Arete Healthcare LLC
Independent Practices | Shared Infrastructure | Clinician-Led Governance
Independent Practices | Shared Infrastructure | Clinician-Led Governance
Arete Original Research
Current Federal Healthcare Program Exclusions Across Healthcare Specialties, with a Focus on Chiropractic Student-Loan Defaults and Historical Completion Trends
Three figures summarize the paper's central finding: the current OIG exclusion record for chiropractic is unusually large in raw volume, but more than half of that stock reflects a historically concentrated student-loan-default cohort. The final figure places that cohort beside national chiropractic completions to show timing, not causation.
Figure 1. Current OIG exclusions by specialty and reason. Bars show exclusions that remained in effect in the August 2026 LEIE snapshot and had effective dates from 1990 forward. Chiropractic has the largest raw count among the 14 mapped specialties, driven substantially by exclusions under Social Security Act section 1128(b)(14) for default on health education loan or scholarship obligations. Source: HHS-OIG LEIE via Mimilabs, August 2026 snapshot; Arete analytic workbook. Visualization was human generated.
Figure 2. Chiropractic exclusion mix by exclusion-effective-date period. Loan-default exclusions account for 83.4% of the 847 currently active chiropractic exclusions with 1990s effective dates, 41.2% of the 832 with 2000-2009 dates, 37.4% of the 356 with 2010-2019 dates, and 0.8% of the 133 with 2020-2025 dates. The chart is a current-stock view; it does not show all exclusions originally imposed in each period because reinstated individuals/entities are removed from the LEIE. Visualization was human generated. Note: the source chart image also includes an illustrative "2020 to 2029 (projected)" bar not discussed anywhere in this paper's text or methods; it is retained here because this is the Author's original chart, but it is flagged for Author review before release (see Disclosures and Version History).
Figure 3. Chiropractic student-loan-default exclusions and completions. The current LEIE contains a large cohort of still-active loan-default exclusions with 1990s and early-2000s effective dates; the annual count peaks at 188 in 1997. National chiropractic completions peak later at 3,809 in academic year ending 2000 and then contract sharply. The HEAL marker provides historical context only: Congress ended authority for new Health Education Assistance Loan originations after September 30, 1998, but the LEIE exclusion-type field does not identify the underlying loan program for each excluded person. Sources: HHS-OIG LEIE via Mimilabs; IPEDS Completions; Congressional Research Service, Student Loan Programs Authorized by the Public Health Service Act: An Overview. Visualization was human generated.
This Arete Original Research paper explores a couple of interesting findings in the previous The Regulatory Record (https://arete.healthcare/research/the-regulatory-record) and The Education ROI (https://arete.healthcare/research/the-education-roi) papers.
The Regulatory Record paper summarized high level National Practitioner Data Bank (NPDB) trends, of which Office of Inspector General (OIG) Exclusions were a category, however Exclusions details were not presented. The Education ROI paper revealed an abrupt reduction in chiropractic completions (graduations) in the early 2000s.
In this paper, chiropractic was found to make up 29% of total 1990-2026 OIG Exclusions for the 14 specialties analyzed. 54.3% of chiropractic exclusions were associated with student loan defaults, 81.5% of total student loan defaults during 1990-2026 for the 14 specialties.
79% of chiropractic student loan defaults occurred in 1990-2002, the period immediately before the abrupt 28.6% reduction in chiropractic completions (graduations) in 2003. While an interesting potential association, there are too many potential factors to imply cause and effect.
The current student loan debt and early career income trajectory identified in The Education ROI, coupled with potential changes in federal student loan availability, makes the association between student loan debt/income and the pipeline of future chiropractors one of the profession's top priorities.
Arete is all-in to do our part to help ensure the success of early, mid and late career chiropractors. For early career chiropractors, several Arete priorities are important. Our New Practitioner Readiness Curriculum is designed to provide foundational knowledge to help chiropractors avoid delays in revenue generation while student loan debt burden is highest. To derisk practice we provide a management services organization (MSO) platform that supports chiropractors with modern, interoperable technology at the lowest price possible. Through participation in the Arete clinically integrated network (CIN), early career chiropractors are able to efficiently enter contractual arrangements with insurers.
