Christopher Kerr says nearly 90 percent of dying patients see dead loved ones before death

The most persuasive ghost story on YouTube right now is not a ghost story at all, which is inconvenient for everyone who came dressed for ectoplasm. Christopher Kerr, a hospice doctor who says he has interviewed roughly 1,500 dying patients, told The Diary of a CEO that nearly 90 percent of people near death have intense dreams or visions, and the recurring cast is strikingly specific: dead loved ones.

The vast majority of people, nearly 90% of people start to have very intense dreams or visions. And the overriding theme is that they’re seeing people who they’ve loved and who they’ve lost. And very specifically, it’s those that have loved them unconditionally purely and those that didn’t aren’t there.

That is the headline. Not “death is mysterious,” which is what people say when they’ve run out of useful things to say at a wake. Kerr’s claim is more concrete and more unnerving: when people are close to dying, many of them report vivid encounters with the dead, and those encounters are often the thing that comforts them most.

Kerr is not a random TikTok medium wearing too many rings. He is a hospice physician with a PhD in neurobiology, and his whole posture is half clinician, half reluctant mystic. That matters. The episode works because he keeps refusing the cheap version of the story, then keeps walking right up to the expensive one anyway.

The dead arrive like hospice care, not horror

The examples Kerr brings are almost aggressively domestic. A man dreams repeatedly of his late wife in light blue. A woman near death sees her husband and daughter at a kitchen table, laughing, the years of separation briefly erased. No thunderclap. No pearly-gate production design. More like the universe has the taste level of a family photo album.

Kerr’s most interesting point is that the patients don’t experience these as symbolic riddles. They don’t wake up asking someone to decode the dog, the kitchen, the old perfume. They experience them as contact, or at least as something with the emotional force of contact.

the thing that gets them most distressed is when you refer to them as dreams. because these are unlike anything they’ve ever experienced. So they’re not dreams in the sense they’re not metaphorical. They’re not fantastical. They don’t require interpretation.

This is where the episode stops being a soothing deathbed anecdote and becomes a genuinely searchable argument. Kerr says these events become more common in the last weeks of life, grow more immersive as death approaches, and in about half of cases happen while patients are awake. He describes people looking through the room or petting a dog that is not physically there. Depending on your priors, that is either beautiful, terrifying, or a customer-service escalation for materialism.

The science is stronger than the afterlife claim

The cleanest read is this: Kerr has a strong case that end-of-life visions are real experiences reported by many dying people, and that hospice care should take them seriously. He has a weaker case, because everyone has a weaker case, that these experiences prove consciousness continues after death.

To his credit, Kerr doesn’t pretend he has smuggled heaven through peer review. When Steven Bartlett presses him on whether this means a wife is “truly accessible” or whether the brain is retrieving memory, Kerr hesitates. Good. A little hesitation is the price of admission here.

I don’t know what this means. I don’t know if this I I wouldn’t be so bold as to say this speaks to an afterlife. What I am more convinced of is that those who have loved us are never gone. They’re accessible to us.

That is both moving and slippery. “Accessible to us” can mean memory, attachment, neurochemistry, spirituality, or the kind of sentence people write in condolence cards because the alternative is too cruel. Kerr later says he believes in a soul and thinks the phenomenon goes beyond brain function. That is belief speaking, not data. But it is belief from a doctor who has spent decades sitting beside the dying, which is not the same as belief from a man selling angel candles on Instagram.

The stronger medical rebuttal is about delirium. If these visions were just the brain glitching at the finish line, the claim would be much easier to file away. Kerr says his studies screened for confusion states, and that delirium looks different: disorganized, frightening, distressing. These visions, by contrast, are usually coherent and calming. In death and dying, that difference is not decorative. It changes how families respond. It changes whether nurses correct a patient or listen.

The practical takeaway is not to win Reddit

Nobody should come away from this episode claiming science has proved Grandma is waiting in a blue dress. That’s not what Kerr can prove. What he can say, persuasively, is that many dying people are not merely fading into static. They are often having vivid experiences of reunion, repair, and comfort, and the rest of us should stop treating that as embarrassing nonsense.

The episode’s sneaky power is that it makes the question of the afterlife feel less like a debate-club prompt and more like a bedside etiquette problem. If a dying person says they just saw their mother, maybe the correct response is not “Actually, that’s probably your neurotransmitters.” Maybe the correct response is to sit down, shut up, and ask what she said.

The holy grail for us, the vital sign that matters for us is comfortable sleep because you can’t sleep if you’re in discomfort and you can’t die if you can’t sleep.