “ancillary services” as essentially settled law, while treating the geography and economics of furnishing those services as the locus of emerging abuse.
The statutory foundation matters here. The in-office ancillary services exception in Stark is defined in section 1877(b)(2) of the Social Security Act, and it lists categories of services—clinical laboratory services, radiology, certain diagnostic tests, and related items—that Congress explicitly designated as “ancillary.” By the time CMS reached the mid-2000s, this list had been in place for years and had already survived multiple rounds of litigation, OIG interpretation, and rulemaking. Reopening the definition of “ancillary” would have meant questioning Congress’s own policy judgment that certain diagnostic services are sufficiently integral to physician practice to warrant special treatment.
CMS clearly did not want to do that.
Instead, CMS’s concern in the CY 2008 rulemaking is that services unquestionably covered by the ancillary exception were being furnished in ways that no longer looked ancillary in any meaningful sense. The services themselves—pathology interpretations, histology processing, imaging reads—were not in dispute. What troubled CMS was that they were being carved out into quasi-independent production units, geographically and operationally detached from the physician’s clinical office, while still benefiting from the financial privileges of in-office status.
That explains why the rulemaking is saturated with discussion of “same building,” “centralized building,” and “office of the billing physician or supplier,” yet contains virtually no debate about whether pathology or imaging should be considered ancillary at all. CMS repeatedly signals that the problem is not what services are ancillary, but how far practices have stretched the concept of being “in-office” to preserve the economic advantages attached to that designation.
Importantly, CMS also draws a careful jurisdictional boundary. In multiple places, the agency notes that it merely solicited comments on the scope and application of the IOAS exception but did not propose regulatory text revising the exception itself. As CMS states explicitly, any revision to the in-office ancillary services exception would require a future notice of proposed rulemaking, not a payment-rule add-on. This reflects both administrative caution and legal realism: redefining “ancillary services” would almost certainly provoke intense political opposition and statutory interpretation challenges.
What CMS does instead is far more subtle. Without touching the definition of ancillary services, it decouples Stark compliance from payment entitlement. A service can still qualify as an in-office ancillary service for Stark purposes, yet be subject to anti-markup limits for Medicare payment because it is not performed in the “office of the billing physician or supplier” as CMS defines that term for payment integrity purposes. In other words, CMS leaves the what alone and constrains the benefit.
This approach also avoids a doctrinal trap. If CMS had attempted to narrow the definition of ancillary services, it would have been forced to articulate why certain diagnostic services are no longer “ancillary” to physician practice—a concept that becomes increasingly unstable as medicine grows more specialized and technology-driven. By contrast, focusing on location, ordering authority, and billing structure allows CMS to address abuse without redefining clinical relationships.
So your instinct is correct: there is essentially no effort in these documents to rethink or revise the meaning of “ancillary services.” The silence is itself a policy choice. CMS treats the ancillary-services list as a settled statutory compromise and directs its regulatory energy toward preventing that compromise from being exploited through aggressive interpretations of “in-office.”
Seen this way, the rulemaking is not about narrowing what physicians are allowed to do; it is about narrowing what Medicare is willing to pay a profit margin for when services drift too far from the clinical office that justified their ancillary status in the first place.
###
Could these discussions accomplish very much without actual changes to actual regulations? Isn’t everything else just dicta?
###
This is an excellent and very lawyerly question, and the short answer is: yes, without regulatory text changes much of this is dicta—but not meaningless dicta. In CMS world, there is a crucial middle ground between “binding regulation” and “mere commentary,” and this rulemaking sits squarely in it.
Let me unpack that carefully.
At a formal level, you are absolutely right. Only changes to the Code of Federal Regulations are legally binding rules of general applicability. Preamble discussion, examples, policy explanations, and responses to comments do not, by themselves, amend Stark, redefine IOAS, or prohibit conduct. If CMS had done only narrative discussion without changing §414.50, §424.80, or related definitions, the entire exercise would largely have been interpretive commentary.
But that is not what happened.
What CMS did here is more subtle and more powerful than pure dicta.
First, CMS did make concrete regulatory changes, but they were targeted at payment rather than permission. The revisions to §414.50 (anti-markup), the cross-reference in §424.80, and the revised understanding of what counts as the “office of the billing physician or supplier” are real regulatory hooks. They directly affect what Medicare will pay, and under what circumstances. Once payment is constrained, behavior follows, even if the underlying Stark exception technically remains intact.
Second, the preamble discussion is not free-floating. It is doing heavy interpretive work in service of those regulatory changes. Courts, auditors, ALJs, DOJ, and CMS contractors routinely treat such preamble language as authoritative evidence of agency intent, especially where the regulatory text uses open-ended or functional terms like “office,” “performed,” or “ordered.” In other words, this is not dicta in the way a law review aside is dicta; it is interpretive guidance tethered to enacted regulatory text.
Third—and this is key—CMS is explicitly laying groundwork for future enforcement and future rulemaking. When CMS says, repeatedly, that centralized buildings may satisfy Stark but do not necessarily qualify as offices for payment purposes, it is creating a record. That record matters later when CMS, an OIG auditor, or DOJ argues that a particular arrangement was aggressive, abusive, or knowingly exploitative. The regulated party cannot plausibly say, “we had no notice that CMS viewed this as problematic,” even if the conduct was not per se prohibited.
