The Rise and Fall of San Diego Hospice — Part 3

False Bearing: The Claim, the Audit, and a Story That Outlived Both

In Part 2, I ended on a line I’d been carrying around for weeks: the $112 million number belongs to one story. The $10 million number belongs to another. They were never the same story.

Here is what I meant.

I trained at San Diego Hospice. I remember how that first number landed. A hundred and twelve million dollars. It came stamped with the authority of the United States government, and for a lot of people that ended the inquiry before it began. The number felt dispositive. It made the institution look as if it had been morally and financially beyond rescue.

Except nobody ever publicly showed the math.

That figure was not a judgment. It was not a final audit. It was not even a calculation anyone has explained in public. The number that was actually calculated, and released years later through FOIA, told a different story. Still ugly. Still serious. Still enough to demand accountability. But a different scale, and a different set of conclusions about what the board and executive leadership could have done while the institution was still alive.

That distinction matters. It changes the question from “how could this possibly have been saved?” to “who decided it was unsalvageable before the real exposure was known?”

The Sequence (Because Timing Is Substance)

Keep the dates straight. The order is part of the argument.

November 2012: CEO Kathleen Pacurar goes to the San Diego Union-Tribune editorial board and discloses that SDH is under Medicare audit. She tells the newspaper the hospice may have to return "$60 million or more." She tells the SDH board of directors the number could be $50 million. Two senior executives resign the following day. Medicare had not communicated a final number. The audit was not complete.

February 4, 2013: SDH files Chapter 11 in the Southern District of California (Case No. 3:13-bk-01179). No government claim amount exists on the public record at this point.

February 12–13, 2013: SDH announces intent to cease operations and transfer patients. Scripps Health prepares to acquire the Hillcrest campus and absorb patients and staff.

February 22, 2013: The Rachac qui tam lawsuit is unsealed. Lori Rachac, a former SDH nurse, had filed the whistleblower complaint under the False Claims Act in December 2012. The suit alleges SDH systematically admitted patients who did not meet hospice eligibility criteria under an "open access" policy dating to 2005.

June 2013: The U.S. Department of Justice files a $112 million proof of claim in SDH's bankruptcy, citing alleged false claims from 2009–2010 and referencing the Rachac case. Local reporting noted that the filing provided no explanation of how the figure was derived. DOJ declined to comment. SDH had already stopped caring for patients four months earlier.

March 2016: After a FOIA request by inewsource and KPBS, the Medicare audit is released. The findings: overpayments exceeding $10 million across roughly two years, with an average length of stay approximately two years — about four times the six-month benchmark embedded in the hospice benefit. The final document was dated 2014, a year after SDH closed. Pacurar, reached by phone, told inewsource she had never seen the audit.

That's the frame. Claim after closure. Audit released three years after that. Two artifacts, produced for different purposes, under different legal standards, on different timelines. The public remembered one number. The other arrived too late to matter.

What a Proof of Claim Does

A proof of claim is a creditor staking its position in bankruptcy. It is how a creditor says, “this estate owes us money, and we intend to recover what we can.” It preserves standing. It protects priority. It gives the creditor a seat in the process.

It does not establish liability. It does not fix the final amount owed. It does not function like a judgment.

By the time DOJ filed $112 million in June 2013, San Diego Hospice was already gone as a clinical enterprise. No patients remained to protect. No operating model remained to repair. The institution had already been put on the path to extinction.

So the claim mattered in a very specific lane. It affected bargaining power inside a liquidation. It shaped creditor dynamics. It helped the government maximize its position relative to everyone else trying to recover from the estate.

The bankruptcy docket reinforces this. The case proceeded under Chapter 11 with a liquidating trust. The U.S. Trustee objected to elements of the proposed plan — notably releases and exculpations for insiders. The court was arbitrating process and creditor priority, not liability magnitudes. In the end, the government received $1 million from the estate's roughly $16 million in assets. A separate False Claims Act settlement of approximately $3.67 million was finalized in 2017.

None of those resolutions approached $112 million. The claim shaped creditor dynamics. It did not drive the pre-closure operating decisions described in Part 2. It couldn't have — it didn't exist yet.

What the Medicare Audit Showed — and Why It Still Stings

The audit documented overpayments exceeding $10 million across roughly two years. Average length of stay near two years. Those facts validate serious eligibility and documentation failures, and they point to a length-of-stay distribution at SDH that was high-risk by any standard. They are real. They are damning. They are not $112 million.

And they weren't new information in the sense that the underlying dynamics were new. Return to the policy spine this series has been tracing since Part 1:

Hospice eligibility hinges on certification of terminal illness with a life expectancy of six months or less if the disease runs its normal course. Poor documentation can convert clinically appropriate care into non-covered care in an audit file — a distinction that matters enormously when you're serving a census of 1,000 patients per day, many with dementia, a disease in which prognosticating death is notoriously unreliable.

