Louisiana investigation into deaths at Memorial

The Louisiana investigation into deaths at Memorial began as part of a statewide inquiry into patients who died at hospitals and nursing homes during Hurricane Katrina, but a report of suspected intentional killings transformed its focus. Prosecutor Arthur “Butch” Schafer and Special Agent Virginia Rider pursued witness accounts and physical evidence, then faced a harder question: what could postmortem drug findings establish about how patients died, and what could they prove about intent? The inquiry’s early course is marked by the distance between an alarming allegation and a defensible conclusion.

A broad inquiry becomes a criminal investigation

After returning home from mourning his daughter, Arthur “Butch” Schafer began examining deaths at health facilities as part of his work as an experienced criminal prosecutor. Louisiana attorney general Charles Foti had publicly pledged a full investigation of the tragedy at St. Rita’s Nursing Home, where more than thirty residents had apparently drowned, and the Louisiana Medicaid Fraud Control Unit opened investigations into every hospital and nursing home where patients had died. The scale of the inquiry reflected a larger concern: deaths at care facilities might have resulted not only from the storm and a chaotic government response, but also from failures of planning, oversight, or evacuation.

The state unit had authority over facilities participating in Medicaid. That fact connected the inquiry to the earlier history of Southern Baptist Hospital, which had resisted joining federal hospital programs in part because its leaders objected to government involvement. By the time investigators contacted Memorial Medical Center’s corporate owners, however, the questions were practical and immediate: the government wanted a list of deceased patients and a copy of the hospital’s disaster plan. The city’s pre-storm evacuation order had not applied to hospitals, which had been instructed to remain open and staffed. If Memorial could show that it had followed its plan, investigators might have been able to close that part of the inquiry and look elsewhere for responsibility.

The first corporate response did not simply supply the requested material. A Tenet Healthcare Corporation attorney asked that the questions be put in writing and sent a corporate summary, newspaper accounts praising Memorial’s employees, and press releases attributing deaths to the conditions inside the hospital and the scale of the disaster. The company’s account emphasized that living patients had been evacuated by September 2, including by privately hired helicopters after government rescue efforts were overwhelmed. The materials framed the deaths as a consequence of conditions beyond the hospital’s control; investigators were still trying to establish what had happened patient by patient.

The allegation from LifeCare

A day after Tenet’s response, an attorney for LifeCare Hospitals of New Orleans notified the state unit that nine LifeCare patients on Memorial’s seventh floor had died under suspicious circumstances. The attorney reported that a physician identified as Dr. Pou had administered morphine when it seemed the patients could not be evacuated successfully. The claim was startling to investigators: allegations of intentional killings in health facilities were rare, and Schafer and his coworkers had not expected such an accusation among the possible crimes they were examining.

Schafer and Rider began interviewing LifeCare witnesses. Diane Robichaux, LifeCare’s pregnant assistant administrator, described the conditions on the seventh floor and recounted that Susan Mulderick had told LifeCare staff the plan was not to leave any living patients behind. Robichaux said that, after returning upstairs, she spoke with Anna Pou. In her account, Pou said the patients would not survive and asked whether Robichaux wanted her staff to remain on the unit. The investigation thus had to examine not only what medication was given, but also what staff understood about evacuation, who was present, and how decisions about the patients were communicated.

Other LifeCare witnesses gave accounts that reinforced one another. They alleged that Pou had come to the seventh floor to end the lives of the nine surviving patients; two recalled that she had described lethal intent directly. The patients had died, but corroboration among witnesses did not by itself answer whether the alleged plan had been carried out, why it had been undertaken, or which other people had participated. Rider therefore listed the records and material evidence investigators still needed: the deceased patients’ medical records, evacuation plans, computer hard drives, employee names and contact information, pharmacy records, and medical waste that might contain discarded vials or syringes.