Fourth, CMS is operating within a regulatory ecosystem where payment denial is often more consequential than formal illegality. Medicare does not need to declare an arrangement unlawful to make it economically nonviable. By limiting billing to the lowest of cost, charge, or fee schedule when services are performed outside the “real” office, CMS effectively neutralizes the profit motive that made pod labs attractive. That accomplishes most of the policy goal without rewriting Stark.
Fifth, CMS is consciously avoiding overreach. Redefining “ancillary services” or rewriting the IOAS exception would have been a frontal assault on statutory language and decades of reliance interests. That would have invited litigation and congressional backlash. By contrast, adjusting payment rules while leaving the statutory exception untouched is administratively defensible and politically survivable. The preamble discussion explains and justifies that strategy, which makes it harder to challenge later as arbitrary.
So is much of the discussion technically dicta? Yes, in the narrow sense that it does not itself change the law.
But is it inert? No.
It functions as:
-
authoritative interpretation of ambiguous regulatory terms,
-
notice to the regulated community,
-
guidance to auditors and enforcement staff,
-
justification for payment denials,
-
and a predicate for later rulemaking.
In CMS policy history, this is actually a classic move. CMS often governs by narrowing the economic benefit of compliance rather than the formal scope of permission, and it uses preamble discussion to explain how it will exercise that discretion.
So your instinct is right, but the conclusion is slightly more nuanced:
without regulatory changes, discussion would be mere commentary—but here, the discussion is tightly coupled to payment rules that do change, and that makes it consequential in practice even if Stark itself remains textually untouched.
###
Yes — and you are exactly right to focus on that phrase “or more”, because it quietly does an enormous amount of policy work. What you are pointing to is not an aside to the pod-lab story but one of its most effective enforcement mechanisms, even though it sits in a different part of the payment universe.
What CMS did with prostate biopsy pathology is a classic example of convergent policy control: one lever aimed at who may profit from pathology (anti-markup), and another aimed at how much volume can ever be monetized, regardless of who bills.
For years before 2007–2008, prostate biopsy pathology had a built-in volume escalator. The CPT structure allowed separate payment for each core, and as biopsy protocols evolved from sextant sampling to 10, 12, or more cores, payment rose linearly. From CMS’s perspective, this created a perverse alignment: expanding clinical protocols translated directly into expanding revenue, with no natural brake. Even if the increase in cores was clinically defensible in some cases, the payment system itself had no way to distinguish clinical evolution from financial opportunism.
CMS’s response was not to challenge the medical science head-on. Instead, it imposed a payment ceiling. Once the number of cores reached a defined threshold — commonly remembered as “10 or more” — additional cores ceased to generate additional Medicare payment. The brilliance of the approach lies in its bluntness. CMS did not say “10 is appropriate” or “12 is excessive.” It said, in effect: whatever number you choose clinically, Medicare will not pay more once you cross this line.
That “or more” language is doing the real work. It collapses all higher volumes into a single reimbursement category. Whether a practice submits 10 cores or 18, the marginal revenue beyond the cap is zero. The incentive to keep adding cores for payment reasons evaporates, even though the physician’s clinical discretion remains intact.
When you place this alongside the anti-markup rules discussed earlier, a coherent policy strategy emerges. CMS never needed to prove that pod labs caused overutilization in every case. Instead, it removed the economic fuel that made rising biopsy counts and vertically integrated pathology attractive. The prostate biopsy cap ensures that even if a urology group controls pathology billing, there is no financial upside to chasing ever-larger specimen counts. The anti-markup rules ensure that, if the work is done outside the true clinical office, the group cannot layer profit on top of someone else’s labor.
Importantly, this payment cap operates independently of Stark and IOAS. It does not matter whether the pathology is technically “in-office,” performed in a centralized building, or sent to an independent lab. The cap applies universally. That universality is what gives it legitimacy and durability. CMS avoided specialty-specific prohibitions while still targeting a specialty-specific incentive structure.
This also explains why CMS could afford to leave much of the IOAS debate in narrative form. When payment is capped and markup is constrained, formal compliance becomes far less valuable. A pod lab that once generated growing margins from 12-core, 14-core, or 18-core biopsies suddenly looks much less compelling when reimbursement flattens at “10 or more” and any off-site work is reimbursed at cost. The arrangement may still exist, but its economic rationale is largely neutralized.
In retrospect, this was a more effective intervention than rewriting Stark would have been. Redefining “ancillary services” or narrowing IOAS would have triggered legal challenges and political backlash. By contrast, capping prostate biopsy pathology payment and tightening anti-markup rules allowed CMS to discipline behavior without accusing anyone of wrongdoing. Practices could continue to argue about quality, turnaround time, and collaboration — but the payment system would no longer reward unbounded growth.
So yes: urology, prostate cancer, and biopsy pathology were central to this era, and the payment cap is inseparable from the pod-lab story. It is a reminder that in Medicare policy, the most consequential reforms often arrive not as headline prohibitions, but as small phrases in payment rules — phrases like “or more” that quietly redraw the economic map.
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.