The abolition of regional health authorities raises questions about the future of the 12 regional cancer registries in England and Wales. Their survival will now depend on their ability to persuade district health authorities that they are essential for purchasing. This in turn will depend on how quickly they can acquaint purchasers with their role and function.

Cancer registration was introduced by clinicians in the 1930s to evaluate new treatments such as radium therapy. It has come to be identified with narrowly epidemiological aims, yet most registries hold a wealth of data that are vital for evaluating health care - for example, on treatment, tumour stage, place of treatment, consultant, and place of death. Although notification is voluntary, the registries achieve over 90% ascertainment for most cancers.1 They contribute population data to the national cancer registration system at the Office of Population Censuses and Surveys, which in turn publishes annual statistics on incidence and mortality and, all too infrequently, survival.2

Most district purchasers use registry data only to monitor trends in incidence and mortality and to ensure that these accord with targets set out in the Health of the Nation. This is because many are ignorant about what registries can do; they know that many clinicians doubt the accuracy of registry data, and they may perceive that a registry duplicates data that they already collect. Since 1993 NHS acute hospital providers have had to make a return on …