Somerset Medical Center—Somerset County Prosecutor’s Office

The relationship between Somerset Medical Center and the Somerset County Prosecutor’s Office was shaped by a delayed handoff: the hospital had conducted its own inquiry into repeated patient emergencies for months before contacting prosecutors, and what it turned over did not give detectives the records or access they needed. The resulting investigation was not simply a transfer of a case from one institution to another. It exposed a conflict between the hospital’s control of clinical information and the prosecutors’ need to test whether the apparent medical anomalies were crimes.

Delayed disclosure and competing definitions

Somerset Medical Center notified the Prosecutor’s Office after a sixth critical-care incident, following five months in which it had investigated internally without identifying a source. At the briefing, hospital representatives avoided asserting that the events were connected or describing them as homicides. The six patients were Joseph Lehman, on May 28; Frances Kane, on June 4; Jin Kyung Han, on June 16; Reverend Florian Gall, on June 28; Francis Agoada, on August 27; and McKinley Crews, whose death four days before the briefing preceded the hospital’s call to prosecutors. The hospital said all six had abnormal laboratory findings and life-threatening symptoms; five had died.

For detectives Tim Braun and Danny Baldwin, the careful distinction between unexplained medical occurrences and a homicide left the central question unresolved: whether they had a crime to investigate. The contrast was consequential. The hospital had access to clinical expertise and had already reviewed the incidents, but its account did not establish a criminal cause. The detectives, meanwhile, had no clinical foundation for interpreting the laboratory results and no independent evidence yet tying the incidents to a suspect. The handoff thus began with both a signal of possible serious harm and a persistent uncertainty about what that signal meant.

That uncertainty also reflected a dispute over jurisdiction. New Jersey Poison Control Center director Steven Marcus told Bruce Ruck that the data were not ambiguous and warned that patients might continue to suffer while Somerset Medical Center delayed acknowledging the facts. Ruck repeatedly argued that the matter belonged with police. In a later discussion with hospital representatives, he raised the possibility of two additional patients with unusual glucose results and asked whether C-peptide and insulin levels had been tested. These exchanges show the institutional boundary in operation: clinical findings needed to be interpreted, but the question of suspected deliberate poisoning called for a criminal investigation. The repeated insistence that it was a police matter was an attempt to move responsibility across that boundary, not itself proof of what had happened to any individual patient.

The handoff and its paper trail

At the first meeting, Somerset Medical Center’s counsel, Paul Nittoly, explained that the hospital had experienced the six incidents and had notified prosecutors after the latest one. The hospital had also conducted a five-month internal investigation, which, according to William Cors, had not identified a source. The account placed the Prosecutor’s Office in the position of having to assess the incidents while relying on a summary prepared by the institution whose investigation had failed to resolve them.

The first delivery of material did little to close that gap. Nittoly had promised to provide the detectives with the internal-investigation materials. The envelope that arrived contained a five-page faxed memo, including a cover letter, with an apparently missing page. Later, when Braun and Baldwin met with Nittoly and the hospital’s private investigator, Rocco E. Fushetto, they expected notes, interviews, recordings, and a contact list for critical-care staff. Nittoly said that the investigation had not identified the responsible person, that no final reports or definite conclusions had been produced, and that the hospital had contacted the Prosecutor’s Office once it recognized the matter as a police issue. He also said there was no interview report to provide.

The detectives’ difficulty was not merely that the hospital’s conclusions were inconclusive. They lacked material that might let them independently reconstruct the inquiry: interviews, background information, and usable medication records. Their information about the Pyxis system did not yield the drug records they sought, and they had no physical evidence from Somerset that could settle the question of what had occurred. The five-month internal review had generated, in the detectives’ assessment, only four pieces of paper and no answers. With the patient events still not formally established as homicides, the absence of a robust evidentiary record left the Prosecutor’s Office with suspicion but little it could readily prove.

