Research and analysis

NRRW data: publications by study team

Updated 22 August 2024

Results from the analysis of National Registry for Radiation Workers (NRRW) data were first released as National Radiological Protection Board (NRPB), and later Health Protection Agency (HPA) published reports accompanied by summary papers in the peer-reviewed scientific press.

The UK Health Security Agency (UKHSA) now only publishes the results of analyses in the peer-reviewed press. The scientific journals now give the opportunity to provide extra information (previously provided in the reports) online on their websites, hence the need to have accompanying reports has reduced in recent years.

UKHSA has a policy of, where funds allow, publishing NRRW results ‘open access’ in journals with the highest possible impact factors. These have included the British Medical Journal and the British Journal of Cancer.

UKHSA staff also give presentations about the study and its results at scientific meetings to bring these to the attention of national and international radiation protection standards-setting bodies.

Summary of shape of radiation dose response for ischaemic heart disease mortality

Wei Zhang, Richard GE Haylock, Michael Gillies, Nezahat Hunter. Shape of radiation dose response relationship for ischaemia heart disease mortality and its interpretation: analysis of the national registry for radiation workers (NRRW) cohort. Journal of Radiological Protection (2024).

Background

Statistically significant increases in ischaemic heart disease (IHD) mortality with cumulative occupational external radiation dose were observed in the National Registry for Radiation Workers (NRRW) cohort. The dose-response relationship of ischaemia heart disease mortality in NRRW cohort exhibited a downward curvature at a high dose level greater than 0.4 Sv. The overall shape of dose-response relationship best described by a linear-quadratic model. This publication aims to investigate the reasons behind the observed downward curvature at higher dose levels.

Methods

The NRRW cohort included 174,541 radiation workers, among whom 9,814 deaths were attributed to IHD. The study has individual level information, including about:

• sex
• date of birth
• date of first employment
• attained age etc

The external exposures were mainly resulted from gamma rays and x-rays but with smaller contribution from neutrons and beta particles. IHD mortality risk in relation to cumulative radiation dose was analysed using Poisson regression methods for grouped survival data. Specifically, radiation risk of IHD were examined in relation to:

• age at first exposure
• time since first exposure
• employment length
• doses received in different age windows

Results

IHD death accounted for 89% of all the heart disease mortality in the NRRW cohort. The excess relative risk (ERR) per sievert (ERR/Sv) and it’s 95% confidence interval were 0.32 (0.04, 0.61) for IHD. The dose response was best described by a linear-quadratic model with a downward curvature. The excess risk in the higher dose category (greater than 0.4 Sv) was responsible for the downward curvature, as it was smaller than expected value derived from a linear dose-response model. We investigated the reason behind the significant drop in excess risk in the high dose category (greater than 0.4 Sv) through various analysis.

Our findings revealed that over 60% of worker’s person-year contribution in the dose category greater than 0.4 Sv came from radiation workers who joined the workforce before age 30. Interestingly, the excess risk for this group of workers was smaller than other age groups. This phenomenon appears to drive the overall excess risk to a lower value in the greater than 0.4 Sv dose category. We also examined IHD risk in relation to radiation dose received across different age-windows. Among workers first exposed less than 30 years old, doses received when workers were less than 35 years old or more than 50 years old did not contribute to increased IHD mortality risk. Increased risk was observed for doses received between 35 and 50 years of age. This led to smaller IHD risk when the risk was averaged over the total cumulative dose among workers first exposed less than 30 years old.

We modelled the dose response by age-at-first exposure. For age-at-exposure of 30 and above years old, a linear dose response was the best fit.  For age-at-exposure less than 30 years old, there was no evidence of excess risk of IHD mortality for radiation doses below 0.1 Sv or above 0.4 Sv; excess risk was only observed for doses between 0.1 to 0.4 Sv.

