The Doctor Trying To Solve The Mystery Of Food Allergies

April 15, 2013

Guests: Charles Graeber - Kari Nadeau

TERRY GROSS, HOST:This is FRESH AIR. I'm Terry Gross. In 2003, police in Somerset County, New Jersey, arrested a hospital nurse named Charlie Cullen who was suspected of injecting patients with lethal doses of a variety of medications. Cullen would turn out to be one of the nation's most prolific serial killers, murdering dozens, perhaps hundreds, of people in nine hospitals over a 16-year period.

Our guest, journalist Charles Graeber, spent six years investigating the Cullen case and is the only reporter to speak to Cullen in prison. Graeber's new book, "The Good Nurse," focuses not only on Graeber's tortured life and crimes but on why it took so long to stop him despite plenty of evidence he was harming patients.

In case after case, Graeber writes, hospital staff believed Cullen was harming patients and pressured him to leave, but failed to alert state regulators or take other steps that might have ended his killing spree. Charles Graeber has written for Wired, GQ, New York Magazine and other publications. "The Good Nurse" is his first book. He spoke with FRESH AIR contributor Dave Davies.

DAVE DAVIES, HOST:Well Charles Graeber, welcome to FRESH AIR. This is a remarkable story both of a serial killer and what drove him to commit these crimes, but also one about the health care system and its vulnerability to someone who abuses patients. And I want to start with one of his - maybe his earliest employment at the Saint Barnabas Medical Center in Livingston, New Jersey.

And you write that in 1991, he'd been there a few years, and some of the staff noticed some weird stuff. What did they find?

CHARLES GRAEBER: Right. Well, he'd been there since '87. It was his first job out of nursing school. He started there right out of his honeymoon. And on the ICU ward and CCU wards, they started finding that patients were mysteriously crashing. They would hook a patient up to an IV that was supposed to have saline or something of the like, and the patient became a magical diabetic and where they were on this terrible diabetic roller-coaster ride, woozing(ph) in and out of consciousness and burning through sugar, and they couldn't out-feed the fire.

And then eventually they'd get so bad that they'd unhook them from the lines and rush them in for more emergency procedures, and once they unhooked the lines, everything changed. They were fine again.

DAVIES: That's the line connected to an IV bag, right?

GRAEBER: Correct, correct. So once they unhooked them from the IV, they were OK, and trial of error, a few of these on the ward, and they started questioning the IV bags themselves and brought them in and found that in fact they'd been compromised. There were extra pinpricks in them, and when they analyzed the IV bags, they found that instead of containing saline, the IV bags that had been stored in the medical closet also contained insulin.

DAVIES: So it appeared someone had taken a hypodermic needle with insulin and injected them into the IV bags, leaving virtually no trace, and then causing catastrophic consequences for the patients.

GRAEBER: Exactly, and it was the only conclusion anyone could come to. But it was beyond thinking, because why would anyone do that, who would do that. Suspicion immediately fell to somebody with a specific vendetta, some sort of family member or - you know, it was really very abstract. The hospital had never dealt with anything like this before.

And eventually nurses were cross-indexed, shifts were cross-indexed, and a few names came together, and one name finally stood out more than any, at least according to the head of security there, Thomas Arnold(ph). And that name was Charles Cullen.

DAVIES: And he was the nurse at the center of this. Did some of these folks die?

GRAEBER: People died at Saint Barnabas. It's very difficult to say who died of what. Saint Barnabas is so long ago, and at the time they didn't catch him. I mean, they had him. They strongly suspected him. Tom Arnold said he knew Cullen was dirty, but could they prove it? So it was very difficult to sort out Cullen's personal death toll from the cadence of mortality at a hospital.

Later speaking to Essex County detectives, the county that Saint Barnabas is in, Cullen said that he was dosing people and with the intent of killing them two or three times a week, not knowing where those bombs went out. And to date, they've only been able to identify one person from the five years definitively that died during his Saint Barnabas time.

So obviously there's a huge fog of time, five years time two or three a week, that's totally unaccounted for.

DAVIES: Right but enormous damage, no doubt. So the investigation focuses on him. What happens?

GRAEBER: Well, he's investigated, pressured, interviewed, tells them essentially there's nothing you can do, you can't hold me, you don't have anything on me, you're just picking on me; leaves, is contemptuous, which really riles the detectives, former cops. They go to the chief of police in the town of Livingston, and the chief, for whatever reason, bounces it right back to the hospital and says this is internal, you guys deal with it, you figure out what happens.