Access, monitoring, and limits on cooperation

As the investigation progressed, cooperation remained constrained by the hospital’s control over its workplace and information. On November 4, 2003, prosecutors began arranging interviews with critical-care nurses in the hope of hearing from staff outside the filtering effect of corporate lawyers. Somerset Medical Center’s lawyers required the interviews to take place inside the hospital with risk manager Mary Lund present. The assistant prosecutor accepted those conditions. Braun objected sharply to the arrangement, while other investigators privately agreed with him; the interviews proceeded under the conditions he had criticized.

The interviews supplied useful background on procedures, scheduling, and the physical layout of the unit, but did not immediately produce a decisive lead. Nurses who had worked with Charles Cullen described him as quiet and somewhat eccentric, yet professionally capable and willing to pick up shifts. The accounts complicated the investigators’ task: familiar colleagues could regard Cullen as a good nurse even while the pattern of patient events placed him under suspicion. The hospital’s access conditions produced information, but not the unmediated inquiry the detectives had wanted.

The Prosecutor’s Office did secure one concrete form of cooperation after Danny Baldwin told Lund that Cullen had a criminal record and a troubled work history, including dismissals from other hospitals over nursing-practice concerns. Baldwin identified him as a strong person of interest and requested the mortality rate on his wards and his work schedule for the year. Lund said the hospital would closely monitor Cullen during his shifts. That response was a protective measure, but the detectives did not consider observation alone sufficient to catch him. They still needed evidence capable of establishing what had happened and connecting it to a person.

Amy Loughren as an evidentiary bridge

The relationship between the institutions changed in practical terms when Amy Loughren, a nurse and Cullen’s friend, became a source of information for the detectives. Her position was precarious: colleagues knew she was close to Cullen, and they remembered her earlier refusal to sign a revised insulin adjustment form because she considered the required estimate of the amount remaining in a vial unreliable. She did not know why the hospital was imposing the protocol, and her supervisor would not explain. As rumors spread that insulin, Cullen, and a death were connected, some nurses avoided her; she became afraid that the investigation might implicate her or threaten her nursing license.

A Pyxis printout shown to Loughren by Baldwin changed what she could accept about her friend. She recognized the record as Cullen’s and understood it as evidence that challenged her instinct to defend him. She could now imagine that he had done something terrible during his shifts, though she could not reconcile that possibility with her relationship to him. The printout did not erase her uncertainty; it made that uncertainty part of her role in the investigation.

Loughren weighed the risk of helping the detectives against her loyalty to Cullen and the possibility that the investigators might be wrong. She feared losing her job and being left exposed by the institution where she worked. Even so, she brought the detectives paperwork, shared coffee with them after shifts, and helped them understand that the Cerner patient records would need to be obtained in full and unaltered. While a detective began subpoena paperwork, Loughren sought a way to help collect the records herself. Her cooperation gave the Prosecutor’s Office a connection to clinical information and workplace practices that its investigators had not been able to obtain through the hospital’s internal inquiry alone.

The relationship’s significance

The interaction between Somerset Medical Center and the Somerset County Prosecutor’s Office never resolves into a simple story of refusal or smooth cooperation. The hospital did report the incidents, provide an initial account, permit interviews, and promise to monitor Cullen. At the same time, the referral came after months of internal investigation; the material supplied was sparse; the detectives could not obtain the medication records and interviews they expected; and staff questioning occurred under conditions set by the hospital and accepted by prosecutors. Each institution operated through different kinds of authority, and the transfer between them repeatedly left investigators short of information.

That gap shaped the investigation’s dependence on people who could cross professional boundaries. Prosecutors and detectives needed clinical evidence but could not interpret or access all of it independently; the hospital possessed clinical systems and controlled access to staff; and Loughren could recognize the meaning of workplace records while also risking her friendship and employment. The relationship’s defining consequence was therefore not an immediate resolution but a prolonged struggle to turn unexplained patient events into evidence that could support a criminal inquiry.