Chapter 12

The Hospitals Weigh In

“For God speaks in one way, and in two, though man does not perceive it. In a dream, in a vision of the night, when deep sleep falls on men… then he opens the ears of men.” — Job 33:14–16 (ESV)

A. The Question on the Table

Picture a cardiac care unit at three in the morning. The lights are low. Monitors beep in a steady rhythm. A nurse sits at a station, watching a row of green and amber lines scroll across her screen. Down the hall, a retired teacher lies sleeping after a mild heart attack.

Then the alarm sounds. The green line goes flat. Within seconds, a team rushes in with a crash cart. Chest compressions begin. A defibrillator charges. For the next several minutes, that teacher has no heartbeat. His brain is receiving no blood. By every clinical measure we know, he is not conscious.

And yet, when he wakes hours later, he tells the nurse exactly what she was doing at her station while the team worked on him. He describes a conversation she had on the phone. He names the color of a clipboard she picked up. None of this should be possible.

Stories like this one are the reason NDE research moved into hospitals. For decades after Raymond Moody first named the near-death experience in 1975, critics raised a fair point: the stories were gathered after the fact, from people who came forward on their own. Maybe only the dramatic cases got told. Maybe memory had polished the details. Maybe the whole thing was selection bias dressed up as evidence.¹

Yet some of the most compelling accounts were reported to physicians in hospital rooms, not years later. J. Steve Miller recounts a case described by Dr. Tom Aufderheide, now a professor of emergency medicine at the Medical College of Wisconsin. Early in his career, in 1980, a patient who had suffered repeated cardiac arrests recovered against all odds. Before leaving the hospital, the man described the events of that night in remarkable detail. He recounted conversations that had taken place outside the room and even the young doctor's private thoughts, all from a vantage point outside his body. Aufderheide could find no explanation for how the patient could have known these things.⁵⁵

The critics had a point. If you only hear from the people who want to talk, you do not know how common these experiences really are. You do not know whether the quiet majority had nothing to report, or simply chose not to speak. And you cannot rule out the possibility that time and retelling reshaped what people remember.

So researchers did something bold. They went into hospitals, enrolled patients before anything happened to them, and waited. When a patient's heart stopped and was restarted, the researchers asked a simple question within days: Do you remember anything from the time you were unconscious?²

The results of those studies are what this chapter is about. They did not settle every question. But they changed the conversation forever. They gave us real numbers, careful controls, and, in a few remarkable cases, verified accounts of awareness during the very minutes when the brain should have been dark.

B. Moving Research Into the Hospital

To understand why prospective hospital studies matter so much, you need to see the gap they filled.

Most early NDE research was retrospective. That means it looked backward. A researcher would place an advertisement, or a patient would contact a university, and the interview would take place months or years after the event. This approach gave us rich, detailed accounts. It also left the door wide open for two problems. The first was selection bias: only the most vivid or the most willing came forward. The second was memory distortion: time changes how we remember things.³

A prospective study works differently. It looks forward. Researchers pick a hospital, identify the units where cardiac arrests are most likely to happen, and set up a system to interview every survivor, not just the ones who volunteer. Medical records are kept in real time. Interviews happen within days of the event, while memory is still fresh.⁴

The design also creates something just as valuable: a built-in control group. Most survivors of cardiac arrest remember nothing at all. That large group of non-experiencers becomes the natural comparison. Were they younger or older? Sicker or healthier? On different medications? By comparing the two groups, researchers can test whether any medical factor explains who has an NDE and who does not.

Cardiac arrest is the ideal setting for this kind of study, because the medical facts are so clear. When the heart stops, blood flow to the brain ceases within seconds. Consciousness is lost almost immediately. The EEG, which measures electrical activity in the brain, typically goes flat within ten to twenty seconds.⁵ If a patient later reports vivid, structured, lucid experiences from that window of time, we have a sharp problem on our hands. The brain was in no condition to produce them.

