The Deathbed Doctor: Weeks Before Death, Patients Start Having Visions! | Dr Christopher Kerr

Dr. Christopher Kerr describes himself as having become a hospice doctor "more by accident": a physician who cares for people no longer deemed treatable. Over roughly 30 years he says he has interviewed about 1,500 dying patients. His stated reason is that medical school trained him "to defeat death, not to care for those who are dying," and that nursing colleagues were more attuned to what patients were actually experiencing. His conclusion is that the patient's perspective "is the one that matters," and that seeing patients as people reveals rich emotional lives at the end.

Steven Bartlett notes he has around 15 of Kerr's research papers in front of him. Kerr summarises the program's scope: screening out confusion states, studying post-traumatic growth within the experience of dying, looking at dying children, and measuring the effect of these experiences on bereavement among those left behind.

The episode's opening is a montage of later moments plus a request to follow the show; it previews claims such as "nearly 90% of people" having intense dreams or visions and the line that death is "this unbelievably truthful place." Readers should note that the episode description gives the figure as "over 80%" while Kerr says "nearly 90%" in conversation; the exact figure and its study population are not specified on air.

According to Kerr, the vast majority of patients — he says nearly 90% — begin to have very intense dreams or visions, and the dominant theme is seeing people they loved and lost. He is specific that it is those who loved them "unconditionally, purely," and that those who did not are absent. He frames dying as "progressive sleep": sleep architecture changes, inner experiences increase, and their frequency rises in roughly the last week or two of life.

He presents the research as an attempt to humanise dying — shifting from a medical model of organ failure toward what patients actually experience. When the team measured which themes brought most comfort, seeing deceased loved ones ranked highest, above dreams involving the living; Bartlett puts a graph of this on screen.

Kerr explains that patients were filmed early because medical audiences might assume dying patients were compromised or "feeble-minded." In the first clip, Paul, three weeks before his death, describes around 15 dreams of his late wife, always in light blue, smiling, occasionally speaking. Kerr highlights the vividness, the sparse dialogue, and "the absolute absence of fear" across the videos.

Asked what Paul's dreams mean, Kerr says his view is changing. He once would have called them rekindled memories, like looking at a picture. Two features now suggest more to him: the remarkable detail (a 95-year-old recalling his mother's perfume and voice from age five) and the presence of communication rather than still images. Pressed by Bartlett on whether the dead are truly accessible, he declines to claim an afterlife — "I don't know what this means" — but says he is more convinced "that those who have loved us are never gone."

Kerr describes a symmetry between birth and death: an infant without language still knows its mother's love, and a neurologically diminished dying person retains a core sense of being connected and loved.

The cases he finds most persuasive are people who carried harm for decades and found peace only in these experiences — an old veteran with PTSD reunited with those who died overseas, and children. In another clip, Florence, five days before her death, describes sitting at a kitchen table laughing with her late husband and daughter; she says it felt as if the years apart "had never existed" and that she is not afraid.

Bartlett raises the obvious objection: people facing death think about family and endings, so they dream about them. Kerr's rebuttal is experiential: patients are most distressed when these are called dreams, describing them as non-metaphorical, needing no interpretation, and real; on a zero-to-ten realism scale, he says the answer "is always a 10." About half, he says, occur while awake — patients looking through the room or petting a dog that isn't there — usually within the last couple of weeks, becoming more immersive as death nears, with time and distance falling away.

On whether this is biology or spirit, Kerr — who also holds a PhD in neurobiology — says no one knows. He argues you can map correlates of love to brain regions and neurotransmitters but "can't make the whole." He then states his personal belief that it goes beyond brain function: he answers yes to believing in a soul that lives beyond the body. This is presented as his belief rather than a research finding.

Bartlett cites a survey that half of Americans say a dead relative has visited them, which the conversation picks up in the next section.