All exclusion figures in this table use the 1990-forward mapped comparator set in the supplied analytic workbook. PM&R is excluded because the LEIE free-text fields do not identify it reliably. The comparator range is descriptive, not risk adjusted.
| Domain | Chiropractic | Comparator context | Interpretation boundary |
|---|---|---|---|
| Current exclusions, 1990–Aug 2026 | 2,178 | 91–860 | Current-stock counts, not incidence; workforce size differs substantially. |
| Student-loan-default exclusions | 1,183 (54.3%) | 0–151 (0%–28.2%) | 1128(b)(14) reflects education-financing obligations, not clinical misconduct. |
| Share of all mapped loan-default exclusions | 81.5% | 0%–10.4% per comparator | Chiropractic dominates this exclusion basis in the current stock. |
| 1990s exclusion mix | 706 / 847 loan default (83.4%) | Varies by specialty | Historical exclusions survive in LEIE only while the exclusion remains in effect. |
| 2020–2025 exclusion mix | 1 / 133 loan default (0.8%) | Mostly 0% loan default | Recent chiropractic stock is instead dominated by fraud-felony, licensure, and program-related categories. |
| Completions: peak to early contraction | 3,809 (2000) → 2,718 (2003), −28.6% | Not compared | Ecological timing comparison; no person- or institution-level linkage. |
| Completions: peak to trough | 3,809 (2000) → 2,219 (2013), −41.7% | Not compared | Many market, institutional, and financing factors may contribute. |
| Recent DC/PT, 2023–2025 | 65 records; 4.3 /10,000 CMS provider-years | PT: 43 records; 1.1 /10,000 CMS provider-years | DC/PT 3.8x; neither profession had a loan-default exclusion. |
Background: Arete's first Regulatory Record report used the National Practitioner Data Bank (NPDB) and found that exclusions were a meaningful component of chiropractic regulatory reporting. This follow-up examines the HHS Office of Inspector General List of Excluded Individuals/Entities (LEIE) directly, with emphasis on why chiropractic exclusions occurred and when their effective dates clustered.
Methods: The primary numerator is the August 2026 LEIE snapshot extracted through Mimilabs. LEIE free-text general and specialty fields were mapped into 14 reliably identifiable healthcare specialties; PM&R was not mapped. Exclusions were grouped by Social Security Act exclusion authority and effective year. The principal comparison uses effective dates from 1990 through August 2026. A chiropractic sub-analysis compares student-loan-default exclusions under section 1128(b)(14) with national IPEDS chiropractic completions for academic years ending 1995-2025. A focused DC/PT sensitivity analysis normalizes current LEIE records with 2023-2025 effective dates to the pooled CMS provider-year denominators used in Part 1, with a separate 2025 BLS employed-worker sensitivity. The analysis is descriptive.
Results: Across the 14 mapped specialties, 7,515 current exclusions had effective dates from 1990 forward. Chiropractic accounted for 2,178 (29.0%), including 1,183 student-loan-default exclusions (54.3% of chiropractic exclusions and 81.5% of all loan-default exclusions in the comparator set). Of 847 current chiropractic exclusions with 1990s effective dates, 706 (83.4%) were loan-default related; only 1 of 133 exclusions with 2020-2025 effective dates was. Among current LEIE records with 2023-2025 effective dates, chiropractic had 65 versus 43 for physical therapy; normalized to pooled CMS provider-years, the corresponding rates were approximately 4.3 versus 1.1 per 10,000 (3.8x). Neither profession had a student-loan-default record in this recent three-year cohort. Chiropractic completions peaked at 3,809 in 2000, declined 28.6% to 2,718 by 2003 and 41.7% to a trough of 2,219 by 2013, then recovered to 3,022 in 2025.
Conclusion: The OIG record shows two distinct chiropractic regulatory eras. The current stock of older exclusions is heavily shaped by unresolved education-financing obligations, while recent exclusions are predominantly conduct- and licensure-related. The historical loan-default wave is highly relevant to current education-economics research because it demonstrates that financing distress has previously had durable regulatory consequences in chiropractic. It does not, however, establish that loan defaults or OIG exclusions caused the later decline in chiropractic completions; the LEIE is a survivor-stock dataset, and the completion comparison is ecological.
The first paper in this series, The Regulatory Record, used the National Practitioner Data Bank (NPDB) to characterize malpractice payments, licensure actions, exclusions, and other regulatory reports across healthcare fields. Chiropractic generated 14,585 NPDB reports during 2010-2025; 13.3% were exclusions, compared with 5.2% for physical therapy. That result established that federal exclusion activity was not peripheral to chiropractic's regulatory record, but the NPDB did not explain the underlying reasons for those exclusions in the same detail available from the HHS Office of Inspector General. [1]
This follow-up therefore changes the data source and the question. Instead of asking how often exclusions appear inside the broader NPDB reporting mix, it asks who remains excluded from federal healthcare programs in the OIG LEIE, why those exclusions were imposed, and how the mix differs by specialty and historical period. The distinction matters because an OIG exclusion can arise from very different pathways: program-related crime, patient abuse or neglect, healthcare fraud, controlled-substance conviction, license action, kickbacks, or default on a health education loan or scholarship obligation. [2]
For chiropractic, that last category changes the interpretation of the record. A large share of the current chiropractic exclusion stock is not conduct-based. It reflects section 1128(b)(14), under which an individual may remain excluded until the education-loan default or scholarship obligation is resolved. OIG's current guidance describes that minimum period as lasting until resolution. [2] That duration creates a very different survival pattern in a current-stock database than a five-year mandatory exclusion that later expires and is removed after reinstatement.
