Radiation Hazards, Dose and Risk

X-rays and gamma rays are ionising radiation. They have enough energy to remove electrons from atoms, producing ions. This can damage cells, including their genetic material. Genes are sections of DNA that carry instructions for cells; a change in genetic material is called a mutation. Some mutations can lead to uncontrolled cell division and cancer.

The important chain of reasoning is therefore: ionising radiation can damage genetic material, this can cause gene mutations, and some mutations can lead to cancer. Cancer is a possible consequence, not an inevitable outcome of every exposure.

Dose measures the risk from an exposure

Radiation dose is a measure of the risk of harm resulting from exposure of the body to radiation. It is not simply a count of waves reaching a person. For ionising radiation, dose expressed in sieverts takes account of differences in the effects of radiation types and the sensitivity of body tissues.

Dose is measured in sieverts, written Sv. Smaller doses are often given in millisieverts, written mSv:

$$1000\text{ mSv} = 1\text{ Sv}$$

To change millisieverts into sieverts, divide by 1000; to change sieverts into millisieverts, multiply by 1000. Comparing values without first checking their units can give a very misleading impression of risk.

In general, a larger dose of ionising radiation means a greater risk of harm. Low doses can increase the risk of cancer later in life, whereas very high doses can cause severe immediate damage and may be fatal. The radiation type matters too, so identifying a wave as hazardous is only the beginning of judging an exposure.

Reading dose and risk data together

A useful conclusion identifies what was measured, compares the figures and states what they imply. The following UK Health Security Agency figures are typical doses and approximate additional lifetime risks of fatal cancer per examination for patients aged 16–69:

The CT chest examination has the largest dose and the largest estimated additional risk in this table. A risk of 1 in 2,500 is greater than 1 in 1,000,000: the smaller denominator means the outcome is expected more often among equally sized groups.

These figures support the conclusion that, for the examinations shown, larger doses are associated with greater estimated additional cancer risk. They do not mean that every patient receiving a CT examination will develop cancer. ‘Additional’ also matters: the figures describe risk added by the examination, not a person's total lifetime cancer risk.

The age qualification limits the conclusion. These estimates should not be treated as exact predictions for every individual or applied unchanged to all age groups.

Drawing justified conclusions from a study

For data comparing exposure and health outcomes, first look for a trend: does the harmful outcome become more common as exposure increases? Quote figures that demonstrate the trend rather than simply saying that radiation is dangerous.

Then check whether the comparison is fair. If groups contain different numbers of people, compare proportions or percentages rather than raw case counts. Also check whether exposure is given per event, per year or over a lifetime; those are not interchangeable comparisons.

Finally, distinguish a correlation from proof of a cause. For example, groups with different UV exposure might also differ in age or other factors affecting their health. A larger sample and comparisons that account for these differences can strengthen a conclusion. State what the data supports, but do not claim more certainty than the evidence allows.

Radiation dose measures the risk of harm from exposure.

$$1000\text{ mSv} = 1\text{ Sv}$$

Larger doses generally mean greater risk. Low doses can increase later cancer risk; very high doses can cause severe immediate harm and may be fatal.

Conclusions from data

Identify the trend and support it with figures. Compare like units, exposure periods and proportions. Check qualifications such as age range and whether a risk is additional or total. A correlation does not establish causation; other differences between groups may affect the results.