As Chris Carter puts it, several of the most careful NDE studies have used cardiac arrest for exactly this reason. Physicians Sam Parnia and Peter Fenwick have noted that cardiac-arrest survivors are the closest model we have to a dying brain. These patients meet two of the three standard criteria used to pronounce a person dead: no cardiac output and no breathing. In the clinical setting of an arrest, they usually develop the third as well: fixed, dilated pupils from loss of brain-stem activity.⁶

Even before these prospective studies began, some researchers were already doing careful work inside hospitals. Cardiologist Michael Sabom was one of them. In the late 1970s and early 1980s, Sabom interviewed cardiac-arrest survivors at hospitals in Atlanta. He compared what they said they saw during their resuscitation to what actually happened in the room. He then compared their reports to a control group of cardiac patients who had not had an NDE but were asked to guess what a typical resuscitation looks like. The NDE group gave far more accurate and specific accounts than the guessers. His findings raised the real possibility that perception was happening apart from the body.⁴⁸

Sabom's work showed what hospital research could do. When medical records and staff accounts are right there to check, a patient's claim can be tested in a way that no retrospective interview ever could. His Atlanta Study patients described specific medical instruments, staff movements, and conversations that matched the records. That kind of controlled verification planted the seed for everything that followed.⁴⁹

This is why the prospective cardiac-arrest studies matter. They take the strongest objections seriously: bias, memory, brain activity. And they still find the experiences.

C. The Heart of the Case

1. Why Prospective Studies Changed the Game

Before the hospital studies, the debate about NDEs had a frustrating quality. Believers pointed to powerful stories. Skeptics pointed to weak methods. Both sides were partly right.

The strength of a prospective study is that it removes the storyteller from the equation. Researchers do not wait for someone to come forward with a tale of bright light and deceased relatives. They go to every survivor, including the ones who had no experience at all, and ask the same questions in the same way. The result is a true picture of how common these experiences are, not a highlight reel of the best ones.⁷

Between 2001 and 2023, four major prospective studies were published in peer-reviewed medical journals. All four used nearly the same design. All four studied survivors of cardiac arrest in hospitals. And all four found the same basic result: a real and consistent percentage of survivors reported experiences during the time they were clinically dead.⁸

That consistency matters. When four independent teams, working in different countries and different decades, land on similar numbers, the finding is not a fluke. It is a pattern.

2. Van Lommel’s Landmark Study

The first and largest of the early prospective studies was led by Dutch cardiologist Pim van Lommel. Published in The Lancet in December 2001, it immediately drew international attention, because The Lancet is one of the most respected medical journals in the world.⁹

Van Lommel and his team enrolled 344 consecutive patients who survived cardiac arrest across ten Dutch hospitals over a four-year period (1988–1992). Every one of these patients had been clinically dead. They would have died permanently if they had not been resuscitated within five to ten minutes. Within days of their resuscitation, researchers asked each one whether they remembered anything from the time they were unconscious.¹⁰

The results were striking. Of the 344 patients, 282 (82 percent) had no memory at all. But 62 patients, 18 percent, reported some kind of experience. Of those, 41 described a core NDE with multiple recognizable features: feelings of peace, a tunnel, a bright light, meeting deceased loved ones, or a life review.¹¹

What made the study especially powerful was what the researchers did not find. They looked for medical explanations. They compared the NDE group to the non-NDE group on every factor they could measure: how long the heart was stopped, how long the patient was unconscious, which drugs were given, and how severe the crisis was. Some patients had been unconscious for five minutes. Others had been in a coma for three weeks. None of these factors explained the difference. Medication played no role. Duration of the arrest played no role. Whether the patient was stopped for two minutes or eight minutes made no difference.¹²

One case from van Lommel's study became famous in the NDE literature. During CPR, a nurse removed a patient's dentures and placed them in a specific drawer so she could insert a breathing tube. The patient was unconscious the entire time. His heart had stopped, and he was being resuscitated for over an hour and a half. A week later, when the man was transferred to a general ward, he recognized the nurse, even though he had been clinically dead throughout their encounter. He told her exactly where his dentures had been placed. He later told van Lommel that during the cardiac arrest he had been floating near the ceiling. He had watched the team work on his body. He was afraid they might give up on him.¹³ This case was later investigated thoroughly and found to be accurate.

The study also followed survivors for years afterward. At two-year and eight-year follow-ups, the NDE group showed lasting changes that the non-NDE group did not. They had a sharply reduced fear of death. They showed a stronger belief in an afterlife. They reported a deeper interest in the meaning of life and greater love for other people. These were not fading memories. They were lasting transformations. And they showed up only in the group that had an NDE, not in the control group of survivors who had the same medical crisis without the experience.¹⁴

3. Parnia’s AWARE Studies

If van Lommel opened the door, Sam Parnia walked through it with a flashlight and a measuring tape.

