Epidemiological Studies of the Porton Down Veterans: a ten-year update on mortality and cancer incidence, 2005-2014

The Porton Down Veterans cohort was originally assembled to see if there were any delayed effects on the health of UK servicemen who took part in the 'human volunteer programme' of chemical warfare agent (CWA) research at Porton Down from 01/04/1941 to 31/12/1989. We showed that the 18,276 Porton Down veterans had no overall excess of cancer compared to 17,600 veterans who did not visit Porton Down, taking follow-up until 31/12/2004. While they had a small (6%) excess of deaths from all causes combined, this could not be directly attributed to CWAs.

The cohort study has not previously been updated and we now propose to take follow-up to 31/12/2014. The additional ten years will greatly increase statistical power because we estimate it will add almost 8,000 further deaths and a proportionate number of additional cancer registrations. This will allow us to study new questions that have arisen since the first study. The cohort remains relevant today because the potential for long-term harm from CWAs continues to be topical because of their use by rogue states and terrorists. The exposures used in the research at Porton Down were very low but were also the best documented CWA exposures in the world and therefore continue to be informative about the effects of exposures at the low end of the spectrum. Some of the survivors of a CWA attack would be expected to have had low exposures.

These new research questions include exploring, for example: brain cancer in relation to nerve agents; oesophageal cancer in relation to CS gas; suicide in relation to CWA exposure. We also plan to assess the availability of data on smoking and alcohol consumption now that military personnel record retrieval has been computerised. If this information can be retrieved, it would greatly strengthen our statistical analyses because we would be able to control in our analyses for confounding by these lifestyle factors. Finally, we will explore the feasibility of linking these cohorts to datasets which only became available to researchers after the first study was completed, such as Hospital Episode Statistics data. If linkage is feasible, this will lead on to future research funding applications, such to study the mental health of older-aged military veterans.

Most of the work required to do this new analysis was done during the first study, so the proposal represents a prudent and efficient use of public funds. Concerns about exposure to CWAs or their treatments are likely to remain long into the future and updating this cohort allows us to examine these concerns and provide reliable evidence for policymakers and service providers. Society has a duty of care to its veterans and the updated cohort can also help to answer new questions about the health of older veterans, as they arise in the future, an important current issue for both the Ministry of Defence and also the Royal British Legion.

The Porton Down Veterans cohort comprises UK servicemen who took part in the 'human volunteer programme' of chemical warfare agent (CWA) research at Porton Down from 01/04/1941 to 31/12/1989. 18,276 Porton Down veterans had a small (6%) excess of deaths from all causes combined, compared to 17,600 veterans who did not visit Porton Down, taking follow-up until 31/12/2004. However, this could not be directly attributed to CWAs. They had no overall excess of cancer.

Since 2004, there has been no further follow-up. Therefore, we propose to take follow-up to 31/12/2014. The additional ten years will add almost 8,000 further deaths and a proportionate number of additional cancer registrations, greatly increasing our statistical power. Updating this cohort to allow updated and new analyses to be undertaken represents a prudent and efficient use of public funds.

We will update the standardised mortality ratios and rate ratios calculated initially. We will explore new questions e.g. the risk of neurological malignancy and non-cancer deaths with nerve agent exposure. We will also explore findings from the initial analysis e.g. the relationship between CS gas exposure and oesophageal cancer. Greater statistical power will allow us to study exposure-response relationships in more detail, e.g. for lung cancer and vesicants, and to study subgroups of causes of death, e.g. suicide.

We will carry out a pilot study of the availability of smoking and alcohol information in the newly computerised military personnel archives and, if feasible, incorporate these confounding factors in nested case-control studies of suicide, oesophageal cancer, and lung cancer.

We will also carry out a pilot study of data linkage to newly-available routinely collected datasets e.g. Hospital Episode Statistics.

Collaboration between KCMHR with academics in the University of Oxford and Lancaster University will ensure that this work has the highest academic impact. These links will also support future work to develop this cohort further. At least four high impact publications are planned from this project. Dissemination of this research will be supported by the King's College London (KCL) press office and through the KCL Science Media Centre.

The presence of an active Stakeholder Advisory Group will ensure that the work has the strongest policy impact, particularly on health protection for emergency preparedness. Veterans, and policy regarding veterans' health, will also benefit.