And they don't figure out what happens. Cullen ends up as a floater. He's basically working at the hospital fulltime, but he's staffed by an outside but wholly owned staffing agency of the hospital, which makes tying him to the hospital and specific shifts all the more difficult. And they're still investigating when suddenly they stop giving him shifts.

It's not technically being fired; he's just no longer on the list, and he moves on. And their problems move on with him.

DAVIES: And it's interesting. He doesn't exactly deny it in the interviews.

GRAEBER: No, no, he doesn't - he very rarely denies anything, except during polygraphs, which he passes.

DAVIES: Right, we can get to that. He was good at that. What's fascinating is he goes down to the next job, and when they ask about previous employment, he lists Saint Barnabas, the place where they'd discovered him effectively killing people.

GRAEBER: Sure.

DAVIES: And any impact?

GRAEBER: No, I mean, in fact it worked out really well because they fired him in the beginning of January, and he managed to get over to applying to his next hospital, which was Warren Hospital, at the end of January of 1992. And so he was able to say dates of employment, X date to January 1992, which made it look like he was currently employed and simply wanted a change, rather than having been canned.

And he was given neutral references, which was the norm and the pattern over and over again. And during a nursing shortage with neutral references, seemingly, you know, a nurse that willing to work nights and weekends and holidays and seemed to bring energy with every shift and a lot of experience, that was a nurse worth hiring.

DAVIES: Now I have to note that this happened in 1991, 12 years before the investigation that ultimately caught him. This was - clearly he'd been doing enormous damage, possibly killing patients, and he moves on, works at another seven hospitals, right?

GRAEBER: Yeah, well total of nine, yeah.

DAVIES: Right, OK, so another eight, then. And in case after case, I mean, this is not the only time in which his conduct is suspicious, and supervisors noticed, and he gets questioned, and he moves on and again and again gets new jobs, accurately listing his previous employers as references. Did you ever find a single case of an employer who warned somebody stay away from this guy?

GRAEBER: Well interestingly enough, the employers that did that were the ones that were most in the wrong for it. At Saint Luke's Hospital in Pennsylvania, some years later, tending towards 2000 almost, he was fired after being caught trashing medicine, basically had been taking good meds and throwing them in the sharps bin of the medical closet.

And upon examination, many of those meds were dangerous meds, and many of them were, in fact, empty, and in fact there'd been no prescriptions for those medications in question.

DAVIES: So it appeared he might have been stealing them for future use?

GRAEBER: Stealing them, stockpiling them for future use, or the suspicion at the time was using them on patients on the ward. And in fact he was doing the latter, and he was caught. Outside counsel was brought in that night, he was grilled. He was moved on. He was offered neutral references and took them and removed from the building, but the hospital administrators called their peers in the immediate circle asking for more information about Charles Cullen and also informing them that Charles Cullen was not considered for rehire at their hospital and shouldn't be at theirs, either.

A judge found later, when this secret behavior came to light, that - well his issue was that it was the ultimate tyranny, deciding that these administrators had decided who would live and who would die by which hospitals they called and which they did not. A hospital they did not call, for example, was Somerset Medical Center, where Charles Cullen moved on and killed at least another 16 patients that we know of.

DAVIES: Now was there any requirement that hospitals report this stuff to the state health department or an agency that regulates nursing?

GRAEBER: Yeah, you really have to look at that case by case, and that's one of the reasons I took on this book and a reason it took as many years as it did. Some incidents, sentinel events, events that result in actions that are of potential harm to patients, are supposed to be reported to the state nursing bstate ents, sentinal t theoard. Other incidents, such as medication errors that are more routine, he had a lot of those, as well, and it's again difficult to sort out which were legitimate mistakes and which were simply the MO of murder. And more of those should have been reported.

Very few were, and the question time after time is: Should more have been reported? Yes, absolutely. And you have to go hospital by hospital, case by case, and really look at which incidents should have been reported. It's also - one needs to be careful, at least I certainly need to be careful. I wanted the reader to be able to put the facts together.

The facts had been buried for a long time. No one was speaking. Charles Cullen speaks only to me. He'd never spoken to any other media. He is not happy that this is coming out. The hospitals have not spoken at all except to give a very basic story about, you know, having been duped and then the last one having caught him.

And the detectives were largely unavailable for comment, as well. So this story, laying it out so that a reader can see the facts laid side by side and decide for themselves the culpability of the hospitals, what they knew, when they knew, what they should have done. And certainly laws have changed in the wake of this.

DAVIES: Right, I want to move on to talk about Charlie Cullen in a moment here, but, I mean, an obvious motive for a hospital to be careful about telling anybody else that they had somebody in their system who harmed or killed patients is fear of lawsuits, right?

