International Variation in Criteria for Internal Mammary Chain Radiotherapy

AIMS

Evidence has emerged that internal mammary chain (IMC) radiotherapy reduces breast cancer mortality, leading to changes in treatment guidelines. This study investigated current IMC radiotherapy criteria and the percentages of patients irradiated for breast cancer in England who fulfilled them.

MATERIALS AND METHODS

A systematic search was undertaken for national guidelines published in English during 2013-2018 presenting criteria for 'consideration of' or 'recommendation for' IMC radiotherapy. Patient and tumour variables were collected for patients who received breast cancer radiotherapy in England during 2012-2016. The percentages of patients fulfilling criteria stipulated in each set of guidelines were calculated.

RESULTS

In total, 111 729 women were recorded as receiving adjuvant breast cancer radiotherapy in England during 2012-2016 and full data were available on 48 095 of them. Percentages of patients fulfilling IMC radiotherapy criteria in various national guidelines were: UK Royal College of Radiologists 13% (6035/48 095), UK National Institute for Health and Care Excellence 18% (8816/48 095), Germany 32% (15 646/48 095), Ireland 56% (26 846/48 095) and USA 59% (28 373/48 095). Differences between countries occurred because in Ireland and the USA, treatment may be considered in some node-negative patients, whereas in the UK, treatment is considered if at least four axillary nodes are involved or for high-risk patients with one to three positive nodes. In Germany, treatment may be considered for all node-positive patients.

CONCLUSIONS

There is substantial variability between countries in criteria for consideration of IMC radiotherapy, despite guidelines being based on the same evidence. This will probably lead to large variations in practice and resource needs worldwide.

Abstract P2-03-10: Risk factors for upgrading and upstaging of pre-operative biopsies in ductal carcinoma in situ

Background: Ductal carcinoma in situ (DCIS), accounts for one fifth of all screen-detected neoplastic breast lesions. Contemporary research in DCIS focuses on separating lesions that need active treatment from those that can be safely left under surveillance. This, in turn, relies on accurate determination of invasive status and DCIS grade at time of initial biopsy. Most previous studies have examined factors associated with upstaging the diagnosis from DCIS to invasive breast cancer (IBC) following surgery, and few have evaluated factors associated with upgrading the diagnosis to a higher grade of DCIS. This is because upgrading has not traditionally influenced clinical management in the way that upstaging has done. However, recent interest in non-operative treatment for low-risk DCIS has meant that accurate determination of grade at time of initial biopsy has become more important. We aimed to compare risk factors for upgrading and upstaging of biopsies in DCIS.

Method: We undertook a cohort study of all women diagnosed with DCIS at a large specialist cancer centre between 2000–2014. Information from the clinical records was abstracted, including the pre-operative mammography (MMG) and pathology information from the initial biopsy. We also abstracted pathology information regarding the excised specimen in order to identify women whose diagnosis was subsequently upgraded or upstaged. We looked for factors that were predictive for upgrading or upstaging.

Result: A total of 641 women were diagnosed with DCIS at initial biopsy. Of these, 72 (11%) were upgraded: 26 (4%) from grade 1 to grade 2, 2 (0.3%) from grade 1 to grade 3 and 44 (7%) from grade 2 to grade 3. A further 115 (18%) were upstaged to IBC: 20 of these (3%) had grade 1 DCIS on initial biopsy, 47 (7%) had grade 2, 43 (7%) grade 3, and for 5 (1%) biopsy grade was not available. Necrosis on biopsy increased the risk of upgrading (with necrosis: 14% upgraded, without: 10% upgraded, p for difference 0.02) and also of upstaging (with necrosis: 23% upstaged, without: 15% upstaged, p for difference <0.01). Lesions measuring ≥50 mm on MMG were more likely to be upgraded than smaller lesions (0-19 mm: 9% upgraded, 20-50 mm: 9% upgraded, ≥50 mm: 19% upgraded, p for heterogeneity <0.01), while lesions measuring 20-50 mm and ≥50 mm were both more likely to be upstaged than lesions measuring 0-19 mm (0-19 mm: 9% upstaged, 20-50 mm: 23% upstaged and ≥50 mm: 21% upstaged, p for heterogeneity <0.01). Fewer 9G vacuum-assisted biopsies than 14G core biopsies were upgraded (9G vacuum-assisted: 7% upgraded, 14G core: 15% upgraded, p for difference 0.01), while the effect of biopsy method on upstaging was not significant (9G vacuum-assisted: 12% upstaged, 14G core: 16% upstaged, p for difference 0.15). Presence of a palpable lump was not significantly associated with upgrading (palpable lump: 13% upgraded, no palpable lump: 10% upgraded, p for difference 0.19) but increased the risk of upstaging (palpable lump: 23% upstaged, no palpable lump: 16% upstaged, p for difference 0.02).

