1. Introduction
United States v. Tekippe addresses how sentencing courts may calculate loss when health care fraud permeates a provider’s billing practices. The central question was whether the record supported a finding of pervasive fraud sufficient to shift to the defendant the burden of identifying legitimate amounts within the claims submitted to an insurer.
Benjamin Tekippe, a chiropractor who operated Oak Tree Chiropractic LLC under the name Metairie Chiropractic & Rehab, submitted reimbursement claims to Blue Cross and Blue Shield of Louisiana (“BCBSLA”). After a seven-day trial, a jury convicted him of six counts of health care fraud and one count of wire fraud. It acquitted him of making a false statement to federal agents.
Tekippe appealed his sentence, not his convictions. He challenged the finding of pervasive health care fraud underlying the loss calculation, relying particularly on declarations from twenty-five patients confirming that he had treated them. The Fifth Circuit affirmed.
The decision applies an established burden-shifting rule rather than creating a new one. Its principal contribution is to explain why evidence that patients received some treatment does not necessarily establish that the corresponding insurance claims were legitimate.
2. Background and Sentencing History
The billing practices and trial evidence
During approximately three and a half years, at least 93% of Tekippe’s BCBSLA claims included the same six billing codes, covering five categories of services: mechanical traction, electrical stimulation, therapeutic exercises, massage therapy, and chiropractic manipulation. His massage-therapy billing exceeded that of every other chiropractor in Louisiana.
BCBSLA required covered services to be medically necessary, performed or supervised by an appropriately licensed provider, and supported by treatment documentation. The government’s evidence indicated substantial departures from those requirements:
- Massage therapists provided whatever massages clients requested, typically full-body massages, without corresponding clinical direction from a chiropractor.
- Two substitute chiropractors testified that they did not perform, supervise, or certify patients’ need for several billed services.
- Employees described unsupervised treatment, inaccurate billing, and instructions to fabricate treatment notes.
- Investigators identified substantially identical records across patients and treatment dates.
- Tekippe requested backdated treatment notes and offered $2 per note; two massage therapists refused.
- Employees who were also patients testified that claims attributed services to them that they had not received.
Patient Logan Camus supplied a particularly concrete comparison: he recalled receiving two or three therapeutic exercises and one or two chiropractic adjustments, while the practice billed fifty-five exercises and fifty-three adjustments.
Tekippe denied fraudulent intent and attributed irregularities to ignorance, oversight, and employee mistakes. Another employee testified that she had neither falsified records nor observed others doing so, and that she had seen chiropractors perform adjustments.
The competing loss calculations
Tekippe submitted more than $2.3 million in claims to BCBSLA, which paid $740,842.36. The presentence investigation report (“PSR”) initially treated the billed amount as intended loss. Adding $12,952 in fraudulent unemployment-insurance payments produced a stated intended loss of $2,312,952 and actual loss of $753,794.36.
The PSR calculated an offense level of 33 and an advisory imprisonment range of 151–188 months. Tekippe argued that intended loss should reflect the reimbursement he expected, rather than the full amounts billed, and that legitimate services should be excluded.
His sentencing memorandum included twenty-five patient declarations. He attributed more than $450,000 in billed services and $158,327.14 in payments to those patients, proposing a BCBSLA loss figure of $582,515.12. The government responded that the declarations did not establish compliant claims and that several declarants’ records themselves showed indications of fraud.
The district court accepted Tekippe’s argument that intended health care loss should correspond to the amount BCBSLA actually paid. It nevertheless declined to deduct further amounts for allegedly legitimate services. Including the unemployment-insurance payments, it adopted a total loss of $753,794.36, an offense level of 31, and an advisory range of 121–151 months.
The court imposed 84 months’ imprisonment and three years of supervised release. It explained that a downward variance was appropriate because the loss-driven guideline sentence was disproportionate to the offense’s nature and circumstances.
3. Summary of the Opinion
The Fifth Circuit held that the finding of pervasive fraud was plausible on the record as a whole and therefore survived clear-error review. It relied on the combined force of standardized billing, unusually high massage claims, testimony about unperformed services, deficient supervision, and widespread record fabrication.
Under the governing sentencing rule, the government ordinarily must establish that claimed loss resulted from fraud. When it demonstrates fraud so extensive and pervasive that separating legitimate benefits from fraudulent ones is not reasonably practicable, the defendant must identify particular legitimate amounts.
Tekippe’s patient declarations did not compel a reduction. They could establish that some treatment occurred without establishing that the particular claims satisfied BCBSLA’s reimbursement conditions. Moreover, even assuming all declarant-supported claims were legitimate, they represented less than one-quarter of the amounts billed and paid and did not negate the broader evidence of pervasive misconduct.