Parnia, a critical care physician and researcher, has spent more than two decades studying what happens to the mind when the heart stops. His early pilot study in the United Kingdom, published in 2001, examined 63 cardiac-arrest survivors. Eleven percent reported an NDE, with 6.3 percent describing a core experience and 4.8 percent a shallower one.¹⁵

But it was his AWARE studies that pushed the research to a new level.

AWARE I (AWAreness during REsuscitation) ran from 2008 to 2012 across fifteen hospitals in the United Kingdom and the United States. The study tracked 2,060 cardiac-arrest events. The vast majority of those patients died. Only 101 survived long enough to be fully interviewed.¹⁶

Of those 101 survivors, 38 percent had some vague sense of awareness during their arrest, though they could not recall specific details. About 9 percent had features of what Parnia calls a "recalled experience of death," the structured, vivid experience that matches the classic NDE pattern. And 2 percent, two individuals, actually recalled seeing and hearing specific details of their own resuscitation from a point above their bodies.¹⁷

"Importantly, we were able to verify that his period of conscious awareness had happened when his heart was not beating and had lasted at least three to five minutes. This was very significant, as it was the first time that a study had identified a case of conscious awareness—in support of what so many people had claimed to have experienced—while there was no heartbeat." — Sam Parnia, Lucid Dying

One of those two patients, a fifty-seven-year-old man, gave an account that could be checked against the medical record. He described floating above his own body. He saw the medical team around him. He recalled being given two electrical shocks to his heart. He heard a nurse named Sarah speaking. He heard the staff say, "We've got him back." All of this was confirmed. And all of it happened while his heart was not beating.¹⁸

AWARE II, published in the journal Resuscitation in September 2023, was even more ambitious. It was the largest study of its kind ever conducted. A team of thirty-three scientists across twenty-five medical centers followed 567 cardiac-arrest patients. For the first time, researchers attached brain-monitoring equipment to patients during active resuscitation, measuring brain oxygen levels and electrical signals second by second.²⁰

AWARE II found that about one in five survivors had features of a recalled experience of death. But the study went further. Nearly 40 percent also had vague perceptions of awareness without clear recall. This supported what Parnia had long suspected: more people may have conscious experiences at the edge of death than are able to remember them afterward.²¹

The survivors' experiences, Parnia found, followed a specific pattern. They were not random or confused. They included a sense of separation from the body, observation of events without pain or distress, and a purposeful review of one's life. This was followed by the sense of going to a place that felt like home, and then returning to the body. This structured arc appeared again and again, across different patients, different hospitals, and different countries.²²

4. Greyson’s Decades of Hospital Research

Bruce Greyson has been studying NDEs longer than almost anyone alive. A psychiatrist at the University of Virginia, he has spent decades interviewing patients in cardiac care units and emergency rooms. He developed the Greyson NDE Scale, the standard research tool used worldwide to measure and compare near-death experiences.²³

In 2003, Greyson published his own prospective study of cardiac-arrest survivors at a cardiac care unit. He interviewed nearly 1,600 patients admitted to the unit over a thirty-month period. Of those whose hearts had stopped, about 15.5 percent reported an NDE.²⁴

That number sits right in the range found by the other studies. Van Lommel found 18 percent. Parnia's early study found 11 percent. Penny Sartori, a British nurse-researcher who ran her own five-year prospective study, found 23 percent.²⁵ The numbers vary, as they always will when different teams study different populations. But the range has been remarkably stable: between 10 and 23 percent of cardiac-arrest survivors report an NDE when asked soon after the event.

What Greyson added over the years was depth. He tracked how people changed after their experiences. He studied their reluctance to talk about it. Many patients kept their NDEs secret for years, afraid they would be thought crazy or attention-seeking. Greyson found that this silence was one of the biggest reasons the true frequency of NDEs had been underestimated for so long.²⁶

He also noticed something else. The experiences his patients reported in the 2000s matched, feature for feature, the experiences that earlier researchers had documented in the 1970s and 1980s. The tunnel. The light. The deceased relatives. The life review. The reluctant return. Decades of medical progress, cultural change, and media coverage had not altered the core pattern.²⁷

This consistency fascinated Greyson. If NDEs were shaped mainly by culture or expectation, you would predict that they would change over time. Movies and books would spread new ideas about what dying is supposed to look like. But the experiences did not change. The experiences that patients reported to Moody in the early 1970s, before any book had made the subject popular, are the same experiences that patients report today. In my own dissertation, I confirmed this finding across seven decades of data. The core features stayed remarkably stable before and after 1975, the year Moody's Life After Life was published.