Conclusion: Our findings suggest that consideration of MMG lesion size and necrosis on biopsy may be helpful in selecting low-risk women for non-operative management of DCIS, as may use of the 9G vacuum-assisted method of biopsy.

Citation Format: Mannu GS, Groen E, Wang Z, Schaapveld M, Lips E, Chung M, Joore I, Leeuwen Fv, Teerstra J, Winter-Warnars GAO, Darby SC, Wesseling J. Risk factors for upgrading and upstaging of pre-operative biopsies in ductal carcinoma in situ [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P2-03-10.

Impact of smoking on mortality and life expectancy in Japanese smokers: a prospective cohort study

OBJECTIVE

To investigate the impact of smoking on overall mortality and life expectancy in a large Japanese population, including some who smoked throughout adult life.

DESIGN

The Life Span Study, a population-based prospective study, initiated in 1950.

SETTING

Hiroshima and Nagasaki, Japan.

PARTICIPANTS

Smoking status for 27,311 men and 40,662 women was obtained during 1963-92. Mortality from one year after first ascertainment of smoking status until 1 January 2008 has been analysed.

MAIN OUTCOME MEASURES

Mortality from all causes in current, former, and never smokers.

RESULTS

Smokers born in later decades tended to smoke more cigarettes per day than those born earlier, and to have started smoking at a younger age. Among those born during 1920-45 (median 1933) and who started smoking before age 20 years, men smoked on average 23 cigarettes/day, while women smoked 17 cigarettes/day, and, for those who continued smoking, overall mortality was more than doubled in both sexes (rate ratios versus never smokers: men 2.21 (95% confidence interval 1.97 to 2.48), women 2.61 (1.98 to 3.44)) and life expectancy was reduced by almost a decade (8 years for men, 10 years for women). Those who stopped smoking before age 35 avoided almost all of the excess risk among continuing smokers, while those who stopped smoking before age 45 avoided most of it.

CONCLUSIONS

The lower smoking related hazards reported previously in Japan may have been due to earlier birth cohorts starting to smoke when older and smoking fewer cigarettes per day. In Japan, as elsewhere, those who start smoking in early adult life and continue smoking lose on average about a decade of life. Much of the risk can, however, be avoided by giving up smoking before age 35, and preferably well before age 35.

Measurement error in the explanatory variable of a binary regression: regression calibration and integrated conditional likelihood in studies of residential radon and lung cancer

In epidemiology, one approach to investigating the dependence of disease risk on an explanatory variable in the presence of several confounding variables is by fitting a binary regression using a conditional likelihood, thus eliminating the nuisance parameters. When the explanatory variable is measured with error, the estimated regression coefficient is biased usually towards zero. Motivated by the need to correct for this bias in analyses that combine data from a number of case-control studies of lung cancer risk associated with exposure to residential radon, two approaches are investigated. Both employ the conditional distribution of the true explanatory variable given the measured one. The method of regression calibration uses the expected value of the true given measured variable as the covariate. The second approach integrates the conditional likelihood numerically by sampling from the distribution of the true given measured explanatory variable. The two approaches give very similar point estimates and confidence intervals not only for the motivating example but also for an artificial data set with known properties. These results and some further simulations that demonstrate correct coverage for the confidence intervals suggest that for studies of residential radon and lung cancer the regression calibration approach will perform very well, so that nothing more sophisticated is needed to correct for measurement error.

