Chapter 3

How NDE Research Works — Methodology, Standards, and the Difference Between Good and Bad Evidence

When critics dismiss near-death experiences, the word that comes up most often is unscientific. NDEs are just stories, the argument goes. Personal testimonies. Anecdotes. Wishful thinking dressed up in medical language. We can't take them seriously, the critics say, because they don't pass scientific muster.

I understand why people say this. If you've only met NDEs through books at the airport bookstore — Heaven Is for Real, 90 Minutes in Heaven, Proof of Heaven — you might assume the whole field rests on personal stories that no one has actually checked. But that picture is wildly incomplete. Once we look at what NDE research actually involves, the "unscientific" charge collapses.

NDE research has been done by cardiologists, oncologists, psychiatrists, professors of medicine, and PhD researchers working in major hospitals. It has been published in The Lancet, Resuscitation, Critical Care, the Journal of the American Medical Association, and dozens of other peer-reviewed journals. As J. Steve Miller and others have shown, well over nine hundred scholarly articles on NDEs were published before 2011 alone.¹ That is not pseudoscience. That is a serious body of medical literature.

Still, here is an important truth: not all NDE evidence is created equal. Some of it is rock solid. Some is suggestive but limited. Some should be set aside entirely. The job of this chapter is to give you the tools to tell the difference. By the end, you should be able to read any NDE book or news article and ask the right questions: Where did this evidence come from? Who collected it? How was it tested? Can it be checked?

I'll walk you through the main tiers of NDE evidence, from strongest to weakest. Then I'll introduce the special category that changes the whole conversation — veridical evidence — and briefly preview the naturalistic objections we'll engage in detail later in the book.

Tier One: Prospective Hospital Studies

The strongest tier of NDE evidence is the prospective hospital study. The word prospective is just a fancy way of saying "set up in advance." Researchers decide what they're looking for before patients have their NDEs. They identify cardiac arrest patients as they come into the hospital, interview them shortly after resuscitation, and compare their experiences with control groups of patients who survived cardiac arrest but did not report an NDE.

Let me introduce four of the most important prospective studies.

Van Lommel's Lancet Study (2001)

Cardiologist Pim van Lommel and his colleagues studied 344 consecutive cardiac arrest patients in ten Dutch hospitals over thirteen years. Of these, 62 patients (18 percent) reported NDEs. The study was published in The Lancet — one of the most prestigious medical journals in the world.² This was a watershed moment. The Lancet doesn't publish folk wisdom. The fact that a top peer-reviewed journal accepted a careful, controlled study of NDEs forced the academic community to take the phenomenon seriously.

Van Lommel's findings struck a heavy blow against the dying-brain hypothesis. If NDEs were simply caused by oxygen loss in dying neurons, every cardiac arrest survivor should have one — every one of them experienced the same brain ischemia (loss of blood flow to brain tissue). Yet only 18 percent did. And those who did couldn't be told apart from those who didn't by medication, length of arrest, or oxygen levels. We'll explore the implications more deeply in Appendix B. Methodologically, the point is this: a prospective, controlled, peer-reviewed study with a real sample size found NDEs to be common, consistent, and unexplained by ordinary brain physiology.

Parnia's AWARE Studies

Dr. Sam Parnia, a critical care physician now at NYU Langone Health, designed AWARE (AWAreness during REsuscitation) as the largest prospective NDE study ever conducted. AWARE I, published in Resuscitation in 2014, spanned fifteen hospitals across the United States, the United Kingdom, and Austria.³ AWARE II followed.

What makes the AWARE design clever is its use of visual targets. Researchers placed images on shelves in cardiac arrest rooms — facing the ceiling, so they could only be seen from above. The logic is simple. If NDErs really do float out of their bodies and observe the room from the ceiling, they should sometimes report seeing these targets. If they don't actually leave their bodies, they shouldn't be able to describe images that no one in the room could see from a normal viewpoint.

Honestly, the visual-target results from AWARE I were modest. Many cardiac arrests didn't happen in rooms with targets. Most patients didn't survive long enough to be interviewed. And of those who did survive, most didn't report full NDEs. But AWARE I did find something striking: 39 percent of cardiac arrest survivors who could be interviewed reported some form of awareness during the time they were clinically dead — even though only a smaller fraction had a complete NDE meeting Greyson's criteria. The famous AWARE denture case (treated in Chapter 6) showed a patient accurately describing events during his resuscitation that he could not have witnessed by ordinary means.

