A phantom based method for deriving typical patient doses from measurements of dose-area product on populations of patients.

One of the chief sources of uncertainty in the comparison of patient dosimetry data is the influence of patient size on dose. Dose has been shown to relate closely to the equivalent diameter of the patient. This concept has been used to derive a prospective, phantom based method for determining size correction factors for measurements of dose-area product. The derivation of the size correction factor has been demonstrated mathematically, and the appropriate factor determined for a number of different X-ray sets. The use of phantom measurements enables the effect of patient size to be isolated from other factors influencing patient dose. The derived factors agree well with those determined retrospectively from patient dose survey data. Size correction factors have been applied to the results of a large scale patient dose survey, and this approach has been compared with the method of selecting patients according to their weight. For large samples of data, mean dose-area product values are independent of the analysis method used. The chief advantage of using size correction factors is that it allows all patient data to be included in a survey, whereas patient selection has been shown to exclude approximately half of all patients. Reduction of the size of the data set may lead to mean dose-area product values that are less reliable indicators of typical practice. The use of size correction factors will be of particular benefit in the analysis of paediatric dosimetry data, where a wide range of sizes exist, even within accepted age bands.