John Campbell, a retired nurse educator with more than three million subscribers, sat down in front of a camera this month and told a large and grateful audience how to catch cancer before it catches them. The video is called Cancer, early warning signs, and it does what his videos have always done well: it treats ordinary people as capable of understanding their own bodies, and it tells them not to wait, not to be brushed off, not to assume the lump or the cough is nothing.
The list he walks through is one of the most repeated in all of medicine. It is also one of the least tested.
Generations of patients have been handed some version of it. The American Cancer Society built the mnemonic CAUTION decades ago and still publishes the same core signs and symptoms of cancer: a change in bowel or bladder habits, a sore that won’t heal, unusual bleeding, a thickening or lump, trouble swallowing, an obvious change in a mole, a nagging cough. The premise underneath it is simple and humane. Notice the sign, see a doctor, survive. Nobody argues with the instinct. The argument is with the word “early.”
For some of the cancers that kill fastest, the warning signs are not early at all. They are the sound of the door already closing. Pancreatic cancer is the clearest case. By the time it produces the back pain, the jaundice, the unexplained weight loss that would send anyone to a clinic, it has usually already spread, and most cases are caught after the window for a cure has closed. The overall five-year survival still sits around 12 percent. The symptoms are real. They are just late.
Gastric cancer tells the same story with unusual candor. The registry record for a South Korean screening trial states the problem in a single sentence in its own trial listing: early gastric cancer is mostly asymptomatic, and a cancer found because of warning symptoms such as bleeding or weight loss “usually can not be cured.” Read that again. The entire logic of the warning-sign poster is that the sign comes first and the tumor is still catchable. For the disease that registry was describing, the sign is the eulogy.
So the question is not whether the signs exist. It is whether teaching people to watch for them actually changes when cancer gets caught. Here the record is unkind to the poster.
England ran one of the largest natural experiments anyone has: the Be Clear on Cancer program, a decade of national campaigns telling people to see a GP about blood in the urine, a persistent cough, a change in bowel habit. It worked, in the sense that campaigns are usually judged to work. When the government’s own evaluators synthesized eleven of the campaigns, primary-care attendances rose for 9 of 10 measured campaigns, and urgent referrals for suspected cancer rose above trend for 9 of the 11. Waiting rooms filled with worried people doing exactly what they were told.
The campaigns moved one number reliably. They did not move the other one. That same synthesis found less evidence of any shift toward earlier-stage disease and no measurable impact on survival. It is not an isolated verdict. A systematic review in the British Journal of Cancer found only limited evidence that public education campaigns reduce the stage at which breast cancer, melanoma, or retinoblastoma is diagnosed, and little good evidence that awareness efforts moved anyone to present earlier. A more recent review focused on lung cancer reached the same shrug: awareness interventions were not associated with a shift in the stage of cancer at diagnosis, and the two outcomes that actually matter, stage shift and mortality, were barely measured at all. That last detail is the quiet indictment. Programs sold as lifesaving mostly never checked whether anyone lived longer.
None of this means early detection is a fantasy. It means the thing that detects cancer early is not a symptom, because a symptom is a body that has already noticed. What works is looking before the body knows anything is wrong.
That is screening, and the best-tested screening has receipts a poster never earned. The National Lung Screening Trial, funded by the National Cancer Institute, scanned heavy smokers with low-dose CT and cut lung-cancer deaths by 20 percent against chest X-ray, 247 lung-cancer deaths per 100,000 person-years versus 309. It is not a free lunch. CT screening finds nodules that would never have killed anyone, and it exposes healthy people to radiation and biopsies they did not need. But it is a real number attached to a real reduction in dying, which is more than symptom awareness has ever produced. The work now is aiming the test better, pointing scans at the people most likely to benefit. A 2025 review of 38 externally validated lung-cancer risk models found the workhorse PLCOm2012 beat simple age-and-smoking rules across populations, with newer entrants like CanPredict and OWL close behind. That is early detection as engineering, not as a slogan.
The uncomfortable part, and the reason the poster never dies, is that awareness is cheap and screening is not. A public campaign costs a fraction of a screening program with scanners, radiologists, and follow-up for every false alarm. It also relocates the responsibility. Tell people to know the signs and the burden sits on the patient who “should have come in sooner,” not on a system that never funded the scan that would have found it first. The message feels like action. For the cancers doing the most killing, it mostly isn’t.
Campbell’s instinct is the correct one, and it is worth more than the list he read from. Take your own body seriously, insist on being seen, refuse to be waved off. The failure is not in the man at the camera. It is in the century-old list itself, which asks people to watch for the moment the news is already bad.
Sources
- International Journal of Cancer – Lai et al., “Reviewing the impact of 11 national Be Clear on Cancer public awareness campaigns, England, 2012 to 2016” (2021)
- British Journal of Cancer – Austoker et al., “Interventions to promote cancer awareness and early presentation: systematic review” (2009)
- npj Primary Care Respiratory Medicine – systematic review of interventions to recognise, refer and diagnose lung cancer symptoms (2022)
- NEJM – National Lung Screening Trial, “Reduced Lung-Cancer Mortality with Low-Dose Computed Tomographic Screening” (2011)
- Acta Oncologica – systematic review and meta-analysis of lung cancer risk prediction models (2025)
- Cancer Research UK – Be Clear on Cancer campaign evaluation summary
- ClinicalTrials.gov – Gastric Cancer Screening and Alarm Symptoms in Early Gastric Cancer (NCT01659632)
- Johns Hopkins Medicine – pancreatic cancer prognosis and survival
- American Cancer Society – signs and symptoms of cancer
- Dr. John Campbell – “Cancer, early warning signs”