The financing history is also relevant to Arete's companion paper, The Education ROI. That study found chiropractic to be a present-day early-career financing-burden outlier: approximately $196,057 in evaluated federal debt versus $61,163 in reported four-year earnings, a 3.21x debt-to-earnings ratio; visible borrower-repayment data also showed limited positive amortization. Yet recent chiropractic graduations have been stable to rising, creating what that paper called a graduation-financing paradox. [3] The OIG data add a longer historical perspective: education-debt distress has previously been large enough to leave a regulatory footprint that remains visible decades later.
The primary objective is descriptive. This paper does not treat exclusion counts as a validated measure of clinical safety, does not infer individual default prevalence from the LEIE, and does not claim that the 1990s exclusion wave caused the subsequent contraction in chiropractic completions. Instead, it asks whether the timing and composition of the OIG record provide useful historical context for the profession's current education-financing debate.
The primary source was mimi_ws_1.hhsoig.leie, the Mimilabs representation of the HHS-OIG List of Excluded Individuals/Entities, using the latest monthly source snapshot available to the supplied analysis (August 2026). The LEIE download is a current-state file: it contains exclusions in effect, and individuals or entities that have been reinstated are removed from the full downloadable database. OIG explicitly describes the file as the complete database of exclusions currently in effect. [4]
This design has an important consequence: the analysis measures the stock of exclusions that remain active in 2026, grouped by original exclusion effective date. It is not a retrospective census of every exclusion imposed in 1990, 1997, 2005, or any other historical year. Older categories with long or unresolved exclusion periods are preferentially retained relative to categories whose exclusion periods ended and were followed by reinstatement.
The LEIE uses free-text general and specialty fields rather than a standardized taxonomy crosswalk. A priority-ordered CASE statement mapped records into 15 intended specialties: chiropractic, physical therapy, orthopedic surgery, pain medicine, physical medicine and rehabilitation (PM&R), family medicine, internal medicine, emergency medicine, nurse practitioner, physician assistant, optometry, clinical psychology, clinical social work, podiatry, and psychiatry. PM&R produced fewer than five reliably identifiable records and was removed from specialty comparisons, leaving 14 mapped specialties. Emergency medical technicians were excluded from the emergency-medicine physician mapping.
The mapping includes records coded under broad organizational labels when the specialty sub-field identifies a mapped profession. Because the general field is truncated and specialty text is inconsistent, some providers may remain unmapped or be misclassified. These limitations are more important for absolute cross-specialty comparisons than for the chiropractic sub-analysis, where the source strings are comparatively distinctive.
Exclusion type was grouped into seven named Social Security Act authorities plus an Other category: program-related crimes (1128(a)(1)); patient abuse/neglect (1128(a)(2)); healthcare fraud felony (1128(a)(3)); controlled-substance felony (1128(a)(4)); license revocation/suspension/surrender (1128(b)(4)); fraud/kickbacks and other prohibited activities (1128(b)(7)); and default on health education loan or scholarship obligations (1128(b)(14)). Remaining exclusion types were grouped as Other. [2]
The student-loan-default category is analytically separated because it reflects an education-financing obligation rather than a clinical or billing event. OIG states that the exclusion period under 1128(b)(14) continues until the default or obligation is resolved. [2] The LEIE field does not identify the specific loan or scholarship program underlying each record.
The primary cross-specialty table uses exclusions with effective dates from 1990 through August 2026, aligning the historical frame with the federal regulatory period examined in Part 1 and avoiding small pre-1990 counts. The supplied detail contains 26 additional chiropractic exclusions with effective dates before 1990; including them would raise the all-time current chiropractic stock from 2,178 to 2,204 but does not change the central findings.
Chiropractic exclusions were also summarized in four complete historical periods: 1990-1999, 2000-2009, 2010-2019, and 2020-2025. Partial-year 2026 was excluded from the final period. Annual chiropractic student-loan-default counts were compared visually with national chiropractic completions from the IPEDS Completions Survey for academic years ending 1995-2025. Completions are field-level totals and are not linked to the individuals in the LEIE.