GRAEBER: Absolutely, and that's a legitimate fear. Hospitals get sued all the time. They have to carry huge insurance burdens, as do doctors. Their first instinct is to cover themselves, and I don't mean to confuse that with a cover-up, I don't know that that was always the case. It is beyond the imagination of most people, most hospital administrators, to ever think that their staff are killing their patients. They're really there to do the opposite.

So it does take a leap of imagination. I do know that the first actions you see time and time again in these hospitals is a legal action rather than an effective investigative reaction. And oftentimes you'll find that comes in - certainly in retrospect to be a real burden of evidence against one guy, Charles Cullen, even before they can say for sure what he's doing, when it starts to really look like this guy is dirty, that's the time he gets moved on. One way or the other, he's pushed out or pressured out.

So did the hospitals know? That's a question a reader needs to ask, and I think I've provided enough evidence that they'll be able to draw that conclusion. But certainly he should have been stopped before he was, and because he wasn't, he killed a lot more people.

DAVIES: We're speaking with investigative journalist Charles Graeber. His new book about serial killer Charlie Cullen is called "The Good Nurse." We'll talk more after a short break. This is FRESH AIR.

(SOUNDBITE OF MUSIC)

DAVIES: This is FRESH AIR, and our guest is investigative journalist Charles Graeber. He has a book about Charles Cullen, who was a nurse who killed an untold number of people over many years at hospitals in Pennsylvania and New Jersey. Graeber's new book about the case is called "The Good Nurse."

Let's talk about Charlie Cullen. Very troubled childhood, right?

GRAEBER: Yeah, he describes it as miserable. This is a really dark period not only for Charlie but also for anyone investigating Charlie Cullen. It's a land of shadows, a world he doesn't talk very much about. It's a world that if you were to ask him straight about his childhood, you would not get very much. You tend to learn more about his childhood talking about other topics and then hearing echoes of his childhood, references.

He grew up the youngest of eight, the youngest by a long shot, sort of an unexpected child. His father died only months after his birth. His mother survived largely on sewing and charity. There was a sickly aunt that had moved in with them. He had two brothers with chemical abuse problems and a host of sisters that came in and out of the house, and he - he felt extremely close to his mother. His mother was the figure that protected him.

But the darkness in that house and exactly what happened in that house set the tone for the rest of his life.

DAVIES: Darkness meaning drug addicts coming in, doing who knows what in rooms, abusing him in some cases?

GRAEBER: It's really hard to say. When asked directly about abuse of that sort in the house, he gets very angry. He's gotten very angry with family members, with ex-wives when they've tried to get him to seek counseling, when they've tried to take him aside, because the pattern - it certainly seems to fit the pattern. He would say.

But he felt unsafe. There were strange men in and out of that house. He had a brother-in-law that came to live with one of his sisters when his sister was pregnant. There was a lot of domestic violence surrounding that. Exactly what happened to the child is not clear. But eventually the sister ran away, but the brother-in-law stayed, and he and Charlie had a tortured relationship that Charlie had reported to at least one, if not two, of his later lovers that he had tried to poison that brother's drink - he put lighter fluid in the vodka, which is sort of an early example of what would become his pattern for life, a way of passively dealing with things.

DAVIES: He serves in the Navy. There are a number of examples of bizarre behavior there and apparent suicide attempts. You talked to him a lot of times. Did you feel like you got a sense of what his issues were? Is there a psychiatric diagnosis of Charlie Cullen?

GRAEBER: Yeah, well, he's not crazy in the sense that clinicians use that term. Well, clinicians don't use that term. He's oftentimes very funny, seems to have great self-humor. When I say funny, he's not cracking jokes and being crass. It's quite the opposite. It's self-deprecating, insightful, sardonic and seems quite with it, and you can understand why he was able to succeed, why he was charming, why, you know, he had no romantic - you know, he had no problem finding romantic partners and the like.

Other times his thinking is circular, quite narcissistic, and then the question is how far does that narcissism go? Is it sociopathic? And the answer to that lies somewhere in - well, you have to ask yourself what sort of a person can kill someone and be there as they die and not have it seem to really affect their day at all or in fact affect their future behavior in any negative fashion for 16 years.

You know, he started - he entered nursing in 1987. He said he killed his first patient in 1987. So, you know, this is a lifelong condition. It's almost a compulsion. And what it meant to him I guess is the real question of that, and I don't think it meant that much.

DAVIES: When he was in the hospital and injecting patients with a whole variety of drugs and killing some, you know, this isn't a case of suffering patients who died peaceful deaths, right. I mean, some of these were pretty horrific.

