The assumption that any additional exposure to ionizing radiation leads to an increase in the risk of stochastic health effects implies that some cases of these effects will be caused by exposure incurred occupationally. The main health effect expected to arise in the exposed individuals is cancer. Such radiation-induced cancers cannot be distinguished from the far larger number of background cancers, and, therefore, causation must be assessed statistically. The probability of causation methodology has been developed to ascertain the likelihood that a particular cancer may be attributed to a particular prior exposure to radiation. Given the pertinent details of an individual case, a probability of causation value is calculated from the appropriate relative risk obtained from radiation risk models derived from the epidemiological study of exposed populations, although there are many uncertainties inherent in a particular probability of causation calculation. In the United Kingdom, the Compensation Scheme for Radiation-linked Diseases has been created to determine whether those individuals occupationally exposed to radiation in the nuclear industry who have subsequently developed a malignant disease should be compensated. The Scheme is a voluntary arrangement based upon the probability of causation methodology, which incorporates various procedures agreed by employer and employee representatives and their advisers. In a pragmatic approach to compensation, the uncertainties of a specific probability of causation calculation are accommodated through generosity factors which favor the claimant and encourage the use of the Scheme. The Scheme, which was introduced in 1982 and modified in 1987 and 1991 in the light of operational experience and revised risk estimates, has provided a successful alternative to litigation from the point of view of both employer and employee.