To provide a workforce-adjusted contemporary comparison without applying a current denominator to decades of surviving exclusions, a focused DC/PT sub-analysis was limited to recent effective dates. The primary numerator counts current LEIE records in the August 2026 snapshot with exclusion effective dates from 2023-2025. Those counts were divided by the same pooled 2023-2025 CMS Medicare-enrolled provider-year denominators used in Part 1: approximately 150,024 DC provider-years and 377,727 PT provider-years, derived from annualized counts of 50,008 and 125,909, respectively. Results are expressed per 10,000 provider-years. [1]
This construction materially narrows the numerator/denominator time mismatch but does not turn the LEIE into an incidence dataset: reinstated providers are absent from the August 2026 snapshot. The resulting values are therefore described as normalized current-exclusion records, not incidence rates. The recent-period numerator is also analytically useful because neither DC nor PT has a student-loan-default record with an effective date in 2023-2025.
A separate 2025 sensitivity analysis divides current LEIE records with 2025 effective dates by May 2025 BLS Occupational Employment and Wage Statistics employed-worker counts: 39,630 chiropractors and 267,330 physical therapists. [8] OEWS excludes self-employed workers; that limitation is particularly consequential for chiropractic because The Education ROI documents a substantially higher self-employment share among chiropractors than physical therapists. [3] The BLS result is therefore treated as directional sensitivity rather than the preferred denominator.
No record-level individual/entity filter was rerun for this focused sub-analysis. The supplied LEIE mapping can include records carried under broad organizational general categories when the specialty field identifies a mapped profession. Because the CMS and BLS denominators count people rather than organizations, the normalized results should not be interpreted as unique-practitioner incidence estimates.
The figure marking September 30, 1998 uses Congressional Research Service documentation that the Health Education Assistance Loan (HEAL) program made loans through fiscal year 1998 and that Congress terminated authority for new HEAL loans after September 30, 1998. [5] This is contextual only. Section 1128(b)(14) covers health education loan or scholarship obligations broadly, and the LEIE extract does not identify which loan program generated a particular exclusion.
No causal or inferential model was used for the exclusion-completion comparison. The analysis reports counts, shares, ratios, period mix, peak and trough values, and simple percent changes. This decision is deliberate: the LEIE is a survivor-stock dataset with reason-specific exclusion-duration bias, while the completions series is an aggregate educational outcome. A time-series regression would create a false impression of precision without resolving the fundamental measurement mismatch.
Across the 14 mapped specialties, 7,515 current exclusions had effective dates from 1990 through August 2026. Chiropractic accounted for 2,178, or 29.0% of the mapped total. Internal medicine (860) and family medicine (852) were next, followed by clinical psychology (650), podiatry (536), psychiatry (535), clinical social work (468), physician assistant (390), and physical therapy (286). The remaining mapped specialties each had fewer than 250 current exclusions.
Raw volume should not be interpreted as a comparative incidence or risk rate. Specialty workforce sizes differ substantially, and the current-stock numerator spans decades while any available workforce denominator would be point-in-time. The supplied Mimilabs methods included an exploratory all-time Medicare-provider normalization, but this paper does not use that all-time rate as a headline measure because it compounds the all-time-numerator/current-denominator mismatch and because Medicare enrollment captures different fractions of different professions. Instead, section 3.5 limits denominator adjustment to the recent 2023-2025 DC/PT cohort.
Of the 2,178 chiropractic exclusions in the 1990-forward current stock, 1,183 (54.3%) were for student-loan default. Chiropractic alone accounted for 81.5% of all 1,452 student-loan-default exclusions across the 14 mapped specialties. The next-highest counts were podiatry at 151 (28.2% of its own 536 exclusions), clinical psychology at 48 (7.4%), optometry at 26 (26.0%), family medicine at 18 (2.1%), and internal medicine at 15 (1.7%). Physical therapy had zero.
Removing student-loan-default exclusions materially narrows, but does not eliminate, chiropractic's raw-volume difference. Chiropractic falls from 2,178 total current exclusions to 995 non-loan exclusions. Physical therapy remains at 286 because it has no student-loan-default records in the supplied extract; the raw DC/PT ratio therefore falls from 7.62x for all exclusions to 3.48x after removing loan-default exclusions. Those ratios remain unadjusted for workforce size and should not be read as comparative risk.
The period mix makes the cohort structure visible. Among the 847 current chiropractic exclusions with effective dates in 1990-1999, 706 (83.4%) were loan-default exclusions. Among 832 exclusions dated 2000-2009, 343 (41.2%) were loan-default related. The share was 133 of 356 (37.4%) for 2010-2019, then fell to 1 of 133 (0.8%) for 2020-2025.