GRAEBER: Yeah, he's often misunderstood, to the extent that anyone knows him at all, he's often called a mercy killer. They called him an angel of death, and that's what that refers to. And the point here is that it was never about the patients. It was never about what the patients needed or even - or wanted, regardless of how appropriate. It was always about what Charlie Cullen wanted and what Charlie Cullen needed.

And if that meant - well, if that meant putting a man with a broken neck and a halo device into insulin shock, that's what happened. If it meant going back and trying time and time again on a patient that he really had his eye on and sending them into atrial fibrillation, over and over again, and, you know, forcing them to be shocked and paddled and coded, time and time again, that's - that was considered fair game.

Whether that was his goal, he would deny that that was his goal - adamantly. But he's also - he's a smart guy and doesn't want to be seen as - well, he doesn't want to been seen as how you see him.

DAVIES: One more thing about Charlie and the killings. I mean, hospitals have what are called code teams, that is to try and resuscitate patients who go into cardiac arrest. They call it a code, right?

GRAEBER: Right.

DAVIES: And Charlie was always among the most active members of that. And in a lot of cases, you write, went in to try and resuscitate patients who were in cardiac arrest very likely because of, you know, medications he had secretly administered to them. And it made me wonder, is - was part of the motivation here, that he wanted to be the hero, the very best (unintelligible), very best member of the code team and demonstrate that by creating these incidents and then rushing in to be the hero?

GRAEBER: Yeah, that's absolutely part of it. Sometimes that's what worked for him. He knew what was wrong the patient when no one else did. He could be the first to go in there. The other residents remember him jumping on the chest of the patient in just sort of the most dramatic fashion and working - they appreciated his enthusiasm and his passion, but it seemed a little over the top. But the truth was he did what others could not do, and he did receive praise for that elevated status.

And so there was absolutely an element of ego in the murder, and he was anonymous, but this was a way of actually still claiming (unintelligible) for some of that - for some of the crime he committed under anonymity.

DAVIES: And he knew how to beat a polygraph test?

GRAEBER: Yeah, I don't - he doesn't speak to that. I know that he did beat a polygraph test. He beat two polygraph tests. And in both cases, he was lying. He also was, you know - again the best I can do with this sort of journalism when you can everything, you can't see behind every wall, and not everyone says everything you need that - in order to put the story together fully.

But I can just put the facts next to each other, and the facts next to each other are that here's a guy that worked on cardiac wards administering drugs that regulated heart rhythm and stress levels, and everything else, and also very possibly didn't genuinely feel the sorts of wild winging emotions that you and I might feel thinking back on murder.

And put those things together, and you end up with a guy who doesn't register on a polygraph.

GROSS: Charles Graeber will continue his interview with FRESH AIR contributor Dave Davies in the second half of the show. Graeber's new book is called "The Good Nurse." I'm Terry Gross, and this is FRESH AIR.

(SOUNDBITE OF MUSIC)

GROSS: This is FRESH AIR. I'm Terry Gross. Let's get back to the interview FRESH AIR contributor Dave Davies recorded with Charles Graeber, author of the new book "The Good Nurse." It's the story of Charlie Cullen, a nurse who became one of the nation's most prolific serial killers, murdering dozens - perhaps hundreds - of patients in nine hospitals over a 16-year period by injecting them with lethal doses of medications. In case after case, hospital staff believed Cullen was harming patients, but failed to take the steps that could have stopped him.

DAVIES: So, how did it all finally come apart? What did Charlie do that finally led to the criminal investigation that put him away?

GRAEBER: Nothing that he hadn't done before. It did finally catch up with him. There was a gathering steam of evidence behind him that, you know, you can only have so many investigations, especially in the age - the mounting age of computers and linked databases to not be able to connect the dots. But what happened was he was at Somerset Medical Center. A number of patients were coding mysteriously, most were digoxin deaths, digoxin being a heart drug like foxglove, it regulates heart rhythm. Others were insulin deaths. They were in the same ward. They were in multiples.

The hospital noticed it. They had a number of meetings. They asked an assistant pharmacist to call the poison control people to see if they could get help figuring out the math of the chemistry, figuring out when a patient would have to be injected with how much dig - dig, which is digoxin, short for digoxin - how much dig a patient would need to be injected with and when in order to start showing these numbers at, you know, X hour. And in the process of asking for help with the math, of course, they saw in the details, the facts of this started to come out, and alarm bells naturally started to ring.

Outside of the hospital, at New Jersey Poison Control, it was Dr. Bruce Ruck who first was alerted, and he alerted his boss, Dr. Steven Marcus. And Marcus very bluntly took it to the administration of Somerset and said, if you're slow rolling this, you're going to look terrible with your pants down. It looks like, you know, it looks like someone's killing patients in your medical center, and if you don't call, I'm going to call. And, by the way, I've got this on tape.