Conclusion

We found that a large proportion of person-year contribution in the dose category greater than 0.4 Sv came from workers first exposed at less than 30 years of age. The ERR of IHD for this group was smaller than other groups, resulting in a reduction of the overall ERR at higher dose level. Among workers first exposed less than 30 years, we observed that doses received when workers were less than 35 years old or greater than 50 years old did not contribute to the increased IHD mortality risk.  The increased risk was associated with doses received between 35 and 50 years of age.  As a result, the averaged IHD mortality risk over the total cumulative dose among workers first exposed less than 30 years old was smaller than other age groups.

Summary of radiation effect on respiratory disease mortality

Wei Zhang, Richard GE Haylock, Michael Gillies, Nezahat Hunter and Erica Zhang. Effects of radiation on respiratory disease mortality: analysis of National registry for radiation workers in United Kingdom: 2023. International Journal of Radiation Biology.

The effects of external radiation exposure on overall respiratory disease mortality and its subtypes (pneumonia, chronic obstructive pulmonary disease (COPD) and other remaining respiratory diseases) have been evaluated using the NRRW cohort.

Background

While some evidence of an effect of radiation exposure on respiratory disease at low dose levels has now emerged, there are differences in the estimated risks between studies across populations. In this publication, we aim to show the effect of radiation on 3 different subtypes of respiratory disease mortality through the analysis of the NRRW cohort of occupationally exposed workers in UK.

Methods

The overall respiratory disease mortality and its subtypes were analysed. They include:

• pneumonia (1,066 cases)
• COPD (1,517 cases)
• allied disease and other remaining respiratory diseases (479 cases)

Most of the radiation doses received by the workers in NRRW cohort are associated with external exposure to X-rays and gamma rays. While 25% of the male workers and 17% female workers were identified as being monitored for internal exposure, estimates of internal dose were not available.

The analyses aimed to identify if there was an association between cumulative external dose and excess risk considering other available factors that are known to affect the respiratory disease risk.

Results

There was very little radiation effect on overall respiratory disease mortality or pneumonia mortality, but evidence of a reduction in risk for COPD and allied disease (ERR/Sv= -0.56, 95%CI: -0.94, -0.06; P=0.02) and an increase in risk for other respiratory disease mortality (ERR/Sv=2.30, 95%CI: 0.67, 4.62; P=0.01) with increasing cumulative external dose were observed. The effects of radiation were more prominent amongst workers monitored for internal exposure.

There was good evidence for a reduction in risk of COPD with increasing cumulative external dose for radiation workers who were monitored for internal exposure (ERR/Sv=-0.59, 95%CI: -0.99, -0.05; p=0.017) but not among the workers who were not monitored (ERR/Sv=-0.43, 95%CI: -1.20, 0.74; p=0.42).

There was also good evidence for increased risk with increasing cumulative dose for the group of other respiratory diseases among monitored radiation workers (ERR/Sv=2.46, 95%CI: 0.69, 5.08; p=0.019), but not among unmonitored workers (ERR/Sv=1.70, 95%CI: -0.82, 5.65; p=0.25).

Conclusion

While there was no good evidence for an overall effect of radiation exposure on respiratory disease risk, there was evidence of an association (in opposite directions) for COPD and the category of other respiratory diseases with external dose among workers who were monitored for internal exposures.

Update to the NRRW third analysis 2022 – extended analysis of solid cancer incidence

Nezahat Hunter, Richard G E Haylock, Michael Gillies, Wei Zhang. Extended analysis of solid cancer incidence among the Nuclear Industry Workers in the UK: 1955-2011. Radiation Research.

The recent publication of the updated third analysis of the NRRW cohort provided the most precise estimates to date of the risks of mortality and cancer incidence from low dose occupational external radiation exposure in the UK. This publication describes a more detailed analysis of solid cancer incidence.

Background

Much of what is known about the long-term carcinogenic effects of radiation exposure comes from the Life Span Study cohort of the A-bomb survivors in Japan. The radiation dose received by these survivors was delivered acutely, primarily gamma-ray (external radiation).

