The authors, led by Dr Laura Woods from the Cancer Research UK Cancer Survival Group at the London School of Hygiene and Tropical Medicine, say their findings "support a large body of evidence that survival after cancer in the UK is indeed lower, and that this is likely to be related to delay in diagnosis, lower investment in healthcare and sub-optimal care."

In a BMJ editorial in August 2010, two renowned cancer scientists, Professor Dame Valerie Beral and Professor Sir Richard Peto, questioned established research that found that breast cancer survival in the UK was lower than in equally developed nations. They claimed that breast cancer survival rates were misleading and appeared significantly worse because of data inputting errors in the National Cancer Registry and because many long-term survivors were never registered.

Woods and colleagues challenge these views, saying that they "are not plausible explanations for the deficit in cancer survival between the UK and many similarly developed countries."

The authors tested Beral and Peto's theory by simulating both errors in the entire national cancer data.

Beral and Peto suggested that because cancer registration is not statutory in England and Wales, the registry could incorrectly record the date of a breast cancer recurrence shortly before death, not the correct date of the initial diagnosis. The true survival time for that patient would then be misrepresented in the registry data.

However, when the research team put this theory to the test, they found that in order to explain the differences in one-year breast cancer survival between England and Sweden, the diagnosis date would have to have been incorrectly recorded by around one year for more than 70% of women known to have died.

Beral and Peto also suggested that many long-term breast cancer survivors are never registered. This would mean, say Woods and colleagues, that a large number of women are diagnosed, treated and survive breast cancer without any of their clinical information reaching the cancer registry.

It would also mean that the same group of women later die of something else with no mention of breast cancer on their death certificate.

Even if 40% of long-term survivors fell through the net, say the authors, this would explain less than half the difference in one-year survival from breast cancer between England and Sweden.

Dr Woods and colleagues say Beral and Peto's proposition "becomes more implausible when one considers the gross error that pathologists and cancer registry staff would have to have been making, for thousands of patients each year, consistently over many years, and in each of the cancer registries across all four UK nations."

The authors conclude: "The question has long since ceased to be whether such unacceptable differences in cancer survival exist. The question is what the underlying mechanisms are and what can be done to improve the outcomes for patients with cancer in the UK."

Link to paper

Source
British Medical Journal