DAVIES: Yeah. This is fascinating, because you - we have these conversations in your book verbatim, because the Poison Control, I guess, routinely tapes calls like this.

GRAEBER: Yeah. They tape their incoming calls, and these were incoming calls.

DAVIES: So a call that the hospital, the pharmacist thought was a request for information in - I mean, soon leads to this Dr. Bruce Ruck saying, listen, you have a police problem. Do you hear me? And you have to wonder, if it hadn't been for that interaction, if it wouldn't have been a criminal investigation. I guess you can't know.

GRAEBER: You can't know. Certainly, the hospital - hospitals have a tendency to do everything by committee. They move very slowly. Again, it takes a huge leap of imagination - although, it seems fairly clear here, but it does take a huge leap of imagination and a great large amount of acceptance to ever believe that someone, you know, on staff somewhere in your hospital killing other people in your hospital. You know, that's not why you're there.

So having said that, they retained outside counsel. They had been aware of this problem for months at this point. Charles Cullen was killing people for months. He'd been spoken to about related issues, some of the means by which he was getting his drugs. He was questioned about those means by an outside attorney that the hospital had hired. So this was an ongoing issue at the hospital. It got pushed to the forefront, certainly, by an outside entity refusing to allow that internal process to continue to drag on. And it's terrifying to speculate as to what would've happened if - had he - had Dr. Marcus, Dr. Ruck not pushed it.

DAVIES: So the Somerset County Prosecutor's Office launches an investigation, and this is not the kind of thing they're used to doing. They don't have a lot of heavy-duty murders there, anyway, and this is a very specialized kind of crime. Two detectives, Tim Braun, Danny Baldwin get assigned. And, of course, they know that, you know, that most of the information is going to be at the hospital. They approach the hospital. What do they get?

GRAEBER: Right. These guys, Tim and Danny, I spent a lot of time with them. They're really great guys, interesting characters, and were really blunt with me. They came from Newark, which was the murder capital of the world when they were there, Newark Homicide. And as Tim put it, me and Danny, you know, we're street guys. We're blood-and-guts guys. You know, give me a - I've got to delete half the words here but, you know, give me a good old-fashioned street to murder, as opposed to this Latin stuff. Because what they were presented with was they weren't even sure if they had a murder. They thought maybe it was just a well-connected guy and some sort of a litigious situation where they couldn't figure out where inappropriate drugs had come from or gone to and...

DAVIES: A well-connected victim trying to make something of it. Right. Right.

GRAEBER: Exactly. A well-connected victim, because it was a - it's a wealthy county, one of the more wealthy counties in America, and as you say, not a lot of murder there. And they proceed because they have to. They've got stacks of medical charts, which apparently the hospital themselves had already gone through. You know, as they would put it, guys with letters after their name had already looked at this stuff for four months, and they didn't find anything. Now they're handing to us like we're going to, you know, trace this thing like a bullet. They were completely confused.

DAVIES: But here's what's fascinating. So they give - the hospital people, having conducted their own investigation, having spoken to many, many nurses and other staff, give these detectives I think a four-page document, which is not very informative, and then tell the detectives they did not take notes in their interviews of the nurses and other staff, which just seems completely unbelievable, doesn't it?

GRAEBER: It did to them.

DAVIES: Yeah.

GRAEBER: They were really upset. Tim and Danny, the detectives were very upset. It didn't strike them as being true or helpful.

DAVIES: Right. Now, there was another thing. There was a technical system that the hospital had for dispensing medication, which keeps a record when a nurse has to log in to either - to request and withdraw medication, or to cancel a previous request. And so there is data on which nurses have requested which medications, date and time, right?

GRAEBER: Right.

DAVIES: And when the detectives ask for that data about these nurses in the patients in question, they are told what?

GRAEBER: Right. It's called a Pyxis machine, and they asked for the records. It would be like asking for all the receipts for the last, you know, period of time that, you know, you say these victims died over this period of time. Can we see everything that happened to them? And they are told flatly that all those records are erased from the system two months after, within two months, which would mean that every single victim that they're looking at has no medication records.

DAVIES: And then when they check with the company that makes the system, they discover what?

GRAEBER: Yes. And quite a few steps later, when their beating their heads against the wall, saying how do we solve a maybe murder with no evidence whatsoever, not even a paper trail - in fact, they're questioning whether they're even murders and contemplating early retirement - they do call the company that owns the thing. And the immediate response is no, of course. It's all the - the data is stored forever. Why would it dump data every two months? That doesn't make any sense. Are you having a problem with your machine? And that creates a whole new dynamic. That really is what cracks the case. It also really pisses off the detectives.