Long-term follow-up of the health of radiation workers in the nuclear industry is important to determine whether risk estimates derived from the A-bomb survivor studies are directly applicable to the lower protracted external exposures experienced by workers.

This analysis uses the updated NRRW-3 data set to examine the shape of fit of various dose response relationships between external radiation exposure and solid cancer incidence to investigate if they provide a better description of the data than the linear function used in previous analysis. It also studies individual solid cancer sites in more detail and evaluates the impact of factors that may modify the dose-response such as age at first exposure, time since first exposure, duration of exposure and so on.

Methods

The analyses are based on the same NRRW-3 cohort updated with an additional 10 years of follow-up (as used in other analyses). As in previous analyses Poisson regression methods were used to evaluate quality of fit of different models. However, in contrast to previous analyses where a fully stratified baseline was used, here a parametric baseline was employed.

Results

A total of 18,310 cases of solid cancers were registered as their first primary cancers and of these 43% of the solid cancer cases occurred during the latest 10 years. Having adjusted for baseline factors known to affect cancer rates, this study demonstrated a non-linear dose-response for solid cancers provided a better description of the data than did a linear function.

The main difference being the reduction in risk at higher doses for the non-linear function. This finding appears to be driven by the workers who were monitored for potential exposure to internal emitters and who had also received relatively high cumulative external doses (above 400 mSv). There was evidence of linear dose-response being a better fit for cumulative doses below this. Among cohort members only exposed to external radiation, a strong linear association was found between external dose and solid cancers.

A similar non-linear pattern was observed for lung cancer. This suggests that the findings from lung cancer may be influencing the overall result for solid cancers. Excluding lung cancer from the grouping of all solid cancers results in an absence of evidence for a non-linear association being a better fit than a linear one. There was no evidence that the dose response function for solid cancers excluding lung varied with time since first exposure, age at first exposure or duration of employment.

Among 23 individual cancer sites, there were statistically significantly increasing linear trends with dose for cancers of the colorectal, bladder and pleura cancer.

Conclusion

It is likely that the absence of data on individual smoking habits and/or internal dose estimates may at least partly explain the non-linear dose response in lung cancer and the overall result for solid cancers. The observed increase in the risk of cancer of colorectal, bladder and pleura cancer should be treated with caution because of the limited corroborating evidence from other published studies.

There appears to be reasonably good agreement between our estimates based on limited data with the solid cancer risk estimate obtained in the A-bomb survivors’ study. Information on internal doses as well as non-radiation factors such as smoking would be very helpful to make more definitive inferences, in particular, for lung cancer. Even in the absence of this additional information, continued follow-up of the NRRW cohort should provide more precise results for these types of cancers and inform on the possibility that the pattern of risk, particularly at higher doses, are chance findings due to small numbers.

Update to the NRRW third analysis 2022 – lymphoma and multiple myeloma

The effect of external radiation exposure on the risks of developing lymphatic cancer, including non-Hodgkin lymphoma (NHL), Hodgkin lymphoma (HL) and multiple myeloma (MM) was evaluated using the NRRW based upon the third analysis cohort but with an additional 10 years of follow-up.

Nezahat Hunter and Richard Haylock. Radiation risks of lymphoma and multiple myeloma incidence in the updated NRRW-3 cohort in the UK: 1955–2011. Journal of Radiological Protection. 42 011517.

Background

The effects of radiation exposure on risk of lymphatic cancers (NHL and HL) and MM have been studied widely among the Japanese A-bomb survivors, among patients receiving radiotherapy and diagnostic irradiation (where radiation exposures were short and at high dose) and among large groups of radiation workers with protracted chronic low dose exposure.

Overall, the strength of the evidence from these studies was mixed. This analysis using the updated NRRW-3 data set provides further evidence about the risk of incidence of lymphatic cancers and multiple myeloma following occupational exposure to protracted external radiation.

