Under the Microscope: Low-level radiation seems to be much less harmful than experts had assumed, writes Prof William Reville
The model that has long been used to estimate the risk of contracting serious ill health from exposure to radiation is the linear no threshold (LNT) model. The data on which this model is based is good for higher levels of radiation, but poor for the lowest levels of exposure, and conservative assumptions were made about the risks of exposure to the lowest levels. There is now much evidence that the assumptions underpinning the LNT model for low-level radiation were much too conservative.
There are two types of radiation - ionising radiation (IR) and non-ionising radiation (NIR). IR is emitted by radioactive substances, and also includes cosmic rays that bombard earth from space, and medical X-rays. NIR covers all other radiation, including visible light, ultraviolet light, infrared, microwaves, radiowaves and more. This article deals only with IR which for simplicity I will just call radiation.
Radiation is greatly feared by the general public, mainly I think because people associate radiation with atomic bombs and the possible end of the world. I believe that the greatest human psychological scarring event of the 20th century was the atomic bombings of Hiroshima and Nagasaki in 1945. When an atomic bomb explodes, a lot of radiation is released but it is the awesome blast of the explosion that mostly causes the devastation. The blast effect at Hiroshima and Nagasaki killed more than 100,000 people immediately. Survivors of the explosion have since been intensely studied in order to correlate their exposure to radiation with subsequent ill-health.
When radiation interacts with matter, it deposits energy in the matter. The amount of dangerous energy absorbed per unit mass of matter is called the radiation dose. The unit of dose is the Sievert (Sv). One thousandth of a Sievert is a millisievert (mSv). Every year in Ireland we each receive an unavoidable dose of natural radiation (cosmic rays, radiation from the earth, including radon, and internal radiation) of about 4mSv. In some parts of the world the annual dose of natural radiation can be as high as 200mSv.
Radiation can cause two main ill-health effects - cancer and hereditary defects. There are two types of cell in the body - the somatic cells that make up our various tissues and organs, eg muscle, nerve, liver, etc, and the germ cells involved in procreation - egg cells in women and sperm cells in men. When radiation deposits energy in the cell it can damage hereditary material (DNA). This can initiate a cancer if the cell in question is somatic and the cancer will be expressed in the individual after a latent period of up to 40 years.
If the radiation damage occurs in the DNA of a germ cell, no ill-health effects will be expressed in the irradiated individual, but there is a risk that, if this individual is sexually active and fecund, that he/she will transmit a hereditary defect that will be expressed in a future generation. DNA damage in itself does not necessarily mean that ill-health will result. Cells can repair damaged DNA. Damage carries a risk that ill health will ensue, not a guarantee. The more extensive the damage the greater the risk.
In 1958 the scientists who studied the atomic bomb survivors proposed the LNT model, which says that the risk of contracting fatal cancer is linearly proportional to dose, ie, doubling the dose doubles the risk. It also says that there is no low threshold dose below which there is no risk. This concept is frequently expressed as: "There is no safe level of radiation," which is often misunderstood to mean that very low levels of radiation are very dangerous. But, what the LNT model says is that a tiny dose of radiation carries a tiny risk, a moderate dose carries a moderate risk and a large dose carries a large risk.
However, there has always been a problem with the LNT model. The Hiroshima and Nagasaki data were mainly for high radiation doses. There was little data for lower exposures, particularly for doses less than 200mSv. The linear relationship between dose and risk clearly applies for higher doses but the relationship at low doses was not clear. A conservative estimate was made that the linear relationship persists all the way down to zero dose, with no threshold dose.
There is now much evidence very low radiation doses, (less than 100mSv) are not nearly as dangerous as originally supposed. This evidence comes, for example, from studies of people who live in areas of the world where background natural radiation levels are high, and from the Nuclear Shipyards Workers Study in which 71,000 workers were studied. Also, the people of Chernobyl have been studied ever since the nuclear accident in 1986, most of whom received radiation doses below 20mSv. The number of deaths directly attributable to Chernobyl radiation currently stands at 56. No significant increase in cancer, other than childhood thyroid cancer, of whom 99 per cent were successfully cured, has been noted.
Overestimating the effects of low-level radiation has several negative consequences. It feeds radiation phobia, which prevents society from rationally debating important issues such as nuclear power. It calls for massive expenditure to protect the public against very low levels of radiation where the risk is probably non-existent, money that could be spent to good effect in other areas. It is time to think again.
• William Reville is associate professor of biochemistry and public awareness of science officer at UCC - http://understandingscience.ucc.ie