Greyson also brought careful attention to the quality of the medical documentation. In a cardiac care unit, every heartbeat is monitored. Every intervention is recorded with a timestamp. When a patient claims to have seen or heard something during a cardiac arrest, the medical records are right there to check. This kind of verification is simply not possible with retrospective stories gathered years after the event.

5. Prevalence and the Underreporting Puzzle

Let me be honest about the numbers. The prospective studies tell us that somewhere between 10 and 20 percent of cardiac-arrest survivors report an NDE when asked soon after the event. That is a real and consistent finding. But it also means that the large majority, 80 to 90 percent, report nothing at all.

Does that mean only a small minority actually have an experience?

Not necessarily. There are strong reasons to think the true rate is higher than the recall rate. Cardiac arrest is an extreme assault on the brain. When the heart stops, blood flow ceases, and the brain is starved of oxygen. When the heart restarts, blood rushes back in, causing what doctors call reperfusion injury. This flooding of oxygen into damaged tissue triggers massive inflammation. On top of that, survivors typically spend days or weeks in intensive care, heavily sedated.²⁸

All of this destroys memory. Memory is one of the most fragile functions of the brain. It is the first thing to fail after a brain injury, and the last to return. The medical term for this is amnesia, and it is expected after any serious disruption to the brain. Parnia has argued that many people may have conscious experiences during cardiac arrest but lose the ability to recall them afterward. It is the same way we lose most of our dreams upon waking.²⁹

AWARE II supported this idea directly. While about 20 percent of survivors had clear features of a recalled experience, nearly 40 percent had vague perceptions of awareness, a sense that something had happened without the ability to bring it into focus. The experience was there. The memory was not.³⁰

Population surveys add another layer. Gallup polls and other surveys suggest that about 4 percent of the total population in western countries reports having had an NDE at some point in their life. That comes to roughly nine million people in the United States, two million in the United Kingdom, and twenty million across Europe.³¹

Silence plays a role too. Greyson and others have documented how many NDErs keep their experiences private for years. Some fear ridicule. Some worry their doctors will think they are mentally ill. Others simply cannot find the words. This means that even the prospective studies, which ask every survivor, may still be missing people who are not yet ready to talk.³²

6. The Correlation That Turns the Tables

Here is where I need to bring in my own research.

In my doctoral dissertation, I analyzed 5,278 near-death experiences, scoring each one for the strength of its evidence that consciousness was working apart from the brain. What I found was a pattern that should not exist if NDEs are simply the product of a failing brain.³³

I found a positive correlation between the severity of the medical crisis and the quality of the verified perception. In plain words: the worse the crisis, the stronger and clearer the experience. The highest evidence scores appeared in cases where the patient was in documented cardiac arrest, often with EEG monitoring confirming a flat line. The experiences that produced the most accurate, confirmed observations of the outside world came precisely when the brain was most impaired.³⁴

On a dying-brain view, this is backwards. If NDEs were caused by a brain shutting down, we would expect the opposite. Random neuron firing, oxygen starvation, or any of the other explanations we will look at in Chapter 24 should produce worse perceptions, not better ones. We would expect the experience to be more confused and less accurate as the brain deteriorates. A brain that is losing oxygen and blood does not produce better perceptions. It produces worse ones. Confusion. Disorganization. Fragmented images. That is what brain damage does to consciousness.³⁵

But the NDE data go the other way. The deepest crises yield the clearest experiences. The most impaired brains produce the most lucid awareness. And the most accurate reports of what was happening in the room come from the patients who, by every measure, should have been the least able to perceive anything at all.