Adjuvant chemotherapy in oestrogen-receptor-poor breast cancer: patient-level meta-analysis of randomised trials

BACKGROUND

The long-term effects of adjuvant polychemotherapy regimens in oestrogen-receptor-poor (ER-poor) breast cancer, and the extent to which these effects are modified by age or tamoxifen use, can be assessed by an updated meta-analysis of individual patient data from randomised trials.

METHODS

Collaborative meta-analyses of individual patient data for about 6000 women with ER-poor breast cancer in 46 trials of polychemotherapy versus not (non-taxane-based polychemotherapy, typically about six cycles; trial start dates 1975-96, median 1984) and about 14 000 women with ER-poor breast cancer in 50 trials of tamoxifen versus not (some trials in the presence and some in the absence of polychemotherapy; trial start dates 1972-93, median 1982).

FINDINGS

In women with ER-poor breast cancer, polychemotherapy significantly reduced recurrence, breast cancer mortality, and death from any cause, in those younger than 50 years and those aged 50-69 years at entry into trials of polychemotherapy versus not. In those aged younger than 50 years (1907 women, 15% node-positive), the 10-year risks were: recurrence 33% versus 45% (ratio of 10-year risks 0.73, 2p<0.00001), breast cancer mortality 24% versus 32% (ratio 0.73, 2p=0.0002), and death from any cause 25% versus 33% (ratio 0.75, 2p=0.0003). In women aged 50-69 years (3965 women, 58% node-positive), the 10-year risks were: recurrence 42% versus 52% (ratio 0.82, 2p<0.00001), breast cancer mortality 36% versus 42% (ratio 0.86, 2p=0.0004), and death from any cause 39% versus 45% (ratio 0.87, 2p=0.0009). Few were aged 70 years or older. Tamoxifen had little effect on recurrence or death in women who were classified in these trials as having ER-poor disease, and did not significantly modify the effects of polychemotherapy.

INTERPRETATION

In women who had ER-poor breast cancer, and were either younger than 50 years or between 50 and 69 years, these older adjuvant polychemotherapy regimens were safe (ie, had little effect on mortality from causes other than breast cancer) and produced substantial and definite reductions in the 10-year risks of recurrence and death. Current and future chemotherapy regimens could well yield larger proportional reductions in breast cancer mortality.

Cardiac Risks of Breast-cancer Radiotherapy: A Contemporary View

For some time, there has been compelling evidence both from randomised-controlled trials and from observational studies, that some of the breast-cancer radiotherapy regimens used in the past have led to increased risk of mortality from heart disease. There is also some evidence that the more recent regimens used in the USA are associated with lower risks than previous ones, but it is not clear whether current regimens are free from cardiac risk, especially in the light of recent evidence from the survivors of the bombings of Hiroshima and Nagasaki, in whom a clear relationship was observed between the risk of mortality from heart disease and radiation dose for doses in the range 0-4 Gy. Mortality from radiation-induced heart disease usually occurs at least a decade after irradiation. Symptomatic heart disease might have a much shorter induction period, but little information about it is available at present. Subclinical vascular abnormalities have been observed within months of irradiation, via myocardial perfusion imaging studies, but little is known about the relationship between these and later overt heart disease. At present, few data relate heart dose and other specific characteristics of breast radiotherapy to cardiac outcome. Further information on these topics is needed to enable estimation of the cardiac risk, that is likely to arise from radiotherapy regimens in current use and from those being considered for future use. Such knowledge would facilitate radiotherapy treatment planning and enable a reduction in cardiac risk while maintaining the known benefit in terms of breast cancer mortality.