Sabom's Atlanta Study

Cardiologist Michael Sabom — a committed Christian and PCA elder — first researched NDEs in the late 1970s and published Recollections of Death in 1982. Then in 1994 he launched the Atlanta Study to extend his research. He interviewed 160 patients over two years. Forty-seven of them had survived clearly defined near-death experiences linked to a real medical crisis (cardiac arrest, shock, accident, or surgery).⁴

Sabom's methodology was rigorous. He required each NDE in his study to meet Bruce Greyson's NDE Scale — a 16-question instrument scoring experiences from 0 to 32, with a minimum score of 7 needed to qualify as a true NDE.⁵ He deliberately avoided members of the International Association for Near-Death Studies, since IANDS members tend to read the same books, attend the same conferences, and influence each other's beliefs. He wanted "virgin cases" — people who hadn't told their stories publicly. He created a Spiritual Beliefs Questionnaire to distinguish traditional Christians from liberal Christians, God-believers, and atheists. And he used two separate control groups: 32 open-heart surgery patients and 81 random cardiac patients who had not had NDEs.⁶

What made the Atlanta Study so important — and what we'll return to throughout the book — was its Christian-friendly findings: NDEs in his sample did not lead Christians away from traditional faith. They deepened it.

Sartori's Welsh Study

Dr. Penny Sartori, a critical-care nurse, conducted a five-year prospective study at Morriston Hospital in Swansea, Wales. Sartori interviewed cardiac arrest patients and tested what she called the "good guesses hypothesis" — the idea that NDErs only seem to give accurate descriptions of their resuscitations because they're guessing well based on hospital TV dramas. To test this, she asked seasoned cardiac patients who had not claimed to leave their bodies to describe what happened during their resuscitations. Their guesses were riddled with errors — wrong equipment, wrong procedures, scenarios pulled straight from medical television. The NDErs' descriptions, by contrast, were strikingly accurate.⁷ Sabom had found the same pattern years earlier in his original work, comparing NDErs' accounts of their own resuscitations against the medical records.⁸

These four studies — Van Lommel, Parnia, Sabom, Sartori — represent the gold standard. Prospective. Controlled. Peer-reviewed. Conducted by medical professionals using standardized tools. This is not pseudoscience. This is medicine.

Tier Two: Retrospective Studies

The second tier is retrospective research. Here, instead of identifying patients in advance, researchers interview people who reported NDEs at some point in their past — sometimes years or decades earlier. Raymond Moody's pioneering work in Life After Life (1975) was retrospective. Kenneth Ring's foundational studies in the 1980s were retrospective. Bruce Greyson's decades of data collection at the University of Virginia largely fall in this category.

Retrospective research has real strengths. It allows researchers to study NDEs in depth, gathering rich qualitative data. Long, detailed interviews can explore aspects of the experience that a quick post-resuscitation interview might miss. Many of the foundational insights in the field — the core features of the NDE, the long-term aftereffects, the cross-cultural patterns — came from retrospective work.

But this tier comes with weaknesses we should name honestly. Memory shifts over time, even for vivid experiences. People who had dramatic NDEs are more likely to volunteer for studies than people who had mild or distressing ones, creating self-selection bias. And the longer the gap between event and interview, the more the experiencer's interpretation may have hardened into a polished story. A 40-year-old account is not the same as a fresh one.

I want to be careful here. Retrospective evidence isn't worthless — far from it. Greyson's standardized scale, applied across decades, has produced extraordinarily consistent results. But retrospective findings are best understood as complementary to the prospective studies, not a substitute for them.

Tier Three: Self-Reported Online Databases

The third tier deserves special attention because it is both massive and often misunderstood. The Near-Death Experience Research Foundation (NDERF), founded by oncologist Jeffrey Long and his wife Jody Long, is the largest collection of NDE accounts in the world — over 5,000 cases gathered through a detailed online questionnaire that asks more than a hundred questions.⁹ The website also offers multiple language translations, drawing experiences from across the globe.

J. Steve Miller has done the most careful analysis of NDERF's strengths and weaknesses. Let me summarize his findings.

Strengths. First, NDERF includes a wide range of medical conditions — not just cardiac arrest but drownings, car wrecks, surgeries, suicides, combat injuries, and serious illnesses. Second, the researcher gets to see entire accounts, not just the snippets a physician might select for a published study. Third, the sample size allows both broad statistics and detailed subgroup analysis — for example, isolating accounts from Hindus, Muslims, or atheists. Fourth — and this is Miller's methodological breakthrough — accounts can be analyzed in the order they were uploaded, which removes the worry of cherry-picking. Miller read 100 consecutively uploaded NDEs to test whether NDEs really teach New Age doctrines (as some critics claimed). They didn't. By taking experiences in upload order, his findings couldn't be dismissed as confirmation bias.¹⁰

Weaknesses. First, there is no direct medical verification of the near-death event — the researcher trusts the submitter's account. Second, fabrication is possible. Third, there is no control group. Fourth, experiences are self-submitted rather than collected by trained researchers in a clinical setting.