Parnia and Fenwick made this same observation in their clinical work. They wrote that the occurrence of lucid, well-structured thought processes during a cardiac arrest raises perplexing questions. Reasoning, attention, and memory recall of specific events all appear at a time when brain function can be described at best as severely impaired, and at worst absent.³⁶

I do not think this finding proves that the soul leaves the body. That is a theological claim, and the data alone cannot carry it. But I do think it does something nearly as important. It makes the dying-brain explanation very difficult to maintain. And it points, strongly, in the direction this book has been building: that consciousness can work apart from the brain, because it does not come from the brain alone.

D. What the Researchers and Scholars Say

The scholars who have done this work do not all agree on what it means. But they agree on what it shows.

Sam Parnia, speaking of the AWARE II findings, has argued that the recalled experiences of death are not dreams, not hallucinations, and not products of a disordered brain. His team used a scientific method called grounded theory to analyze survivors' testimonies, and found that the experiences follow a structured narrative arc, not a random jumble of images. They are, in his words, "a unique and real experience that emerges with death."³⁷

Bruce Greyson has stated plainly that no one physiological or psychological model can by itself explain all the common features of an NDE. He has called it "paradoxical" that heightened, lucid awareness and logical thought processes occur during a period of impaired blood flow to the brain. This, he says, raises "particular perplexing questions for our current understanding of consciousness and its relation to brain function."³⁸

Van Lommel, reflecting on his own study, went further. He concluded that the concept that consciousness and memories are located in the brain, assumed but never proven, needed to be reexamined. He asked how a clear consciousness outside one's body could be experienced at the moment when the brain no longer functions during clinical death, with a flat EEG.³⁹

Chris Carter, surveying the full body of evidence, argued that the reports of enhanced mental processes and accurate perception of the environment at a time when brain activity should be absent quite clearly challenge the idea that consciousness is produced entirely by the brain.⁴⁰

Not all researchers draw the same conclusions from these findings. Some, like Kevin Nelson, believe that the experiences can be explained by the brain's arousal system, particularly REM sleep intrusion during the crisis. Others suspect that tiny pockets of brain activity may persist even when the EEG reads flat. These are fair questions, and I will take them seriously in Chapter 24, where we examine the naturalistic explanations one by one.⁴¹

Michael Sabom, the cardiologist who pioneered hospital-based NDE research, came to his own work as a skeptic. He heard about NDEs at a Sunday school class and did not believe a word of it. But when he began asking his own cardiac patients, he was startled. One man described the defibrillation paddles, the staff positions, and even the number of shocks given to his heart, all confirmed by the medical record. Sabom eventually concluded that something was happening during these arrests that existing medical science could not explain. His years of bedside research convinced him that the near-death experience deserved serious scientific attention, and his controlled comparisons remain among the most careful in the literature.⁵¹

Not every important voice in this conversation is a formal researcher. Laurin Bellg, a critical-care physician, spent years at the bedsides of dying patients in hospital ICUs. She did not set out to study NDEs. But patient after patient confided in her, often tearfully, afraid that their doctor would think they were losing their mind. What struck Bellg was the consistency of their accounts and the depth of the changes these experiences produced in the people who had them. Her observations, recorded in Near Death in the ICU, add the voice of a frontline physician who came to take these experiences seriously. She did so because she kept seeing them in her own patients, case after case, year after year.⁵²

J. Steve Miller, in his survey of the evidence for NDEs, took a different approach. He walked through each naturalistic explanation, measured it against the data, and asked a simple question: which theory best fits the facts? His conclusion matched that of Carter, Parnia, and Greyson: no purely brain-based explanation accounts for the verified perceptions that hospital patients have reported during documented cardiac arrest.⁵⁴

But what all researchers agree on is that the prospective studies have established something that was only suspected before: NDEs are a real, consistent, reproducible phenomenon in cardiac-arrest survivors. They are not the invention of cranks or the product of selection bias. And they happen, by the patients' own reports, during the window when brain function is at its lowest.

E. Counter-Objections

The evidence is strong. But strong evidence still draws pushback, and some of it is worth hearing.

