Guidance

Nuclear Decommissioning Authority (NDA) radiation epidemiology programme: information sheet

Updated 23 July 2018

Background

There are international and national guidelines and regulations, which are designed to protect the health of workers by restricting occupational exposure to ionising radiation. These guidelines are largely based on long-term health risks found from studies of survivors of the atomic bombings of Hiroshima and Nagasaki in 1945 and from certain groups exposed to radiation for medical reasons.

However, the atomic bombing survivors were exposed mostly to high doses of radiation over a short period of time, whereas radiation workers are generally exposed to lower doses over protracted time periods. The guidelines developed to protect radiation workers had made assumptions by extrapolating the results of the atomic bomb survivor studies.

In the 1970s, a number of epidemiological studies were set up to investigate the health effects of long-term low-dose exposure to ionising radiation in radiation workers. Two such studies were set up by British Nuclear Fuels Limited (BNFL) and the UK Atomic Energy Authority (UKAEA) in 1978. These studies were subsequently transferred to the NDA as part of the reorganisation of the nuclear industry. Both are long-term follow-up studies of the health of radiation workers from the former BNFL and UKAEA sites that now make up the NDA radiation epidemiology programme.

The BNFL programme was managed by the company Chief Medical Officer, before it was contracted out to a private company, Westlakes Scientific Consulting (WSC), in 1993. The management of this contract was passed to NDA in 2005. The programme was paused in 2010, following the liquidation of WSC, and restarted in 2011 by the NDA in contract with the Health Protection Agency (HPA, now Public Health England (PHE).

The UKAEA programme was managed by the company Chief Medical Officer and operated by UKAEA-employed researchers until 2008, when the research team was spun out to become part of AEA Technology, RWE Nukem and then Nuvia Ltd. In 2010, the UKAEA’s management of the research ended and, in 2011, the NDA contracted the Health Protection Agency (now PHE) to manage the Nuvia contract alongside the former BNFL programme.

Study team

The NDA Radiation Epidemiology Programme is currently being managed by PHE on behalf of the NDA. PHE’s Radiation Epidemiology Group based at the Centre for Radiation, Chemical and Environmental Hazards (CRCE) was identified as being the organisation with the most extensive skills, knowledge and experience in epidemiological and radiobiological research involving radiation workers.

The UKAEA database is currently being operated and maintained by Nuvia Ltd and the BNFL database is directly operated and maintained by the PHE Radiation Epidemiology Group.

The functions of the HPA were transferred to PHE in 2013. The Centre for Radiation, Chemical and Environmental Hazards (CRCE) is part of PHE and has its headquarters at Chilton in Oxfordshire. It provides the focus for research and advice on radiological protection as well as chemical and environmental hazards.

Study design

Epidemiological studies collect data on people’s exposure to particular events and also collect data on the occurrence of particular diseases in those people. Statistical analyses compare different levels of exposure to see if there is any difference in the pattern of diseases.

Studies in the NDA Epidemiology Programme link details of a worker’s occupational radiation exposure with health data collected later, such as cause of death or cancer incidence. Periodically, specific analyses (or ‘studies’) are undertaken to investigate specific issues or to provide an update on previous studies by using the additional data collated since a previous analysis. Statistical analyses are used to consider what conclusions may be drawn concerning risks of mortality and cancer incidence from comparisons between the study group (or ‘cohort’) and the national population, and between groups of workers with different levels of radiation exposure.

Large databases have been established and maintained over many years for the collation and management of data for the NDA Programme. As events occur (such as radiation exposures, cancer incidence or death), relevant data is collected and the databases are updated.

Study population

The programmes include radiation workers that were employed by UKAEA and BNFL. Workers employed before the study began in 1978 have also been included in the study population. The BNFL programme currently holds details of over 60,000 workers. The UKAEA programme currently holds details of over 75,000 workers.

For records on the UKAEA database, relevant radiation dose information, employment and follow-up data continues to be added for all workers. Workers on the former BNFL sites who have started more recently have not yet been added to the database. This work had been on hold when WSC ceased operating and it is now intended that the more recent workers will be added to the study cohort as soon as the new database managers, working with the NDA and the governance group, are able to establish suitable arrangements for database operations and data handling.