The court also rejected the contention that the district court refused to consider the declarations because they were submitted after trial. The sentencing court had expressly addressed Tekippe’s proposed reduced figure and observed that it would not change the applicable loss-enhancement bracket.
4. Analysis
A. Precedents Cited and Their Influence
United States v. Dennis, 41 F.4th 732, 743 (5th Cir. 2022)
United States v. Dennis supplies the basic division between legal and factual appellate review. Guideline interpretation and application receive de novo review, while factual findings receive clear-error review.
That distinction framed the appeal. Tekippe’s sole appellate issue concerned the factual finding that fraud was pervasive, rather than a newly disputed interpretation of the loss guideline. Accordingly, the appellate court did not independently reweigh the evidence as though making the original sentencing determination.
United States v. Barnes, 979 F.3d 283, 310–12 (5th Cir. 2020)
United States v. Barnes served two purposes. First, it specifically supported clear-error review of a finding of pervasive health care fraud. Second, it provided an evidentiary analogy: unusually frequent use of a small group of diagnostic codes, compared with other providers, could support a finding of pervasiveness.
Tekippe’s near-uniform service bundles and exceptional massage billing similarly suggested a systematic practice rather than unrelated mistakes. The court did not treat statistical abnormality as conclusive by itself; it evaluated those patterns alongside direct testimony and evidence of fabricated documentation.
United States v. Puig-Infante, 19 F.3d 929, 942 (5th Cir. 1994)
United States v. Puig-Infante states that a factual finding survives clear-error review when it is “plausible in light of the record read as a whole.”
This formulation explains why evidence favorable to Tekippe did not require reversal. The existence of some actual treatment, or testimony denying observed falsification, did not make the district court’s contrary overall assessment implausible. The appellate question was whether the finding was supportable, not whether every item of evidence pointed in the same direction.
United States v. Hebron, 684 F.3d 554, 560, 563 (5th Cir. 2012)
United States v. Hebron supplied the opinion’s central substantive rule. It explains that guideline loss generally means the greater of actual loss and intended loss. More importantly here, it establishes the conditional shift in responsibility for identifying legitimate amounts:
Where fraud is so extensive and pervasive that separating legitimate benefits from fraudulent ones is not reasonably practicable, the defendant must make a showing that particular amounts are legitimate.
Without that showing, the sentencing court may reasonably treat the entire benefits claim as intended loss. Applied here, the rule supported declining further deductions from the BCBSLA payments after the district court had already reduced the health care loss calculation from gross billings to actual reimbursements.
The qualification matters: the government does not obtain a burden shift merely by proving some fraudulent claims. It must first establish the breadth of fraud and the practical difficulty of separating legitimate amounts.
United States v. Malone, 828 F.3d 331, 337 (5th Cir. 2016)
United States v. Malone recognizes that even hearsay can be sufficiently reliable to support a sentencing finding.
Tekippe argued that testimony about his records and practices lacked sufficient detail and corroboration. The citation explained why such criticisms did not necessarily disqualify the evidence at sentencing. The court relied on testimony from relevant witnesses, supported by the broader record. It did not hold that unreliable allegations must be accepted.
United States v. Mazkouri, 945 F.3d 293, 303–04 (5th Cir. 2019)
United States v. Mazkouri establishes that sentencing facts need only be proved by a preponderance of the evidence. It also recognizes that a PSR’s distillation of the facts may suffice to demonstrate pervasive fraud.
This precedent answered Tekippe’s suggestion that the government had not adequately proved fraud beyond the specific conviction counts. The sentencing determination did not require a separate jury verdict establishing every additional fraudulent claim beyond a reasonable doubt. The district court could assess the scheme-wide evidence under the sentencing standard.
United States v. Jones, 664 F.3d 966, 984 (5th Cir. 2011)
United States v. Jones rejected an argument that treatment had value to patients when the services had no value to the defrauded payer, Medicare.
Its relevance here was the distinction between patient benefit and insurer entitlement. A patient’s receipt of treatment did not establish that BCBSLA owed reimbursement for every billed service. Although the payer here was a private insurer rather than Medicare, the analogy supported examining compliance with the victim payer’s coverage rules rather than relying solely on patients’ appreciation or receipt of care.
B. Legal Reasoning
1. The court assessed a scheme, not isolated claim errors
The evidence reinforced itself across several dimensions. Standardized coding suggested routine bundled billing; patient accounts showed that particular billed services were not performed; employee testimony described unsupervised care; and copied or fabricated notes undermined the records used to substantiate reimbursement.