Long-stay scrutiny wasn't new. SDH itself had appeared in Operation Restore Trust in 1997, where the OIG focused on stays exceeding 210 days and found improper payments. As Part 1 documented, the founding medical director said she successfully appealed that audit — a response that may have inoculated the institution against the belief that long-stay risk required a structural fix rather than a one-off defense.

MedPAC's work — which the Commission had been publishing since its March 2009 recommendation and modeled more concretely in its June 2013 report — kept highlighting the structural problem: flat per-diem payment creates disproportionate profitability for long stays. The cost of care follows a U-shaped curve, higher at admission and near death, lower in the middle. A flat daily rate pays the same regardless of where a patient sits on that curve. The longer the stay, the longer the low-cost middle — and the wider the margin.

That payment design made LOS a board-level governance responsibility at every hospice in the country. At SDH, where the census was large and the culture valued saying yes to patients, it was an existential one.

The audit numbers are bad. They are also the kind of bad that, had they been known in real time, could have been addressed through documentation remediation, targeted clinical review, and negotiated repayment — the tools other hospices have used when facing similar findings. The numbers represent a fixable problem. The $112 million claim represented something else entirely: a legal position in a proceeding that was already winding down a dead institution.

How the Larger Number Became the Story

Once a nine-figure number enters the public record with the imprimatur of the United States Department of Justice behind it, the narrative congeals. Speed beats nuance every time.

The $112 million claim was filed in June 2013 and covered immediately. The audit result didn't surface until March 2016, after a FOIA request that took three years to process. That's the information vacuum in which folklore forms: a huge, unexplained number at the front end; the smaller, harder, more instructive number years later, arriving after most people had stopped paying attention.

And there was a compounding factor. The numbers SDH's leadership put into the public sphere in November 2012 — $60 million to the Union-Tribune, $50 million to the board — primed the environment for a nine-figure claim to feel like confirmation rather than surprise. The institution disclosed before it had a confirmed number. The board received these estimates and, as far as the public record shows, neither demanded independent financial counsel nor challenged the disclosure approach. The vacuum that leadership created was filled first by its own estimates, then by DOJ's claim, and finally — far too late — by the audit itself.

Go back to Part 2's central argument: dead reckoning is what happens when you navigate without an external fix. The number problem is the financial manifestation of the same failure. SDH's leadership — CEO and board together — was operating on estimated position. The $50 million estimate. The $60 million estimate. Imprecise, unconfirmed, and the basis for irreversible decisions. When the actual position arrived years later, it turned out to be more than an order of magnitude lower than the claim that had defined the public narrative — and roughly a fifth of what the board had been told to expect.

The tragedy is not that the government filed an aggressive claim. Creditors in bankruptcy file aggressive claims. That is what the process exists for. The tragedy is that the number became the story — and the story became the reason nobody questioned whether the institution had to die.

Where I Might Be Wrong

A few things that would change my calculus:

• The $112 million claim may have been based on a methodology — extrapolation from sampled claims, treble damages under the False Claims Act, or a combination — that, if disclosed, would make the number more defensible than it currently appears. DOJ has never explained the calculation. Without that explanation, I'm reading the number as a creditor's litigation ceiling, not an evidentiary finding. If the methodology surfaces, I'll revisit.
• The audit's $10+ million figure is itself a product of sampling and extrapolation. Different sample selection, different timeframe, different extrapolation method could move the number meaningfully in either direction. The audit is the best available estimate; it is not ground truth.
• SDH's internal financial models may have contained information that justified the magnitude of alarm Pacurar communicated to the board and the press — exposure calculations that incorporated additional audit years, potential exclusion from Medicare, or the cost of the qui tam defense. I don't have access to those models. It is possible the $50 million estimate was more analytically grounded than the public record suggests.

Why This Distinction Still Matters

This is not ancient hospice archaeology for people who enjoy old files and old scandals. The oversight environment now is, if anything, less forgiving.

California hospice fraud is under scrutiny. The policy incentives around long stays still exist. Documentation and recertification audits remain relentless. The next institution-level crisis will arrive in a system that is more suspicious, more politicized, and probably less patient with ambiguity than the one San Diego Hospice faced in 2012.

So the lesson here is painfully current.

Boards and executives have to read the artifact in front of them for what it is. A claim is a claim. An audit is an audit. A judgment is a judgment. Those distinctions are not semantic. They determine what decisions are rational, what disclosures are fair, and how much runway remains for clinical continuity.

I trained at a place that taught me how to sit with hard truths at the bedside. This series is my attempt to do that same work with the institution itself. No nostalgia. No martyrdom. No mythmaking. Just the record, the failures, and the hope that the next board under this kind of pressure chooses analysis over panic and continuity over collapse.

Next: Part 4 — We move out of the file and back to the bedside: long-stay live discharges, the loss of a rare inpatient unit, and what a responsible wind-down looks like when transfer becomes unavoidable.

I am a palliative care physician, educator, and professional strategery expert known for turning rounds into rants and rants into teaching points.