A warrant and the limits of physical evidence

Investigators concluded that they needed to search the hospital, and that a consensual search was unlikely. Rider reviewed Louisiana’s second-degree murder statute and prepared an affidavit based on accounts from Robichaux and other witnesses, including Therese Mendez, Kristy Johnson, and Steven Harris. The warrant application sought items that could help prove second-degree murder, a charge whose state-law definition included a specific intent to kill. Investigators decided not to seek the death penalty; manslaughter or negligent homicide remained possible lesser charges. A judge granted the warrant, and on October 1 Rider and Schafer joined a large team of state and federal agents and crime-scene investigators at Memorial.

The physical investigation confronted a serious obstacle. Frank Minyard, the Orleans Parish coroner, initially told Rider that autopsies would be of little use because the bodies had decomposed in the heat for more than a week; blood could not be collected to test for lethal drug levels. A volunteer pathologist pointed out that drugs might still be detectable in more durable tissues, including liver, brain, and muscle, or in fluid collected in the abdomen. At the attorney general’s request, Minyard agreed to seek autopsies and tissue samples from patients who had died at health facilities across Orleans Parish after Katrina. Investigators sent samples from eighteen Memorial bodies to National Medical Services, Inc., a forensic toxicology laboratory in Pennsylvania.

The first reported results sharpened the investigation without settling it. Of the eighteen Memorial cases discussed with the laboratory director, nine tested positive for morphine and roughly five or six for midazolam. Later, among forty-one Memorial and LifeCare bodies analyzed, twenty-three tested positive for morphine, midazolam, or both. Those findings warranted closer examination, but they were not interchangeable with proof of a particular dose, a specific mechanism of death, or criminal intent.

Expert disagreement and an unresolved question

At a review of the evidence, Robert Middleberg, the laboratory director, joined investigators and nurses who had organized the drug results into charts and tables. He cautioned that interpreting drugs in postmortem tissue was less straightforward than interpreting blood tests from living patients. Prolonged heat could alter measured concentrations, he advised, so laboratory findings had to be considered alongside each patient’s medical history. The experts examined whether the drugs might reflect a single large dose or repeated administration and accumulation, how kidney or liver disease could affect that process, and whether the timing and known effects of the drugs fit the recorded circumstances of death.

Alice Hutzler’s case showed why individual records mattered. The ninety-year-old LifeCare patient had been treated for pneumonia, and morphine and midazolam were detected in her liver, brain, and muscle tissue. Her chart, however, showed neither drug prescribed and had been kept current until a few hours before her death was recorded. The records also described her as resting comfortably on Wednesday afternoon and contained no documented complaints that indicated a need for those medications that night. The evidence raised pointed questions about her treatment, while remaining part of a broader, case-by-case assessment rather than an automatic answer to the cause of every death.

Minyard consulted Steven Karch, a specialist who argued that drug levels in a cadaver might not reflect the levels present before death. After examining the evidence, Karch recommended leaving the medical cause of death undetermined, reasoning that the bodies had remained in extreme heat for many days. Other experts disputed the weight Minyard gave that position. Michael Baden argued that uncertainty about the precise dose did not erase the fact that morphine and Versed were present shortly before death and that there was no competing cause of death. Cyril Wecht said the levels in some cases were exceptionally high and described a small experiment in which drug levels in samples exposed to hot, humid conditions did not differ significantly from refrigerated samples. The experiment was limited and unpublished, leaving it as one piece of evidence rather than a conclusive resolution.

Minyard imagined the case becoming a battle among forensic experts at trial and feared that reasonable doubt could defeat it, with consequences he believed would harm New Orleans’s recovery. His judgment placed the scientific uncertainty inside a civic and legal dilemma: the presence of drugs, the condition of the bodies, the medical records, and the witness accounts each bore on different parts of the case, but they did not all answer the same question. The investigation’s early narrative therefore ends not with a settled verdict, but with evidence being tested against competing interpretations of what it could prove.