Epidemiological studies of UK test veterans: II. Mortality and cancer incidence

An epidemiological study was set up in the 1980s of UK participants in the UK atmospheric nuclear weapons testing programme. A large cohort of test participants was established along with a closely matched comparison or control group. Three analyses of mortality and cancer incidence have been carried out. This review describes the development of the evidence on possible effects on test participants with especial emphasis on the most recent analysis. Other sources of evidence, particularly from studies of other groups of test participants, are also considered. It was concluded that overall levels of mortality and cancer incidence in UK nuclear weapons test participants were similar to those in a matched control group, and overall mortality was lower than expected from national rates. There was no evidence of an increased raised risk of multiple myeloma among test participants in recent years, and the suggestion in the first analysis of this cohort of a raised myeloma risk relative to controls is likely to have been a chance finding. There was some evidence of a raised risk of leukaemia other than chronic lymphatic leukaemia among test participants relative to controls, particularly in the early years after the tests. Whilst this could be a chance finding, the possibility that test participation caused a small absolute risk of leukaemia other than chronic lymphatic leukaemia cannot be ruled out.

Epidemiological studies of UK test veterans: I. General description

This review gives a general account of how and why epidemiological studies of UK participants in the nuclear weapons test programme were set up. There is a short description of the circumstances in which the tests were planned and executed and a discussion of the general considerations involved in designing studies to show whether the health of test participants suffered as a result of the tests. The companion review article summarises the results of the epidemiological studies.

The incidence of factor VIII and factor IX inhibitors in the hemophilia population of the UK and their effect on subsequent mortality, 1977-99

BACKGROUND

Previous studies of the development of inhibitors and their impact on mortality have been small.

OBJECTIVES

To examine the development of inhibitors in people with hemophilia in the UK and their effect on subsequent mortality.

PATIENTS

6078 males with hemophilia A and 1172 males with hemophilia B registered in the UK Haemophilia Centre Doctors' Organisation database, 1977-98.

RESULTS

In severe hemophilia A inhibitors developed at rates of 34.4, 5.2 and 3.8 per 1000 years at ages <5, 5-14 and 15+years; cumulative risks at ages 5 and 75 were 16% and 36%. In hemophilia A the rate of inhibitor development decreased during 1977-90, but increased during the 1990s. In severe hemophilia B inhibitors developed at rates of 13.3 and 0.2 per 1000 years at ages <5 and 5+ and cumulative risks at ages 5 and 75 were 6% and 8%. With HIV, inhibitor development did not increase mortality. In severe hemophilia without HIV, inhibitor development doubled mortality during 1977-92, but during 1993-99 mortality was identical with and without inhibitors. In severe hemophilia without HIV but with inhibitors, mortality from causes involving bleeding decreased during 1977-99 (P = 0.001) as did mortality involving intracranial hemorrhage (P = 0.007).

CONCLUSIONS

These data provide estimates of the rate of inhibitor development in hemophilia A and hemophilia B, and they show that the rate of inhibitor development has varied over time, although the reasons for this remain unclear. They also show that in severe hemophilia the substantial increase in mortality previously associated with inhibitors is no longer present.

Health effects of residential radon: a European perspective at the end of 2002

Surveys of indoor radon concentrations, when taken together with estimates of the risk of lung cancer from studies in miners of uranium and other hard rocks, suggest that residential radon is responsible for many thousands of deaths from lung cancer each year in Europe. The vast majority of these deaths are likely to occur in individuals who also smoke cigarettes. Because of the skewed nature of the distribution of the indoor radon concentrations in most populations, most of the deaths will occur in individuals who are exposed at moderate rather than at very high radon concentrations. In order to enable appropriate policies to be developed for managing the consequences of exposure to radon, more reliable estimates of the risk of lung cancer resulting from it are needed. To achieve this, a European Collaborative Group on Residential Radon and Lung Cancer was initiated and its findings should be published in 2004.

