For more than 110 years we have been able to generate ionising radiation and for 3.8 billion years life has been exposed to natural radioactivity. We know that exposure to high doses at high dose rates is detrimental to health and we still appear to be on a learning curve in that respect. From believing in the early 1970s that hereditary disease and leukaemia were the only late health effects we now know that solid cancers and non-cancer diseases are also caused. This is primarily due to the study of the survivors of the atomic bombings in Japan.

What is much more debated is “what are the effects of low doses of radiation delivered at low dose rates”? This is the kind exposure that arises from the environment both in the form of natural radiation and man made sources, such as releases from nuclear reactors. This is an exposure range in which epidemiology is rather a limited and blunt tool. Regulatory bodies which control the exposure of the public to these sources of radiation assume that no exposure, however small, is without a finite risk. This is known as the “Linear No Threshold” or LNT, hypothesis.

There is much misunderstanding as to what this means but essentially we can say that it dictates that at low doses, <50mGy, the dose response will be linear from zero dose. What LNT does not stipulate is the slope of that line and there is no implications regarding how to interpolate from this dose range to the higher doses (>200mGy) where the epidemiological data are more secure. The International Commission on Radiological Protection, ICRP, assumes, for cancer, that this initial dose response would have half the slope of the direct linear extrapolation of the Japanese bomb survivor data. This is contested as there is no specific evidence to support it.

Some argue that there is evidence that at low doses there is in fact a threshold and some even that low dose exposures can be beneficial.

My current view is that although there are a few specific circumstances where there appears to be a threshold, bone cancer after internal exposure to radium-226 is an example, in general the LNT hypothesis applies. In this area selective use of the evidence can be used to support almost any argument.

Of course for very low doses the risks are also very low and thus, there being few aspects of life that are without risk, some level of risk should be regarded as acceptable.

This emphasises the importance of understanding the effects of ionising radiation in order to confidently assess risk in the many circumstances in which unavoidable exposures might arise.