Methods

The study comprises a large cohort of 172,452 workers, of whom 90% were men with follow-up from 1955 through to the end of 2011 who experienced a range of accumulated low doses; two-thirds of workers’ doses were less than 10 mSv and 6% of workers’ doses more than 100 mSv.

Poisson regression was used to estimate the excess relative risk (ERR) per unit of cumulative exposure to external radiation.

Results

The statistical precision of the estimates in this study was improved for NHL and MM with a longer follow-up of 10 years more in comparison with the previously published NRRW-3 study.

The number of cases was doubled in comparison with the original third NRRW for NHL from 305 to 711 cases, for HL 67 to 113 and for MM from 149 to 279 cases. Of these cases, 33, 5 and 10 cases are among females respectively.

There was evidence of an increasing trend with dose in incidence from NHL and MM. However, this association lost its significance in NHL and the evidence was weakened in MM when restricting the data used to doses under 0.5 Sv.

For HL there was no indication of any association with dose.

There was no evidence for temporal effects in the ERR for NHL or MM risk with attained age or time since exposure, indicating these effects are less marked in this study following radiation exposure for males of working age.

Conclusion

This study provides some new evidence for increased risk of NHL and MM among occupational exposed radiation workers.

However, the observed increase in the risk of cancer of NHL and MM should be treated with caution because despite the increase in numbers of cases compared to the original NRRW-3 analysis, the association is still driven by a very small number of cases at relatively high doses.

Continued follow-up in the NRRW cohort should provide more precise results for these types of cancers and inform on the possibility that these raised risks are chance findings due to small numbers, particularly at higher doses.

Update to the NRRW third analysis 2019 – leukaemia

In the third analysis of the NRRW cohort (NRRW-3), evidence was found of increased risk of leukaemia (excluding chronic lymphocytic leukaemia) associated with external radiation exposure that was consistent with estimates from the Life Span Study (LSS) of the Japanese atomic bomb survivors. However, the uncertainty on these estimates was large and there remained questions around how these risks varied over time and across sub-types of leukaemia.

These issues are examined in an updated analysis of leukaemia outcomes in the NRRW-3 cohort.

Background

In the first few years following the atomic bombings of Hiroshima and Nagasaki, an excess of leukaemia was observed amongst survivors and this excess persists in long term follow-up studies of the survivors.

While these studies helped to establish a link between leukaemia and acute radiation exposures, uncertainty remains around the translation of these risks to different populations and to the protracted low dose and dose-rate exposures typically received by workers in the nuclear industry.

The primary aim of the study was to examine any potential association between both leukaemia mortality and incidence rates and protracted low dose radiation exposure using the updated NRRW-3 cohort.

Due to indications from previous studies of the atomic-bomb survivors that radiation risks varied with both age and time since exposure, an aim of the study was also to examine temporal variation in the excess risk by each of these factors in the NRRW-3 cohort.

Methods

This study analysed leukaemia risk from occupational external radiation exposure in the updated NRRW cohort which comprised 173,081 workers first employed during the years 1947 to 1999 with follow-up until the end of 2011.

The study examined deaths and cancer incidence from all leukaemia combined, all leukaemia excluding chronic lymphatic leukaemia (non-CLL) and for the 4 main subtypes of leukaemia: acute lymphoblastic leukaemia (ALL), chronic lymphocytic leukaemia (CLL), acute myeloid leukaemia (AML), chronic myeloid leukaemia (CML), as well as unspecified leukaemia.

Using standard methods of statistical modelling, the dose-response relationship between external radiation exposure and leukaemia rate was derived.

Main results

Male workers showed evidence of raised rates of non-CLL leukaemia in relation to cumulative occupational radiation dose from external sources.

Evidence in female workers was limited by low power because of the low number of leukaemia cases.

The leukaemia risks estimates were consistent with the risk coefficients upon which radiation protection standards are based.

The raised non-CLL risks were driven primarily by increased risks for CML.

Much of the evidence for raised risks comes from workers who received cumulative doses in excess of 100 mSv, although there was no evidence of non-linearity in the dose-response for doses below this level.