100 years of observation on British radiologists: mortality from cancer and other causes 1897-1997

Radiologists and radiotherapists were one of the earliest occupational groups to be exposed to ionizing radiation. Their patterns of mortality provide information on the long-term effects of fractionated external radiation exposure. British radiologists who registered with a radiological society between 1897 and 1979 have now been followed-up until 1 January 1997, and the mortality experience examined among those who registered with a society after 1920, when the first radiological protection recommendations were published. The observed number of cancer deaths in those who registered after 1920 was similar to that expected from death rates for all medical practitioners combined (SMR=1.04; 95% CI 0.89-1.21). However, there was evidence of an increasing trend in risk of cancer mortality with time since first registration with a radiological society (p=0.002), such that in those registered for more than 40 years there was a 41% excess risk of cancer mortality (SMR=1.41; 95% CI 1.03-1.90). This is probably a long-term effect of radiation exposure in those who first registered during 1921-1935 and 1936-1954. There was no evidence of an increase in cancer mortality among radiologists who first registered after 1954, in whom radiation exposures are likely to have been lower. Non-cancer causes of death were also examined in more detail than has been reported previously. There was no evidence of an effect of radiation on diseases other than cancer even in the earliest radiologists, despite the fact that doses of the size received by them have been associated with more than a doubling in the death rate among the survivors of the Japanese atomic bombings.

Lung cancer and cigarette smoking in Europe: an update of risk estimates and an assessment of inter-country heterogeneity

Ten case-control studies have been carried out in 6 European countries to investigate the major risk factors for lung cancer. Carcinogenic effect from cigarette smoke was the most relevant interest in our study, which has included 7,609 cases of lung cancer and 10,431 controls, mainly population based. The results indicate elevated odds ratios (ORs; 23.9 among men and 8.7 among women) with attributable risks exceeding 90% for men and close to 60% for women. A large, and statistically significant, variability of the results across countries was detected after adjusting for the most common confounding variables, and after controlling, at least in part, for the instability of the ORs due to the small number of non-smokers in some of the study subsets. This pattern of lung cancer risk associated with cigarettes smoke, across different European regions, reflects inherent characteristics of the studies as well as differences in smoking habits, particularly calendar periods of starting, and it is likely to have been influenced by effect modifiers like indoor radon exposure, occupation, air pollution and dietary habits.

Lung cancer and cigarette smoking in women: a multicenter case-control study in Europe

The association between cigarette smoking and lung cancer risk in women was investigated within the framework of a case-control study in 9 centres from 6 European countries. Cases were 1,556 women up to 75 years of age with histologically confirmed primary lung cancer; 2, 450 controls with age distribution similar to cases were selected. The predominant cell type was adenocarcinoma (33.5%), with similar proportions for squamous-cell type (26.4%) and small-cell carcinoma (22.3%). Overall, smoking cigarettes at any time was associated with a 5-fold increase in lung cancer risk (odds ratio 5.21, 95% confidence interval 4.49-6.04); corresponding figures for current smoking habits were 8.94, 7.54-10.6. The association showed a dose-response relationship with duration of the habit and daily and cumulative lifetime smoking. A significant excess risk of 70% was associated with every 10 pack-years smoked. After 10 years of smoking cessation, the relative risk decreased to 20% compared to current smokers. The following characteristics were associated with a higher relative risk: inhalation of smoke, smoking non-filter cigarettes, smoking dark-type cigarettes and starting at young age. The association was observed for all major histological types, being the strongest for small-cell type carcinoma, followed by squamous-cell type and the lowest for adenocarcinoma. The proportion of lung-cancer cases in the population attributable to cigarette smoking ranged from 14% to 85%. We concluded that women share most features of the association between cigarette smoking and lung cancer observed in men.

