Somerset Medical Center’s practice of conducting an internal investigation first

Somerset Medical Center’s practice of conducting an internal investigation first placed the hospital’s own inquiry ahead of prompt disclosure to outside authorities after several patients suffered unexplained drug-related crises. Administrators said they needed more information before reaching a judgment; the New Jersey Poison Control Center warned that delay could endanger patients and violate the hospital’s reporting obligations. The dispute shows how an internal review could be presented as caution while also restricting information and postponing external scrutiny.

Origins of the internal-first approach

The dispute arose after the hospital faced a cluster of unusual patient events involving insulin and digoxin. New Jersey Poison Control Center director Steven Marcus believed the evidence pointed toward deliberate harm and warned that waiting to acknowledge the facts might expose more patients to danger. The hospital’s pharmacy head, Stuart Vigdor, told Bruce Ruck that the administration had taken over the investigation and had asked him not to speak with outside agencies while internal investigators worked. The decision therefore affected not only what the hospital would investigate, but who could discuss the incidents with people beyond its walls.

The hospital did not describe the matter as settled. Vigdor emphasized that the investigation was ongoing and that the administration did not yet know for certain whether the patients had been harmed deliberately. William Cors likewise said the hospital was trying to obtain more information before making a judgment. He acknowledged that the institution was weighing the risk of throwing the hospital into chaos against its responsibility to prevent further harm. The appeal to avoid a premature conclusion had a plausible investigative purpose, but it also left the central question—whether patients had been deliberately poisoned—inside the hospital’s own process.

Dispute over urgency and authority

Steven Marcus rejected the idea that Somerset could wait until its internal review was complete. He told administrators that suspected incidents had to be reported to the state within twenty-four hours and that the hospital was already out of compliance. His objection was informed by earlier cases in which, he argued, hospitals delayed legal investigations and lost the opportunity to track a suspect who moved on. He stressed that clinical expertise did not make hospital physicians effective forensic investigators, distinguishing the need to treat patients from the need to preserve and investigate evidence.

Somerset’s representatives continued to frame the delay as an effort to avoid a rush to judgment. William Cors explained that the hospital had sought an expert opinion from Poison Control and that doing so had put the center in an uncomfortable position. Marcus answered that the wider pattern made the matter a police and patient-safety concern, not simply a clinical puzzle for hospital administrators. The clash was thus partly about jurisdiction: whether the hospital could retain control of the inquiry until it felt ready to report, or whether the warning signs already required state and law-enforcement involvement.

The conflict became more consequential when Steven Marcus told the administrators that their conversations had been recorded. He had already contacted Eddy Bresnitz, a state epidemiologist and assistant commissioner, describing a cluster of hospital illnesses that might have resulted from a criminal act. He also emailed Amie Thornton, an assistant commissioner of health, summarizing four troubling clinical events and Somerset’s unwillingness to report them until it had completed its own investigation. His warning that an eventual report would look worse if he had to make it himself made clear that external notification was no longer a hypothetical next step.

Reporting and the limits of the review

A few hours after Marcus disclosed the recordings, Mary Lund reported the four patient incidents to the New Jersey Department of Health and Senior Services by fax and email. The report covered cases involving digoxin and insulin and set out the checks the hospital had already made. Staff had reviewed possible manufacturer recalls and adverse drug interactions, checked the IV equipment and bedside monitors, and repeated laboratory tests to rule out lab error. The hospital had also tightened controls on digoxin, following similar restrictions on insulin, so that nurses would be more accountable for these commonly used drugs.

Somerset still presented human error as the likeliest explanation. Administrators reasoned that ordinary medication errors were possible and that an innocent mistake should leave evidence in the records. They said they were reviewing both the Pyxis MedStation 2000 drug system and Cerner patient charts, but had not found an innocent explanation. At the same time, the state was told that human-resource factors were being evaluated and that independent investigators were interviewing staff. The internal inquiry therefore did not amount to an admission of intentional poisoning; it proceeded through competing possibilities, documentation checks, tighter controls, and interviews while the evidence remained uncertain.

That uncertainty did not remove the costs of keeping information contained. Amy Loughren had already challenged a new insulin sign-out procedure that required nurses to electronically attest to how much remained in a vial. She considered the estimate unreliable and demanded to know why insulin was suddenly being treated as a dangerous drug, but her supervisor would not explain what had happened. The policy’s unexplained arrival gave her a warning without the information needed to understand it. Later, when nurses knew she was Charles Cullen’s friend and associated her earlier protest with the insulin issue, some avoided her; she became frightened that the investigation might threaten her own position. Those consequences do not establish why the hospital withheld details from her, but they show how limited disclosure could leave frontline staff exposed to suspicion and fear as well as uncertainty.

Consequences for staff and investigators

When a detective showed Amy a Pyxis printout from the night Reverend Florian Gall died, she recognized it as Charles Cullen’s record. The document broke through her previous instinct to defend her friend: she could now imagine that he had done something terrible, though she struggled to reconcile the evidence with the person she believed she knew. Her later cooperation placed her job and her loyalty at odds. She feared that investigators might be wrong or manipulating her, yet she began sharing paperwork and information with them, including her knowledge of Cerner records. The transition from hospital employee to confidential police source grew out of a process in which the relevant information reached her piecemeal, through the detective rather than through a direct explanation from her workplace.

For the investigators, the distinction between an internal inquiry and a prosecutable case remained critical. The Pyxis system appeared to preserve access records for only thirty days, and the hospital had told detectives that the records for Gall’s death were gone. Because Gall had been dead for months, the missing data seemed to block the connection between suspected poisonings, drug access, and Cullen. The episode does not prove that Somerset’s internal-first policy caused the loss of those records, but it illustrates the stakes of delay and controlled access: clinical concern and an internal investigation were not by themselves enough to establish who had obtained a drug or administered it.

Somerset’s approach therefore functioned both as an investigative procedure and as a barrier that outside agencies had to challenge. The hospital eventually reported the incidents, reviewed its systems, tightened medication controls, and arranged interviews; the New Jersey Poison Control Center and state officials, however, had to press for timely reporting and independent scrutiny. The account leaves the tension unresolved rather than treating internal review as inherently illegitimate: careful verification mattered, but so did the speed, transparency, and preservation of evidence required when patient harm might be deliberate.