DAVIES: Right. And I have to say, I mean, when I read the book, I am as horrified by what hospital administrators do and don't do as I am by what Charlie Cullen does in a lot of cases here. And you have a case here where Charlie Cullen had clearly been doing some very, very scary stuff, and the hospital had investigated it. And not only did they not volunteer the information to law enforcement, not only did they not share it with law enforcement, they told them things that were untrue, which, by any common understanding of the term, impeded the investigation. Am I missing something here?

GRAEBER: No.

(LAUGHTER)

DAVIES: Right. All right. I have to just get that out and make it clear. And I also want to note that you did - as you write in the book - contact each of the hospitals where Charlie Cullen had worked and done stuff. And Somerset Medical Center gave you a statement, which I think we should read here in full. It said: (Reading) Somerset Medical Center fully cooperated with all interested parties and agencies throughout the course of the Cullen investigation. At this time, we are devoting the full extent of our resources and efforts on delivering the highest quality of care to the members of our community.

That's the beginning and end of what you got from them?

GRAEBER: I received one more communication. The risk manager, who had been the point person for those communications about the medical records and the like, that was obviously someone that I was very interested in speaking with. And that person was - I was informed not going to be available to be spoken with, but no change in the bottom line.

DAVIES: All right. And we should say that there were never any actions, any official or criminal actions taken against hospital administrations for any of these events, right?

GRAEBER: That's correct.

DAVIES: OK.

GRAEBER: As far as I know, that is correct.

DAVIES: So what does Charlie Cullen stand convicted of? And do we know how many people he may have actually killed?

GRAEBER: We'll never know how many people Charlie Cullen ultimately killed. Charlie Cullen doesn't know how many people he killed. He initially could recall 40, and also said there was a large part of his life that was a fog, during which he would have no ability to recall. But during that fog - those fogs lasted years, and he said that there were probably multiples a week to other detectives.

He was convicted of at least 29, and then civil court, another, I believe, eight. The specifics are confusing, because the further back you go, the fewer records there are. There are no medical records, really, that come out of St. Barnabas. There are very few records of that initial investigation. There are no bodies that one could look at. There was no Pyxis. And so during that five years, you've got one murder, even though he - Cullen himself has said there are far more...

DAVIES: That's the first hospital where he worked in the burn unit and was so active. Right. Right.

GRAEBER: Exactly. So, but fast forward 16 years and look at the last six months, right before he's caught, where there are bodies in the ground that can be autopsied, there are medical records they discover, both the patient charts and, you know, all the machines, everything computerized, everything available they discover finally, and you've got 16 murders in the last six months. So we'll never know exactly...

DAVIES: It could be hundreds.

GRAEBER: According to some psychological profilers that I have intimate knowledge of this case, it's likely hundreds.

DAVIES: Well, Charles Graeber, thanks so much for speaking with us.

GRAEBER: Thank you, Dave.

GROSS: Charles Graeber spoke with FRESH AIR contributor Dave Davies. You can read an excerpt of Graeber's book, "The Good Nurse," on our website, freshair.npr.org.

Coming up, Dr. Kari Nadeau talks about her research into treating children with severe food allergies. This is FRESH AIR.

TERRY GROSS, HOST:This is FRESH AIR. No one is certain why food allergies are on the rise. But now, one of every 13 children has a food allergy. Nuts, soy, milk, egg, wheat and shellfish are some of the foods that most commonly set off allergic reactions. In some cases, the reaction can be so severe, that it results in the throat swelling up and closing, leading to death. For a child with a severe food allergy, every meal that isn't made under appropriate supervision can be hazardous. And since so much social life is based around food, family life becomes restricted.

My guest Dr. Kari Nadeau is one of the scientists at the forefront of food allergy research. She directs the Stanford Alliance for Food Allergy Research, SAFAR, at Stanford University School of Medicine. She's an associate professor of allergies and immunology at the school and the Lucille Packard Children's Hospital. Dr. Nadeau is currently conducting a clinical trial testing a technique for desensitizing children with multiple severe food allergies.

Dr. Kari Nadeau, welcome to FRESH AIR. Before we talk specifically about the research you're doing, 90 percent of food allergies are caused by cow's milk, egg, soy, wheat, peanuts, tree nuts, shellfish and fish. Why? Like, do those foods have something in common?