Three quarters of the 1990-forward chiropractic loan-default stock was concentrated in a nine-year span: 892 exclusions had effective dates from 1993 through 2001, representing 75.4% of all 1,183 chiropractic loan-default exclusions. The annual count rose from 46 in 1993 to 107 in 1994, reached 188 in 1997, and remained elevated at 118 in 1998, 87 in 2000, and 102 in 2001 before falling to 40 in 2002 and much lower counts thereafter.
The persistence of these old records is itself a finding. The median effective year of the 1,183 current chiropractic loan-default exclusions is 1998; 59.7% have effective dates in the 1990s, 75.7% by 2001, and 88.7% by 2009. Because 1128(b)(14) exclusions can continue until the default or obligation is resolved, the 2026 LEIE retains a substantial stock of decades-old unresolved exclusions that would be invisible in a recent-incidence view.
The 2020-2025 chiropractic stock no longer resembles the 1990s pattern. Of 133 exclusions with effective dates in this period, 43 (32.3%) were healthcare-fraud felonies, 38 (28.6%) license revocation/suspension/surrender, 26 (19.5%) program-related crimes, 13 (9.8%) patient abuse/neglect, 7 (5.3%) fraud/kickbacks, 3 (2.3%) controlled-substance felonies, 2 (1.5%) other reasons, and only 1 (0.8%) student-loan default.
Physical therapy shows a different recent mix but the same absence of loan-default exclusions: among 73 current PT exclusions with 2020-2025 effective dates, 32 (43.8%) were program-related crimes, 18 (24.7%) license actions, 14 (19.2%) healthcare-fraud felonies, 7 (9.6%) patient abuse/neglect, and none were student-loan default (the remaining 2 records were controlled-substance felonies and fraud/kickbacks). This reinforces the central interpretation that the historical student-debt cohort is a chiropractic-specific feature of the current LEIE stock rather than a generic therapy-profession pattern.
Among current LEIE records with exclusion effective dates in 2023-2025, the workbook contains 65 chiropractic records and 43 physical therapy records. Neither profession had a student-loan-default exclusion in this three-year period. Using the pooled CMS denominators from Part 1, these counts correspond to approximately 4.3 current exclusion records per 10,000 DC provider-years and 1.1 per 10,000 PT provider-years, a DC/PT ratio of 3.8x. The contemporary difference therefore persists after workforce adjustment and is not attributable to the historical chiropractic loan-default cohort.
In the 2025-only sensitivity view, 15 DC and 10 PT current records had 2025 effective dates. Dividing by BLS OEWS employed-worker counts yields approximately 3.8 records per 10,000 chiropractors and 0.4 per 10,000 physical therapists, a 10.1x DC/PT ratio. The much larger BLS ratio should not be interpreted as the more precise estimate: OEWS excludes self-employed workers and therefore omits an especially important segment of the chiropractic workforce. The consistent direction across CMS and BLS is more informative than the difference in magnitude.
Table 2. Recent DC/PT OIG Exclusions Normalized to Workforce Denominators.
| Measure | Chiropractic | Physical Therapy | Interpretation boundary |
|---|---|---|---|
| Current LEIE records, effective 2023–2025 | 65 | 43 | August 2026 survivor stock; neither profession had a student-loan-default record in this period. |
| CMS provider-years, 2023–2025 | ~150,024 | ~377,727 | Pooled Part 1 denominator; Medicare-enrolled providers, not active licenses. |
| CMS-normalized records /10,000 provider-years | 4.3 | 1.1 | DC/PT ratio 3.8x; primary recent-period sensitivity, not incidence. |
| Current LEIE records, effective 2025 | 15 | 10 | Single-year numerator for BLS sensitivity. |
| BLS employed workers, 2025 | 39,630 | 267,330 | OEWS employment excludes self-employed workers. |
| BLS-normalized records /10,000 employed workers | 3.8 | 0.4 | DC/PT ratio 10.1x; sensitivity only, especially limited for chiropractic. |
National chiropractic completions increased through the late 1990s even as the current LEIE records show a rising loan-default exclusion cohort. Completions were 2,968 in 1995, 3,654 in 1997, 3,735 in 1998, and reached 3,809 in 2000. They then fell to 3,284 in 2002 and 2,718 in 2003, a 28.6% decline from the 2000 peak. By 2005, completions were 2,560 (-32.8% from 2000); the series reached a trough of 2,219 in 2013 (-41.7%).
The timing is notable but not sufficient for causal attribution. The loan-default exclusion series is not an incidence series; it is the subset of historical exclusions that survived into the August 2026 LEIE. The completion series measures graduates, not applicants or enrollments, and it is not linked to excluded borrowers or to individual institutions. A plausible interpretation is that the exclusion record is a historical signal of financing stress in a profession whose educational pipeline later contracted. A stronger claim — that defaults caused the contraction — is not supported by these data.