Contractors are generally not included in the databases. Some contractors may be included by virtue of their previous employment by BNFL or UKAEA. The feasibility of including more contractors into the study is a topic which the programme will address in the future.

Employment and follow-up information

In brief, employers provide records of employment episodes, occupational exposure to ionising radiation and some personal identifying data (including name, date of birth, sex, National Insurance Number and National Health Number). The mortality and cancer incidence data are provided through the National Health Service Central Registers (NHSCRs), which are maintained by NHS Digital. The NHS Digital Medical Research Information Service conducts record-linkage work that enables transfer of this data to the NDA programme databases. The personal identifying data supplied by the employer makes this record-linkage process possible.

Results

The analyses to date have shown that on average UK radiation workers have a greater life expectancy and more healthy life than the UK general population. This can be explained by the so called ‘healthy worker effect’, where occupational groups often display better health than the general population because those who are chronically ill are less likely to be at work. Nonetheless, radiation epidemiology studies have observed small increases in the risk of some cancers with increasing radiation exposure.

The health risk estimates derived from these studies have shown consistency with those used by international bodies setting radiation protection standards. This indicates that the measures currently in place in the UK workplace should provide adequate protection of the health of workers from cancer risk

Other similar studies

In addition to the NDA Radiation Epidemiology Programme there is the National Registry for Radiation Workers (NRRW) which is funded and managed by PHE. The NRRW is a long-term follow-up study of the health of radiation workers. It is concerned with UK radiation workers and its target study population is all radiation workers for whom records of occupational exposure to radiation are being kept and whose employer organisations have agreed to support the epidemiological research project. The cohort includes workers from a wide range of employing organisations including most of the nuclear industry and other organisations including the Ministry of Defence and GE Healthcare (formerly Amersham International), as well as a number of smaller organisations in the research and industrial sectors.

The most recent analysis of the NRRW, published in 2009, concluded that

This analysis provides the most precise estimates to date of the risks of mortality and cancer incidence following occupational radiation exposure and strengthens the evidence for raised risks due to these exposures. The cancer risk estimates obtained here are consistent with values used by national and international bodies in setting radiation protection standards.

The authors also recommended the continued follow-up of the workers in all groups to see whether radiation-associated risks vary over time or by age, and also proposed the study of specific cancers and causes of death in more detail.

The NRRW is the largest UK epidemiological study of radiation workers and includes most UK nuclear workers as well as a wide range of radiation workers from other industries. There are similar projects in other countries and there has been some collaborative work on an international basis. In all cases, the primary aim is to establish whether current protective measures are sufficient or whether there could still be areas where workers could be better protected.

The NDA programme allows research specifically related to the BNFL and UKAEA radiation workers to be carried out. The databases contain additional data on internal radiation exposures (for example from plutonium and uranium) not currently held by the NRRW. They also contain information on non-radiation workers within the nuclear industry, providing a useful comparison groups for studies.

Data security

The databases associated with the programme are registered under the Data Protection Act and information security is taken very seriously by the research teams. The databases are held on a secure system and are accessible only to authorised users who require various credentials and authorisations to access the database. Personal identifiable information collected for the programmes is not made available for any other purpose and can only be viewed by those members of the research team that need to access the data and database.

When a dataset is extracted from the database for analysis, the extracted records will be pseudonymised. This means that all identifying details that are not required for this stage of the work (such as names) will be removed. The records will instead be allocated a reference identifier which would enable the database staff to link the record back to the database in the event that there were specific data queries. The reference identifier would not, however, allow anyone else to identify the specific individual. If the data were fully anonymised (having no possible link back to the database) it would not be possible to investigate inconsistencies and this could then impact on the quality of database used for the research.

The databases associated with the UK programmes are considered an important international scientific resource. For the NDA radiation epidemiology programme, requests to use data in conjunction with other datasets will be considered, but this will only be allowed with the approval of the governance group who will expect appropriate ethical considerations have been made and relevant ethical approvals obtained. The outcomes of such national and international collaborative research can help to address some of the questions which cannot be addressed by smaller studies. In all cases, the data which may be shared after such agreements will not contain information which enables any individual to be identified – it will always be pseudonymised.