KARI NADEAU: That's a great question. We're still trying to understand those answers. I think that we know that at certain points in life, cow's milk is more prevalent in children with food allergies, compared to peanut or shellfish or fish, for example. But that all of those foods, they don't have one protein in common. So we think that there's something about the foods and about how they're being delivered that people are becoming allergic to them. And importantly, is that 30 percent of people with food allergies can actually be allergic to more than one of those foods. So we don't exactly know why those foods, and that research is still underway.

GROSS: What do you mean when you say how they're being delivered might be a factor?

NADEAU: Yeah. We're looking at whether or not other dietary components and whether or not environmental factors - for example, different environmental exposures like pollution and tobacco and whether or not they're being delivered orally or through breast milk, or with other environmental features associated with them, we don't know the answers to how they're being delivered in the gut and why, to some children, they're seen as toxins, as it were, just seen as allergens that their body responds to, versus in other children, they're fine and they're used as nutrients. So we don't understand what flips the switch between a food allergen versus a food nutrient in children that are taking those foods early in life.

GROSS: What happens in a severe allergic reaction to food that makes it life-threatening?

NADEAU: In some cases, what happens is the food allergen is ingested and the body sees it as foreign. And it sees it as foreign as if it were an allergen that could cause a severe issue in the body. And so what happens is this huge chemical response occurs within minutes and you can have powerful chemicals released by the body that can hurt the body. And those chemicals are called histamine and there are many other chemicals as well.

But the majority of those chemicals lead to swelling and that swelling can occur internally in the body as well as externally and it can cause a lot of itching. So people get hives. But they can also get swelling internally to - unfortunately in some cases - these are rare but it can happen - it can lead to swelling internally of the throat and in the lungs, and respiratory issues are some of the major ones that we worry about in a severe reaction.

And these reactions can occur within minutes. And so one has to be ready very quickly with an EpiPen to be given right away if someone has any respiratory distress.

GROSS: And an EpiPen is a shot of - it's a little pen injector that has epinephrine, which is a form of adrenaline, right?

NADEAU: Right. Exactly. It's very simply that.

GROSS: You direct the Stanford Alliance for Food Allergy Research at Stanford University. So you're doing state of the art research on food allergies, and one of the approaches you're investigating is an approach that's been used for a long time to desensitize people with allergies to like grasses or pollen or cat dander. Would you describe this approach to food?

NADEAU: Sure. What we are doing is what one would call something that has been done over the past 100 years with pollens and with other allergens like bee sting allergies as well as to animals. So what you do is you give the person back the item that they're allergic to. In the case of foods, there are many different types of therapies being tested right now by many groups around the country as well as the world.

But what it usually involves is a small amount, let's say specks, of the food. Very small amounts of the food that you can barely see in a dish. And then over time you increase that dose ever so slightly. And we want to make sure that the patient can tolerate the dose well and this is safe. And then every two weeks you come back into a clinic and you increase that dose.

And this should all be done in a research setting with a very well trained team of individuals that are ready to help the patient for any type of reaction. So this type of therapy over time is called desensitization. And by the end of a number of months to years, one can get up to, if, again, they succeed and comply with the study, one can get up to about a serving's worth of food.

GROSS: Now, one of the things that you're doing that's very cutting edge is that you're trying to desensitize individuals to several foods at a time. So instead of just, OK, I'm going to desensitize you to corn and then we'll desensitize you to wheat and then if you do well with that we'll desensitize you to cow's milk - you're taking maybe three things that the person - three foods the person is allergic to and trying to give them trace elements and then tiny bits more of each of those foods at the same time.

NADEAU: That's right. People have been doing work in monotherapy around the country as well as around the world...

GROSS: So monotherapy is exposing to one food at a time.

NADEAU: Exactly. So people - peanut therapy was being given at University of Arkansas, University of North Carolina, and then milk and egg was being looked at in Mount Sinai. And milk was being looked at at Johns Hopkins. And so a lot of exciting data was coming out and still is. But for me, I had one patient come up to me and just say, you know, I have so many allergies, there's no way I could do the monotherapy in my lifetime.

It would take 10 to 20 years. And there are many people with multiple food allergies; 30 percent of people with food allergies have more than one food allergy. And so I thought about it a little bit more and then I called - I'll never forget stopping on a highway. It really bothered me because I kept on pondering this and I was driving to my aunt's house in New Hampshire.

And so I stopped along the road and I called my colleague Wes Burks and I had a conversation with him. And I said do you think it's possible - do you think that one could actually simultaneously give multiple foods in oral immunotherapy? And he also said, OK, let me think about it. I'll get back to you. And then let's really design this trial so that, again, safety is paramount. And how could we do this best?

So I thought, OK, if we're going to give multi maybe for some people we should also give this anti-IGE therapy. And that's when I called Daila Metzu(ph) and talked to him.

