How is scientific evidence assessed?

In order to provide the best advice on the protection of the Australian people and environment from the effects of radiation, ARPANSA reviews, on an ongoing basis, the relevant scientific research.

This fact sheet explains how ARPANSA assesses scientific evidence in order to establish evidence-based and risk-informed standards, to make recommendations and to provide advice. Risk in this context can be defined as the likelihood of developing a disease or injury from radiation exposure, either this occurs in people or in the environment.

What are the types and hierarchy of scientific data?

When considering effects from exposure to radiation, various disciplines of science, including biology, epidemiology and medicine, as well as physics, engineering and social sciences, need to be considered. All of these disciplines play important individual and collective roles in identifying possible adverse effects on health and in providing information on the need for, and approaches to, protection. The relevance of such studies to health and environmental protection varies.

Epidemiological studies look into the distribution of disease in human populations. The factors that influence this distribution provide direct information on the health of people exposed to an agent such as radiation. Such studies are given the highest ‘weighting’. Since this type of study is based only on observations on the subject population and is not a controlled experiment, it is prone to errors which can influence the result. Therefore a cause and effect relationship may be difficult to be drawn, except when the evidence from other studies is strong.

Provocation studies are experimental studies using human volunteers, and can give a valuable insight into any physiological responses provoked by a particular agent. For ethical reasons, this type of study is normally restricted to healthy people and is often not feasible (for example it is not ethical to expose people to large amounts of radiation in order to see what happens).

Other experimental studies include studies on animals, tissues and cells. These are also important but are given less weight. Animal studies can often be expected to provide information regarding potential health outcomes, but because of the differences between species the data may not be extendable or applicable to providing information on risk to humans. Studies carried out at the cellular level are normally used to investigate possible ways that an agent may interact with living organisms, but are not generally taken alone as evidence of effects.

Studies of effects in the environment, in particular after accidents or in localised areas of release of radioactive substances, can provide information on sensitivity to radiation among species other than humans, occupying the natural environment.

Drawing a conclusion from the scientific evidence

No single study considered in isolation will provide a meaningful answer to the question of whether or not radiation of any type can cause or contribute to adverse health effects in people or in the environment. In order to make an informed conclusion from all the research studies, it is important to weigh the science in its totality. The final conclusion is not necessarily a definitive yes or no, but may express the likelihood for the existence of a cause and effect relationship. The scientific community uses a system to decide which research results should be published in reputable scientific journals called peer review. Peer review subjects scientific research papers to independent scrutiny by other qualified scientific experts (peers) before they are made public.

Scientific evidence is deemed to be established when it is consistent and generally accepted by the broader scientific community. This usually follows an evaluation of the available data by expert scientific bodies using a health risk assessment approach. In a health risk assessment all the available studies, with either positive or negative effects, need to be evaluated and judged on their own merit, and then all together in a weight of evidence approach. It is important to determine how much a set of evidence changes the probability that exposure causes an outcome. Generally, studies must be replicated or be in agreement with similar studies. The evidence for an effect is further strengthened if the results from different types of studies (epidemiology and laboratory) point to the same conclusion.

In some cases, it might not be possible to quantify the health risk based on observations; in such cases risk models may be used to support informed decisions regarding the level of risk and its mitigation.

Health authorities like ARPANSA or international organisations such as the World Health Organization or the International Commission on Radiation Protection assess all of the available evidence and provide appropriate advice. Individual studies are usually published as peer-reviewed journal papers however conclusions from these can often vary. Opinions on health risks are also often provided in the media and on the internet however the scientific value of these can often be questionable and subjective.

Science continually evolves and ARPANSA reviews on an ongoing basis the research into the potential health effects of radiation and consults with relevant stakeholders in order to provide accurate and up-to-date advice.

Health risk assessments influencing public health policy

The table below gives examples of health risk assessments done by WHO on specific radiation types that had a major influence on health policies around the world.