Objection 1: The percentages are modest, not universal. Ten to twenty percent is real, but it also means that 80 to 90 percent of cardiac-arrest survivors report nothing. Does this suggest the experience is unusual rather than meaningful? I think this objection underestimates the memory problem I described above. Cardiac arrest destroys memory. Many people may have experiences they simply cannot recall. The 10 to 20 percent figure is the floor, not the ceiling. AWARE II found that when you include vague perceptions of awareness, the number rises to nearly 60 percent.⁴²

Objection 2: Recall bias inflates or deflates the numbers. Could some survivors be adding details after the fact, turning a hazy feeling into a dramatic story? Or could others be suppressing real memories out of fear? Both are possible. The prospective design helps by interviewing survivors within days, before memory has time to develop or fade. But no study is perfect. Recall bias is a real limitation, acknowledged by every researcher in the field. It is one reason why the brain-monitoring component of AWARE II was so important. It does not rely on memory alone.⁴⁴

Objection 3: Publication bias favors positive results. Studies that find something interesting are more likely to get published than studies that find nothing. This is true across all of science, not just NDE research. But the prospective NDE studies are a partial answer to this problem. They are designed to report all results, positive and negative. The large majority of their patients reported nothing, and the researchers published that fact alongside the minority who did.⁴⁵

Objection 4: No one has ever reported the hidden shelf image. Critics point out that in both AWARE studies, no patient reported seeing the hidden visual targets placed on shelves near the ceiling. Doesn't this prove that people are not really floating above their bodies? Not quite. The math works against the test. Very few patients survive cardiac arrest. Fewer still report an NDE. Fewer still have the out-of-body component. And of those, the arrest must happen in a room with a target, and the patient's vantage point must include the shelf. In AWARE I, the two verified out-of-body cases both occurred in rooms where no target had been placed.⁴⁶ The failure to capture an image so far is a limitation of sample size, not a disproof of the experience.

F. What the Hospital Rooms Reveal

Let me gather the threads.

The prospective hospital studies have done something no amount of storytelling could do on its own. They have shown, under controlled conditions and with careful methods, that near-death experiences are a real and consistent feature of cardiac arrest. Between 10 and 20 percent of survivors report them when asked within days. The experiences follow a structured pattern. Some include verified perceptions of events that were confirmed by medical staff. And the strongest experiences tend to occur during the worst crises, not the mildest ones.

These are not the results that a dying-brain explanation predicts. A brain starved of oxygen and blood does not become sharper. It becomes duller. A brain in cardiac arrest does not organize vivid, logical, sequential narratives. It falls silent. Yet the patients tell us, again and again, that something was very much awake during the minutes when their brains were dark.

In my own research, I found the same pattern. The severity of the crisis and the clarity of the experience move together. That correlation, confirmed by hospital data from three countries and repeated across more than two decades of study, is a stubborn fact that every theory of consciousness must face.

I am not saying the hospital studies prove that the soul leaves the body. I am saying they make it very hard to insist that consciousness is nothing more than brain activity. Something is going on during cardiac arrest that our current models of the brain cannot explain. Whether that something is the soul stepping free, or a process we have not yet discovered, is a question that science alone cannot settle. But for those of us who already trust Scripture's teaching that we are more than a body (see Chapter 2 and Chapter 3), the hospital evidence is a powerful confirmation.

Think about what that means for the person you love who died last year, or the diagnosis you are carrying, or the fear that wakes you at 2 a.m. The same hospitals that treat our bodies and measure our brain waves are finding, year after year, that something inside us does not stop when the machines say we should be gone. The researchers do not all call it a soul. But the data they are collecting fits what the Bible has been telling us for thousands of years: you are more than your body, and the part of you that matters most does not die when your heart stops.

The psalmist wrote, "Where shall I go from your Spirit? Or where shall I flee from your presence? If I ascend to heaven, you are there! If I make my bed in Sheol, you are there!" (Psalm 139:7–8, ESV).⁴⁷ The God of Israel is present in the highest places and the lowest. He is present in the recovery room and in the moment the heart goes still. And He is present, these studies suggest, even when every medical instrument says we are gone.

"For God speaks in one way, and in two, though man does not perceive it. In a dream, in a vision of the night, when deep sleep falls on men, while they slumber on their beds, then he opens the ears of men and terrifies them with warnings, that he may turn man aside from his deed and conceal pride from a man; he keeps back his soul from the pit, his life from perishing by the sword." — Job 33:14–18 (ESV)

Job's friend Elihu knew what the hospital researchers are rediscovering: that God speaks to people in the deep places, at the edges of consciousness, and sometimes at the very border of death. The prospective studies have given us reason to believe that millions of people have heard that voice. What they heard, and how to weigh it, is the work of the chapters ahead.