The raised non-CLL risk was somewhat reduced when compared with previous NRRW results. This reduction in risk with extended follow-up was what you would expect to see based on previous observations that leukaemia risks are observed to decrease with age or time since exposure.

In this study, there was limited evidence of temporal variation in leukaemia risk with either age or time since exposure. However, the study has limited power to detect these effects at the current time.

Although raised risks of leukaemia were observed in relation to external dose, these increases were small. The leukaemia mortality rates in cohort remained below those of the general population while there was no evidence of that leukaemia incidence rates were above those in the general population.

Conclusions

This study provides further evidence that leukaemia risks may be increased by low-dose and protracted external radiation exposure.

The risks are generally consistent with those observed in the atomic bomb survivor studies and consistent with risk coefficients on which international radiation safety standards, including the dose limits and constraints used to control exposures, are based.

While there was some indication of temporal variation in leukaemia risk that are different to those derived from the LSS cohort, the power to detect such effects is limited and further follow-up of this or other cohorts incorporating further follow-up may be useful in this regard.

Update to the NRRW third analysis 2019 – heart disease

In the third analysis of the NRRW, published in 2009, there was only enough data to perform a simple analysis of heart diseases as a single group among cohort members. With 10 years extra follow-up there were now 11,014 heart disease deaths reported, allowing a more detailed analysis.

Zhang W, Haylock RGE, Gillies M, Hunter N. Mortality from heart diseases following occupational radiation exposure: analysis of the National Registry for Radiation Workers (NRRW) in the United Kingdom. Journal of Radiological Protection, 39 (2).

Background

Studies of high dose and high dose rate exposure to ionising radiation, like studies of breast cancer patients, have shown increased risk of heart disease (HD) in subsequent years. However, evidence of increased HD risk following external radiation doses accumulated at a low dose rate over many years, as occurs with occupational exposure, remains equivocal.

This study provides direct evidence about HD risk from cumulative external radiation exposures in an occupational setting.

Methods

HD mortality was studied in relation to cumulative occupational radiation dose among 174,541 subjects.

These subjects were mainly exposed to protracted low doses over a few years, and were followed up until the end of 2011.

HD as a single group was studied, as were the major HD subtypes.

Results

There were 11,014 HD deaths included in the analysis. Of these, 9,814 deaths or 89% were attributed to ischemic heart disease (IHD).

The ERR/Sv for all HD as a single group was significantly raised (ERR/Sv = 0.37, 95%CI: 0.11, 0.65; p = 0.005).

Among the major subgroups of HD, statistically significantly raised risk was found for IHD (ERR/Sv = 0.32; 95%CI 0.04, 0.61; p = 0.02).

For IHD, the raised radiation-associated risk first appeared at least 20 years after the first radiation exposure, and a categorical analysis also showed that the risk falls below the expected value based on a linear trend for cumulative doses greater than 0.4 Sv.

This small risk appears to be mainly associated with workers who started employment at a younger age and were employed for more than 30 years, reflecting possible healthy worker survivor effect.

Conclusions

This analysis demonstrates a statistically significant increase in the risk of HD mortality overall and for IHD with cumulative external dose among UK radiation workers in the NRRW.

Further follow-up of this cohort and collecting more information about potential confounders associated with lifestyle will help derive more accurate estimates of radiation-associated risk at low exposure levels, as experienced by radiation workers and the public.

Update to the NRRW third analysis 2018

This analysis is based on the same cohort considered in the third analysis but with follow-up extended by an additional 10 years. The extra follow-up provides increased statistical power and should enable the detection of potential radiation-related risks with greater precision than reported previously.

Haylock RGE, Gillies M, Hunter N, Zhang W, Phillipson M. Cancer mortality and incidence following external occupational radiation exposure: an update of the 3rd analysis of the UK national registry for radiation workers. British Journal of Cancer, 119, pp. 631–637.