Study participation

The NDA programme is a records based project, which means that details for each individual are drawn from existing records and it does not require individuals to be contacted for data or to complete a questionnaire. This approach reduces the impact on the individual, but it also reduces the risk of bias in the study due to the response of participants.

The NDA programme operates an ‘opt-out’ policy by which any individual may request that their details are either not included in any study or not included in a particular study. Of course, we do hope that the number of people choosing to take this option will continue to be very low. This will ensure that the research findings, which are intended to inform both radiation workers themselves as well as the scientific and medical community, will be more valid than if a significant number of workers opt-out. New studies are announced in a number of ways (such as a company newsletter) to allow individuals to opt-out. The size and historical nature of the databases means that it is not practicable to contact participants individually when a new study is proposed. It is possible for participants to opt-out of the study by contacting the PHE Radiation Epidemiology Group by emailing radiationworkerepidemiology@phe.gov.uk.

The Trade Union representatives on the NDA-PHE governance group strongly encourage all workers to remain in the programmes which were set up to protect the health of all workers.

The NDA and PHE are very grateful for the ongoing interest and support of the study members and their representatives. This support is essential for the studies to be undertaken and the NDA and PHE believe it is essential that the workforces are appropriately informed about the results of work that is undertaken with their data.

Ethical approval

The UKAEA and BNFL studies have been given a favourable opinion by NHS Research Ethics Committees in 2013 and 2016 respectively. The study has also received support from the Confidentiality Advisory Group (CAG). The CAG considers applications for accessing personal data without consent under section 251 of the NHS Act 2006 and the Health Service (Control of Patient Information) Regulations 2002, and provides recommendations to the Health Research Authority.

Independent oversight and review

An independently chaired governance group provides oversight of the programme. The governance group includes members from NDA, Sellafield Ltd, Staff and Trades Unions and PHE. For specific analysis, proposals are (where necessary) separately reviewed by appropriate Ethics Committees.

Approval from both an Ethics Committee and the governance group is required for new studies to begin. Governance group approval is not granted without full agreement of all parties on the governance group. The research findings are published in reputable scientific journals. These journals operate a peer review system whereby reports are examined by independent expert scientists who provide a view on the quality of the report, the findings, and adherence to ethical principles and whether the work adds usefully to the scientific literature. If they think that the work has not been undertaken properly they suggest that the paper is not published at all. Examples of reputable journals that have published radiation worker studies in the past include the British Medical Journal, the Journal of Radiological Protection and the British Journal of Cancer.

PHE is committed to providing pre-publication briefings to relevant stakeholder groups for all its epidemiological worker studies. This ensures that the constituent workforces have been notified of the results in advance of publication. PHE researchers will provide a written brief or, if it is considered appropriate, will present the results and address questions in advance of the publication. The governance group will help to provide a view on the most appropriate way of presenting the results.

Any other research group requesting permission to work with these datasets would have to guarantee the same level of commitment to communication before governance group approval could be granted.

Publications

A large number of reports have been published in scientific journals and as company specific reports.

Following the liquidation of WSC, the BNFL studies were paused whilst the programme was transferred to PHE. Where feasible we have re-started studies which had previously received approval from the governance group and any necessary ethical approval. Ongoing studies will be described in separate publications in due course.

A selection of publications that have used data from the NDA radiation epidemiology programmes are given below:

The cancer mortality and incidence experience of workers at British Nuclear Fuels plc, 1946 to 2005. Gillies M, Haylock R. Journal of Radiological Protection 34.3 (2014): 595.

The non-cancer mortality experience of male workers at British Nuclear Fuels plc, 1946 to 2005. McGeoghegan D, Binks K, Gillies M, Jones S, Whaley S. Int J Epidemiol 37 (3): (2008) 506 to 518.

Mortality of employees of the United Kingdom Atomic Energy Authority, 1946 to 1997. Atkinson W D, Law D V, Bromley K J, Inskip H M. Occupational and environmental medicine 61.7 (2004): 577 to 585.