GROSS: And what's anti-IGE therapy?

NADEAU: Anti-IGE is a therapy that was approved for asthma and it's used to actually try to inhibit one pathway of the allergic response. And so by giving it, it's kind of a cover that helps the immune system be able to take the food at higher doses much faster in this type of oral immunotherapy study. And so people were able to get to their higher dose of food compared to people who didn't take the anti-IGE. And they were able to do that much faster.

GROSS: My guest is Dr. Kari Nadeau. She directs the Stanford Alliance for Food Allergy Research. We'll talk more after a break. This is FRESH AIR.

(SOUNDBITE OF MUSIC)

GROSS: This is FRESH AIR. Let's get back to our interview about children's food allergies with Dr. Kari Nadeau, director of the Stanford Alliance for Food Allergy Research. Nearly one in 10 preschoolers has a food allergy and the rate of children with food allergies has more than doubled over the past decade. So what are some of the theories about why the number of children with food allergies is growing?

NADEAU: We have been looking at why this increase in incidence and prevalence has occurred across the United States. There are approximately 15 million Americans with food allergy. One in 13 under the age of 18 has a food allergy diagnosis. And not only are we seeing an incidence in prevalence but we're also seeing an increase in emergency room visits. For example, there's about 90,000 per year that were recently published.

And the costs are increasing as well. About 500 million is spent per year on the cost for food allergy ER visits and for all those other medical needs. So with that in mind, people are working hard at trying to understand the causes. We think right now that it's probably multi-factorial. There's no easy answer, as you would probably expect.

Studies funded by FARE and NIH have shown so far that there's probably a gene environment interaction so that there's something about inheriting your allergies. If a person who has an allergic family, their child might have a food allergy, for example. But that's not necessarily always the case. There are families in which they have no allergies in the family and for the first time they have a child and that child has food allergy.

There also seem to be interactions with the environment. We're doing research through the Children's Environment Health Centers and the NIH about the role of tobacco and pollution as well as diet. And recent studies have shown that perhaps a more Mediterranean-type diet, less trans-fatty sort of fast food type diet might actually help in decreasing the risk of food allergies. But that's really over the interlay of genetic factors as well.

So it's an excellent question. More research needs to occur to understand why food allergies are increasing in our population and then really to try to find out how to better diagnose them so that we can prevent them and to perhaps, if they are in our control, change behaviors.

GROSS: So when you're talking about genetics being an issue for food allergies or it possibly being an issue, do you mean the genes passed on by the parents? Or do you mean some kind of change in the genetic structure of the child after the child is born?

NADEAU: Right.

GROSS: Because there's a lot of research into that now about how certain, you know, toxins in the air or toxins in food, toxins you inhale, can perhaps make slight modifications of the genes that have major repercussions.

NADEAU: That's exactly right. There are genes that we inherit from our family and our grandparents and those are genes that are ingrained in the sequence and that's where we inherit certain things like maybe perhaps our height, our color of our eyes, for example. And then there are things like atopy or allergies that can be passed from the family. And that is ingrained in the genes.

However, there are also what you're mentioning - how the environment can affect the genes and how the environment can modify the genes itself. And that field is called epigenetics. And what we're finding is through work, not only of our own, but also through other groups sponsored by the NIH and other organizations, that pollution and tobacco smoke and diet can affect the genes themselves on a chemical basis.

And that can be inherited down through to the children as well as the grandchildren. So there's a combination of genes as well as this epigenetic or how the environment can affect the genes. And we know that those two things play an important role in any one individual. For example, we've published data now out of Stanford in which, through people that received immunotherapy for grass pollens, for example, that their genes changed over time during the therapy so that they became modified to a non-allergic state. Which is very exciting because what you would then take the next step in thinking is that perhaps the children of the people that received immunotherapy might be able to have those same non-allergic type genes passed on to them. So we don't know that yet. We still have to do research.

GROSS: Dr. Nadeau, I wish you good luck with your research and thank you for sharing some of your research with us. Much appreciated.

NADEAU: Thank you. Thank you so much for the opportunity. It's wonderful to talk to you.

GROSS: Dr. Kari Nadeau directs the Stanford Alliance for Food Allergy Research at the Stanford University School of Medicine. To find out more about her research, you can go to our website freshair.npr.org, where you'll find a link to her lab's website and a link to a recent New York Times magazine article about her. That's freshair.npr.org, where you can also download podcasts of our show.

Transcripts are created on a rush deadline, and accuracy and availability may vary. This text may not be in its final form and may be updated or revised in the future. Please be aware that the authoritative record of Fresh Air interviews and reviews are the audio recordings of each segment.