Together, these features supported a systemic explanation. The court was not merely extrapolating from an unusually large practice or a few documentation defects.
2. Loss estimation did not require perfect reconstruction
Section 2B1.1 permits a reasonable estimate of loss. Its commentary recognizes that the sentencing judge is particularly well positioned to assess the evidence and estimate the financial harm.
Requiring claim-by-claim reconstruction despite pervasive falsification could make sentencing loss depend on the very records the fraud rendered unreliable. The burden-shifting rule addresses that practical difficulty while retaining the government’s initial obligation to establish the necessary foundation.
3. Actual treatment did not establish legitimate reimbursement
The twenty-five declarations addressed whether Tekippe personally treated the patients. They did not necessarily establish what services were provided on each occasion, whether all billed services were medically necessary, whether required supervision occurred, or whether documentation complied with BCBSLA’s rules.
This distinction was decisive. A legitimate encounter could coexist with inflated or otherwise improper claims. The declarations also did not resolve the government’s specific concerns about indications of fraud in several declarants’ records.
The opinion should not be read as holding that every coverage or documentation violation independently establishes criminal fraud. Rather, in a scheme already supported by extensive fraud evidence, proof of treatment alone did not establish the legitimacy of disputed loss deductions.
4. The decision did not impose a trial-only deadline for defense evidence
Some district court remarks suggested that the defense’s burden should have been borne at trial. The Fifth Circuit nevertheless rejected Tekippe’s characterization that the sentencing declarations had been disregarded.
Its reasoning rested on the record showing that the district court considered the proposed reduction. The appellate decision therefore does not establish a categorical rule barring defendants from presenting evidence of legitimate claims at sentencing.
5. The proposed reduction would not change the loss bracket
The district court observed that Tekippe’s proposed figure of $582,515.12 still fell within the same fourteen-level loss-enhancement bracket under § 2B1.1(b)(1)(H). That observation reinforced the conclusion that the declarations had been considered.
It also limited the proposed deduction’s effect on the guideline calculation. Nevertheless, the Fifth Circuit addressed and upheld the pervasiveness finding rather than relying solely on the unchanged enhancement bracket.
5. Potential Impact
Claim-specific proof becomes essential after a justified burden shift
Defendants seeking deductions will generally need evidence connecting particular amounts to reimbursable services. General patient declarations may be insufficient without corresponding proof of the services performed, medical necessity, supervision, and compliance with payer requirements.
Billing patterns matter most when corroborated
The opinion supports combining statistical indicators with patient testimony, employee accounts, and documentation evidence. It does not authorize treating high billing volume or standardized coding alone as proof that an entire practice is fraudulent.
Appellate challenges face substantial factual deference
Once the district court makes a supported pervasiveness finding, an appellant must show more than competing evidence or possible innocent explanations. The finding must fail the whole-record plausibility standard.
The ruling remains narrower than a general presumption against providers
The decision does not presume all claims fraudulent whenever some fraud occurs. Its burden-shifting mechanism depends on extensive, pervasive misconduct that makes separation impracticable. Nor did the appeal resolve a general dispute over using gross charges rather than expected reimbursements: the district court had already accepted Tekippe’s reduction to the amounts BCBSLA actually paid.
6. Complex Concepts Simplified
- Pervasive fraud
- Fraud sufficiently widespread throughout a scheme that identifying legitimate transactions separately is not reasonably practicable.
- Burden-shifting
- After the government establishes the required pervasive-fraud foundation, the defendant must identify and substantiate particular amounts claimed to be legitimate.
- Actual loss and intended loss
- Actual loss concerns reasonably foreseeable financial harm that occurred; intended loss concerns harm the defendant intended to cause. The guideline generally uses the greater amount.
- Clear-error review
- A deferential appellate standard under which a factual finding stands if it is plausible when the whole record is considered.
- Preponderance of the evidence
- The sentencing standard requiring a fact to be more likely true than not, rather than proved beyond a reasonable doubt.
- SOAP notes
- Treatment records documenting subjective complaints, objective findings, an assessment, and a treatment plan.
- Downward variance
- A sentence below the advisory guideline range based on the court’s assessment of the appropriate punishment. Here, the court imposed 84 months despite a 121–151 month range.
7. Conclusion
United States v. Tekippe reinforces an established sentencing principle: once the government demonstrates pervasive fraud that makes legitimate and fraudulent amounts impracticable to separate, the defendant must substantiate proposed legitimate amounts.
Its central practical lesson is that proof of patient treatment is not necessarily proof of a legitimate insurance claim. Applying that distinction and deferential factual review, the Fifth Circuit upheld the loss determination and affirmed Tekippe’s sentence.