Methods used to calculate exposures in two epidemiological studies of power lines in the UK

Two epidemiological studies of cancer (one for children and one for adults) and proximity to high-voltage power lines are being performed in the UK. We describe the methods used to calculate exposure to magnetic fields in these studies. We used grid references derived from addresses for subjects and compared these to the grid references of pylons to calculate distances to power lines. We gathered relevant data on the power lines and used these to calculate magnetic fields. This required information on loads in the years from 1962 to the present. The only such information now available is a prediction of loads made for each year in advance, and the use of predictions rather than actual loads is the biggest source of error in these calculations. For a recent year, we compare the loads used in these studies with actual loads to assess the accuracy of our calculations. These calculations were performed by industry, and we describe the steps taken to ensure their trustworthiness, including conducting them all blind to the case-control status of the subjects.