The historical record is especially salient because The Education ROI identifies substantial current chiropractic financing burden despite recent completion growth. That paper reports mean evaluated-institution debt of approximately $196,057 and four-year post-graduation earnings of $61,163, a 3.21x debt-to-earnings ratio. It also estimates, from privacy-coarsened College Scorecard borrower data, a 74% forbearance midpoint and only 13% positive-amortization midpoint for chiropractic, while explicitly cautioning that COVID-era administrative forbearance limits interpretation. [3]
The two datasets should not be treated as a longitudinal continuation of the same borrowers. The OIG cohort largely reflects older education-financing obligations, while the Scorecard measures contemporary graduates under different federal loan structures. Their relevance lies in triangulation: current debt-to-earnings pressure is not merely an abstract ratio; chiropractic has a documented historical cohort in which unresolved education-financing obligations became federal-program exclusion events with multi-decade persistence.
The central finding is not simply that chiropractic has many OIG exclusions. It is that the meaning of those exclusions changes dramatically by era. The 1990s current stock is dominated by education-financing default; the 2020s stock is dominated by healthcare-fraud felonies, licensure actions, program-related crimes, and other conduct-based categories. Combining the two periods into one undifferentiated exclusion rate obscures both stories.
This distinction also changes how this paper should be read alongside Part 1. The NPDB paper correctly treated a field's regulatory reporting position as one input into its administrative risk profile, not evidence of clinical safety. [1] The OIG analysis sharpens that boundary: even within the same broad label of “exclusion,” the underlying mechanism can range from patient abuse to unresolved educational debt. A profession can therefore look like a regulatory outlier for reasons that are partly unrelated to care delivery.
The recent-period normalization addresses a different question from the historical stock analysis. Raw 2023-2025 current-record counts are only modestly different — 65 for DC and 43 for PT — but the CMS workforce denominator is substantially smaller for chiropractic. After normalization, the DC measure is approximately 3.8 times the PT measure. Because neither profession has a student-loan-default record in this recent cohort, the difference reflects the contemporary mix of licensure and conduct-related exclusion categories rather than the legacy education-financing cohort.
The denominator sensitivity also reinforces why the paper should not present one normalized value as an exact risk estimate. CMS and BLS define the workforce differently, and the 2025 BLS view produces a much larger DC/PT ratio because OEWS excludes self-employed workers — a particularly consequential limitation for chiropractic. The CMS result is therefore the primary recent-period comparison for continuity with Part 1; the BLS result is directional confirmation. Neither should be read as exposure-adjusted clinical risk or person-level exclusion incidence.
The most defensible historical statement is that a large cohort of chiropractic loan-default exclusions with 1990s and early-2000s effective dates remains unresolved enough to still appear in the August 2026 LEIE. That is directly observed. By contrast, the number of exclusions originally imposed in 1997 cannot be recovered from the current full LEIE because reinstated individuals are removed and OIG's public supplement archive is limited. [4]
This matters because student-loan-default exclusions have an open-ended resolution rule. A 1997 loan-default exclusion can remain in the current file decades later, whereas a five-year exclusion for another reason may have disappeared after reinstatement. The 83.4% loan-default share among current exclusions with 1990s effective dates therefore reflects both the original burden of student-loan-default exclusions and their greater probability of remaining visible today. The paper preserves that distinction prominently rather than reading the 83.4% figure as an original-imposition rate.
The completion timeline adds value because it shows that a major education-financing distress signal appeared before a large reduction in the number of chiropractic graduates. The sequence is consistent with several possible mechanisms: weaker demand after negative borrower outcomes became visible; tighter or changing financing availability; school closures or consolidation; shifts in applicant expectations; labor-market conditions; or broader changes in higher education. The supplied data cannot distinguish among them.
The HEAL program provides relevant historical context because new loan authority ended after fiscal year 1998. [5] But the timing should not be overread. The LEIE does not identify the loan program underlying each 1128(b)(14) exclusion, and the decline in completions could reflect numerous contemporaneous changes. The strongest language this paper uses is therefore “preceded” or “temporally preceded,” not “predicted,” “drove,” or “caused.”
The Education ROI paper concluded that present-day chiropractic is simultaneously a high-burden field and a graduation-growth exception: recent completions have increased despite unfavorable early-career debt/earnings. [3] The OIG history complicates that paradox in a useful way. It suggests that the profession has previously experienced a period in which education-financing distress was sufficiently severe to generate a large, durable federal exclusion cohort, followed by a substantial contraction in completions.