Background

This study provides direct evidence of cancer risk from low dose and dose rate occupational external radiation exposures.

Methods

Cancer mortality and incidence were studied in relation to external radiation exposure in the National Registry for Radiation Workers. A cohort of 167,003 workers followed for an average of 32 years was analysed using Poisson regression methods.

Results

Mortality and incidence risks were significantly raised for the group of all malignant neoplasms excluding leukemia (ERR/Sv mortality = 0.28; 90%CI: 0.06, 0.53, ERR/Sv incidence = 0.28; 90%CI: 0.10, 0.48) but with narrower confidence bounds compared with the previous analysis of this cohort reflecting the increased statistical power from the additional 10 years of follow-up information.

The linear trends in relative risk for both mortality and incidence of these cancers remained statistically significantly raised when information relating to cumulative doses above 100 mSv was excluded (ERR/Sv mortality = 1.42; 90%CI: 0.51, 2.38 and ERR/Sv incidence = 1.18; 90%CI: 0.47, 1.92).

Conclusions

This study improved the precision of the cancer risk estimates seen in the third analysis of the NRRW cohort.

The overall results remain consistent with the risk estimates from the LSS and those adopted in the current International Commission on Radiological Protection (ICRP) recommendations.

NRRW third analysis 2009

The third analysis of the NRRW was published in 2009. A scientific paper was published in the British Journal of Cancer and a much more detailed report was published by the HPA.

Muirhead CR, O’Hagan JA, Haylock RGE, Phillipson MA, Willcock T, Berridge GLC and Zhang W. Mortality and cancer incidence following occupational radiation exposure: 3rd analysis of the National Registry for Radiation Workers. British Journal of Cancer, 100, 206 to 212 (2009).

Muirhead CR, O’Hagan JA, Haylock RGE, Phillipson MA, Willcock T, Berridge GLC and Zhang W. HPA-RPD-062: third analysis of the National Registry for Radiation Workers: occupational exposure to ionising radiation in relation to mortality and cancer incidence.

NRRW second analysis 1999

The second NRRW analysis was published in 1999 as a full NRPB report and a shorter scientific paper in a peer-reviewed journal.

Kendall G M, Muirhead C R, MacGibbon B H, O’Hagan J A, Conquest A J, Goodill A A and others. Mortality and occupational exposure to radiation: first analysis of the National Registry for Radiation Workers. British Medical Journal 1992; 304:220.

Kendall GM, Muirhead CR, MacGibbon BH, O’Hagan JA, Conquest AJ, Goodill AA, Butland BK, Fell TP, Jackson DA, Webb MA, Haylock RGE, Thomas JM and Silk TJ. First analysis of the National Registry for Radiation Workers: occupational exposure to ionising radiation and mortality. Chilton, NRPB-R251 (London, HMSO) 1992.

Muirhead CR, Goodill AA, Haylock RG, Vokes J, Little MP, Jackson DA, O’Hagan JA, Thomas JM, Kendall GM, Silk TJ, Bingham D, Berridge GL. Occupational radiation exposure and mortality: second analysis of the National Registry for Radiation Workers. Journal of Radiological Protection. 1999, March;19(1):3-26.

NRRW first analysis 1992

The first NRRW analysis was published in 1992 as a full NRPB report and a shorter scientific paper in a peer-reviewed journal.

Kendall G M, Muirhead C R, MacGibbon B H, O’Hagan J A, Conquest A J, Goodill A A and others. Mortality and occupational exposure to radiation: first analysis of the National Registry for Radiation Workers. British Medical Journal 1992; 304:220.

Kendall GM, Muirhead CR, MacGibbon BH, O’Hagan JA, Conquest AJ, Goodill AA, Butland BK, Fell TP, Jackson DA, Webb MA, Haylock RGE, Thomas JM and Silk TJ. First analysis of the National Registry for Radiation Workers: occupational exposure to ionising radiation and mortality. Chilton, NRPB-R251 (London, HMSO) 1992.