Miller's response to these limits is wise. He developed a "Gray Scale" of red flags to filter out questionable accounts: the person wasn't actually close to death, signs of mental illness, drug use, second-hand reports, sales motives, false memories, vague or dreamlike content, low Greyson scores, and so on.¹¹ When applied carefully, NDERF becomes a powerful research tool — not a substitute for prospective studies, but a complement to them, especially for questions that require very large sample sizes.

Tier Four: Popular Testimonial Books

The bottom tier — and where most public confusion comes from — is the popular testimonial book. Heaven Is for Real. 90 Minutes in Heaven. Proof of Heaven. Dozens of others. Some are written by competent physicians with documented medical histories. Others are mass-market trade books with little real verification.

A cautionary tale: Alex Malarkey was the boy at the center of The Boy Who Came Back from Heaven, written by his father Kevin Malarkey. The book sold over a million copies. Then Alex — old enough by then to speak for himself — issued a public statement admitting that he had made the story up, and the publisher pulled the book from shelves.¹² The lesson is hard to miss. Profit motives can corrupt witness. Family members can mislead. Even sincere belief is no substitute for verification.

A more delicate case is Eben Alexander's Proof of Heaven. Alexander is a neurosurgeon whose NDE during bacterial meningitis became a cultural phenomenon, and Shaun Tabatt rightly notes that the book is "extremely influential."¹³ But it has serious theological problems that conservative Christians need to take seriously.

Here is what I want you to take from this. Popular testimonial books are not the primary basis on which biblical Christians should evaluate NDEs. The professional research literature is far more reliable. When critics like John MacArthur and Hank Hanegraaff dismiss NDEs based largely on popular books like Alexander's, their concerns are partly valid — those books really do contain significant theological error. But it is a mistake to let bad popular books discredit the careful medical research. As Miller wisely notes, MacArthur and Hanegraaff were overwhelmed by questionable theology in popular books and underwhelmed by the actual evidence — and they failed to engage the prospective hospital studies that should have shaped their conclusions.¹⁶ We'll return to this in Chapters 31–33.

The Critical Category: Veridical Evidence

Now let me introduce the category that, in my judgment, changes the whole conversation. A veridical NDE is one in which the experiencer reports specific, verifiable information about events occurring during their clinical death — information they could not have known through normal sensory channels.

Think about what that means. If a patient with eyes taped shut and a flatlined EEG accurately describes the surgical instruments used during her procedure — instruments she had never seen and could not have seen — then the hallucination hypothesis is in deep trouble. Hallucinations don't produce accurate information about the external world. A dying brain can't perceive what no living brain in the room can perceive from a normal viewpoint.

The well-known cases — Maria's tennis shoe (Kimberly Clark Sharp), the AWARE denture case (Pim van Lommel), Sabom's cardiac arrest patients, Pam Reynolds — are explored in detail in Chapters 6 and 7. Here, the methodological point is simply this: veridical NDEs constitute a distinct evidential category, and they raise the bar far higher than popular skeptics ever address.

A Quick Word on Naturalistic Explanations

Skeptics have offered many naturalistic explanations for NDEs over the years: oxygen deprivation (cerebral anoxia), DMT release, temporal lobe stimulation, REM intrusion, expectation and cultural conditioning, false memory, residual brain activity, and more. Each of these will be addressed comprehensively in Appendix B. Here, I just want to flag the central problem they all share.

None of them can explain veridical information. A hallucination — whatever its biological cause — cannot tell you what color shoe is on a third-floor window ledge that you've never seen. A burst of DMT cannot generate accurate descriptions of conversations happening in a waiting room three floors below the operating theater. Whatever else NDEs may involve, they involve genuine perception of real events at moments when the brain should not be capable of perception at all. That fact, more than any other, is why I find the NDE evidence compelling.

Conclusion: Reading the Evidence Wisely

So here is where we land. NDE research is not pseudoscience. It is published in peer-reviewed medical journals, conducted by competent physicians and PhD researchers, and built on decades of careful methodology. But not every NDE source is equally reliable. The strongest evidence comes from prospective hospital studies — Van Lommel, Parnia, Sabom, Sartori. Retrospective studies and large self-reported databases like NDERF provide essential complementary data, especially when researchers like Miller use consecutive-upload methodology to guard against bias. Popular testimonial books should be read with caution and never substituted for the professional literature.

The category that towers above all others is veridical evidence. When patients report verifiable details about events they could not have witnessed by ordinary means, we are looking at something hallucination cannot explain. As a Christian, I think we have nothing to fear from this evidence. We have everything to gain by engaging it carefully — and a great deal to lose by dismissing it without ever reading the prospective studies.

In the next chapter we will turn to a closely related category: deathbed experiences, those visions of heaven and reunions with deceased loved ones that occur in the dying process itself. The methodological habits we've built in this chapter — asking where the evidence comes from, who collected it, how it was tested, and whether it can be checked — will serve us well there too.