That does not mean the same sequence will recur. Current loan programs, institutional structures, earnings, borrower protections, and labor markets differ. But it does strengthen the case for monitoring leading indicators of graduate financial stress — debt burden, repayment status, default, enrollment, and institution-level completion trends — rather than relying on completions alone as the measure of program demand or sustainability.
Current-stock survivor bias. The LEIE full database contains exclusions in effect and removes reinstated individuals/entities. Historical effective-date counts therefore are not historical incidence counts.
Reason-specific duration bias. Section 1128(b)(14) exclusions can remain in effect until an education-loan default or scholarship obligation is resolved. Other exclusion authorities often have finite minimum periods. The current file therefore preferentially retains unresolved loan-default cases.
Free-text specialty mapping. The LEIE general and specialty fields are truncated and inconsistently populated. Fourteen specialties were mapped; PM&R could not be identified reliably. Misclassification and undercounting remain possible.
Historical stock cannot be normalized cleanly. The main numerator spans decades. Dividing the full surviving stock by a current Medicare, NPI, or BLS workforce count would create a prevalence-like ratio with a mismatched time base. The paper therefore limits denominator adjustment to the recent 2023-2025 DC/PT sub-analysis; that sensitivity does not solve the historical denominator problem.
Denominator definitions differ. CMS counts Medicare-enrolled provider-years, while BLS OEWS estimates employed wage-and-salary workers and excludes the self-employed. The two normalized views are sensitivity analyses and should not be interpreted as interchangeable estimates.
Numerator and denominator units are not perfectly aligned. The supplied LEIE specialty mapping can include some records carried under organizational general categories when specialty identifies a mapped profession. Because a record-level individual/entity filter was not rerun, the normalized DC/PT measures are not unique-practitioner incidence rates.
Ecological completion comparison. IPEDS completions are national field-level counts. They are not linked to excluded individuals, loan histories, schools attended by excluded borrowers, applicants, or enrollment cohorts.
Loan-program identity is unavailable. The exclusion type identifies default on a health education loan or scholarship obligation but does not identify the specific program. HEAL is historical context, not a record-level classification in the supplied LEIE extract.
Confounding and structural change. The 2000s completion contraction could reflect financing policy, school entry/exit, demographic demand, labor-market expectations, reimbursement, institutional strategy, or other unmeasured factors.
Cross-paper cohort mismatch. The historical OIG cohort and contemporary College Scorecard borrowers in The Education ROI are not the same people or financing regime. The comparison is triangulation, not longitudinal follow-up.
Partial-year 2026. Main cross-specialty counts include the August 2026 snapshot; period comparisons end at 2025 so 2026 is not treated as a complete year.
This work is classified as a public-data workforce and health-services analysis under the Arete Research Library Governance Framework (§2); it is not PBRN-sponsored human-subjects research. This analysis was produced by Arete Healthcare without external funding; Arete supports independent outpatient chiropractic and physical therapy practices and has a commercial and professional interest in workforce infrastructure, interoperability, and payer contracting. AI tools assisted with data review, calculation, drafting, and figure generation. Every factual assertion and calculation requires responsible human verification before publication, and AI reconciliation does not substitute for the named Independent Data/Methodology Reviewer or DAWG approval.
Independent verification (Fresh-Rebuild Process, Governance Framework §9): every headline statistic in this document was independently recomputed from the supplied source workbook (OIGDetail, Completions, and BLS sheets) during this rebuild, rather than carried forward from the prior draft on authority. All recomputed figures matched the prior draft's reported values exactly; no discrepancies were found, and no reconciliation appendix is included for this version as a result. This clean-verification outcome is itself recorded here rather than left unstated, consistent with this project's disclosure standard.
Report Number / Dataset Number: ARETE-OR-2026-011 / ARETE-DATA-2026-011, confirmed and permanent per Arete_00_Suite_Index.
Figure 2 note: the original chart image used for Figure 2 includes an illustrative “2020 to 2029 (projected)” bar that is not discussed in this paper's Methods, Results, or the Figure 2 caption. It is retained unedited as the Author's original chart pending an Author decision on whether to crop it, add supporting text, or leave it as-is; see the internal AI Assistance and Reproducibility Record maintained under the Arete Research Library Governance Framework.
Arete's Take (Advocacy/Commentary, Governance Framework §1.3): authored directly by the Author, carried into this document verbatim rather than originated by AI. It has received approval from at least two of three Arete Executive Reviewers, independent of and in addition to DAWG's approval of the Findings sections above.
Recommended citation: Arete Healthcare. The Regulatory Record — Part 2 — OIG Exclusions. Arete Original Research ARETE-OR-2026-011, Version 2.1. Approved for Release, September 2026. Published September 2026. https://doi.org/10.13140/RG.2.2.12995.77600
1. Arete Healthcare. The Regulatory Record: National Practitioner Data Bank Volume, Mix, and Severity Across Healthcare Fields of Licensure, 2010-2025. ARETE-OR-2026-006, v3.4, September 2026.
2. U.S. Department of Health and Human Services, Office of Inspector General. Background Information & Exclusion Authorities. https://oig.hhs.gov/exclusions/background-information-exclusion-authorities/
3. Arete Healthcare. The Education ROI: Cost, Debt, Earnings, and Graduation Trends Across Postsecondary Credentials — With Focused Analyses of Chiropractic and Physical Therapy. v1.11, September 2026.
4. U.S. Department of Health and Human Services, Office of Inspector General. LEIE Database & Supplement Downloads. https://oig.hhs.gov/exclusions/leie-database-supplement-downloads/
5. Congressional Research Service. Student Loan Programs Authorized by the Public Health Service Act: An Overview. R46720. https://www.congress.gov/crs_external_products/R/PDF/R46720/R46720.3.pdf
6. National Center for Education Statistics. Integrated Postsecondary Education Data System (IPEDS), Completions Survey. https://nces.ed.gov/ipeds/
7. Mimilabs. HHS-OIG LEIE and CMS provider data tables used in supplied reproducibility SQL; source vintage August 2026 for LEIE.
8. U.S. Bureau of Labor Statistics. Occupational Employment and Wage Statistics, May 2025 national cross-industry estimates; chiropractor and physical therapist employment counts as supplied in the analytic workbook.
Terms and abbreviations used in this paper.
| Term | Definition |
|---|---|
| LEIE | List of Excluded Individuals/Entities maintained by HHS-OIG. The full downloadable database contains exclusions currently in effect; reinstated individuals/entities are removed. |
| OIG | Office of Inspector General, U.S. Department of Health and Human Services. |
| NPDB | National Practitioner Data Bank. Part 1 of The Regulatory Record analyzed its broader adverse-action and malpractice-reporting data. |
| 1128(b)(14) | Permissive exclusion authority for default on health education loan or scholarship obligations. |
| HEAL | Health Education Assistance Loan program. New HEAL loan authority ended after September 30, 1998; the LEIE does not identify HEAL at the individual record level. |
| IPEDS | Integrated Postsecondary Education Data System, used here for national chiropractic completion counts. |
| Current-stock measure | A cross-sectional count of exclusions still in effect at the snapshot date, not a count of all exclusions originally imposed in a historical year. |
Counts use exclusions with effective dates from 1990 through August 2026 and reflect the current August 2026 LEIE stock. They are not historical incidence counts.
| Specialty | Current exclusions | Student-loan default | Loan-default share |
|---|---|---|---|
| Chiropractic | 2,178 | 1,183 | 54.3% |
| Internal Medicine | 860 | 15 | 1.7% |
| Family Medicine | 852 | 18 | 2.1% |
| Clinical Psychologist | 650 | 48 | 7.4% |
| Podiatry | 536 | 151 | 28.2% |
| Psychiatry | 535 | 4 | 0.7% |
| Clinical Social Worker | 468 | 2 | 0.4% |
| Physician Assistant | 390 | 2 | 0.5% |
| Physical Therapy | 286 | 0 | 0.0% |
| Nurse Practitioner | 244 | 0 | 0.0% |
| Pain Medicine | 199 | 1 | 0.5% |
| Emergency Medicine | 126 | 1 | 0.8% |
| Optometry | 100 | 26 | 26.0% |
| Orthopedic Surgery | 91 | 1 | 1.1% |
| Total | 7,515 | 1,452 | 19.3% |
| Period | Total | Loan default | License | Program crime | HC fraud felony | Controlled subst. | Abuse/neglect | Other* |
|---|---|---|---|---|---|---|---|---|
| 1990–1999 | 847 | 706 (83.4%) | 40 | 35 | 1 | 0 | 5 | 60 |
| 2000–2009 | 832 | 343 (41.2%) | 180 | 45 | 101 | 15 | 26 | 122 |
| 2010–2019 | 356 | 133 (37.4%) | 75 | 43 | 59 | 7 | 25 | 14 |
| 2020–2025 | 133 | 1 (0.8%) | 38 | 26 | 43 | 3 | 13 | 9 |
*Other combines fraud/kickbacks and remaining exclusion reasons for compact display. 2026 is excluded from the final period. Source: supplied Pivots sheet derived from LEIE SQL; independently recomputed from OIGDetail during this rebuild.