Risks of leukemia in Japanese atomic bomb survivors, in women treated for cervical cancer, and in patients treated for ankylosing spondylitis

The dose-response relationship for radiation-induced leukemia was examined in a pooled analysis of three exposed populations: Japanese atomic bomb survivors, women treated for cervical cancer, and patients irradiated for ankylosing spondylitis. A total of 383 leukemias were observed among 283,139 study subjects. Considering all leukemias apart from chronic lymphocytic leukemia, the optimal relative risk model had a dose response with a purely quadratic term representing induction and an exponential term consistent with cell sterilization at high doses; the addition of a linear induction term did not improve the fit of the model. The relative risk decreased with increasing time since exposure and increasing attained age, and there were significant (P < 0.00001) differences in the parameters of the model between datasets. These differences were related in part to the significant differences (P = 0.003) between the models fitted to the three main radiogenic leukemia subtypes (acute myeloid leukemia, acute lymphocytic leukemia, chronic myeloid leukemia). When the three datasets were considered together but the analysis was repeated separately for the three leukemia subtypes, for each subtype the optimal model included quadratic and exponential terms in dose. For acute myeloid leukemia and chronic myeloid leukemia, there were reductions of relative risk with increasing time after exposure, whereas for acute lymphocytic leukemia the relative risk decreased with increasing attained age. For each leukemia subtype considered separately, there was no indication of a difference between the studies in the relative risk and its distribution as a function of dose, age and time (P > 0.10 for all three subtypes). The nonsignificant indications of differences between the three datasets when leukemia subtypes were considered separately may be explained by random variation, although a contribution from differences in exposure dose-rate regimens, inhomogeneous dose distribution within the bone marrow, inadequate adjustment forcell sterilization effects, or errors in dosimetry could have played a role.

Intercomparison of passive radon-detectors under field conditions in epidemiological studies

The Ardennes and Eifel region is a geologically distinct area covering parts of Germany, Belgium, France, and Luxembourg where enhanced concentrations of radon occur in some houses and other buildings. An international case-control study is being conducted to examine the role of radon in the etiology of lung cancer in this area. The radon detectors used are issued by different laboratories involving a variety of detector types and processes. A series of intercomparisons in houses was therefore conducted under similar conditions of exposure in the field. In most situations the different detectors gave similar results. Nevertheless, in some situations open and closed detectors yielded different results. Therefore, estimates of radon exposure have to be adjusted if results are to be pooled.

Some aspects of measurement error in explanatory variables for continuous and binary regression models

A simple form of measurement error model for explanatory variables is studied incorporating classical and Berkson cases as particular forms, and allowing for either additive or multiplicative errors. The work is motivated by epidemiological problems, and therefore consideration is given not only to continuous response variables but also to logistic regression models. The possibility that different individuals in a study have errors of different types is also considered. The relatively simple estimation procedures proposed for use with cohort data and case-control data are checked by simulation, under the assumption of various error structures. The results show that even in situations where conventional analysis yields slope estimates that are on average attenuated by a factor of approximately 50 per cent, estimates obtained using the proposed amended likelihood functions are within 5 per cent of their true values. The work was carried out to provide a method for the analysis of lung cancer risk following residential radon exposure, but it should be applicable to a wide variety of situations.

Multicenter case-control study of exposure to environmental tobacco smoke and lung cancer in Europe

BACKGROUND

An association between exposure to environmental tobacco smoke (ETS) and lung cancer risk has been suggested. To evaluate this possible association better, researchers need more precise estimates of risk, the relative contribution of different sources of ETS, and the effect of ETS exposure on different histologic types of lung cancer. To address these issues, we have conducted a case-control study of lung cancer and exposure to ETS in 12 centers from seven European countries.

METHODS

A total of 650 patients with lung cancer and 1542 control subjects up to 74 years of age were interviewed about exposure to ETS. Neither case subjects nor control subjects had smoked more than 400 cigarettes in their lifetime.

RESULTS

ETS exposure during childhood was not associated with an increased risk of lung cancer (odds ratio [OR] for ever exposure = 0.78; 95% confidence interval [CI] = 0.64-0.96). The OR for ever exposure to spousal ETS was 1.16 (95% CI = 0.93-1.44). No clear dose-response relationship could be demonstrated for cumulative spousal ETS exposure. The OR for ever exposure to workplace ETS was 1.17 (95% CI = 0.94-1.45), with possible evidence of increasing risk for increasing duration of exposure. No increase in risk was detected in subjects whose exposure to spousal or workplace ETS ended more than 15 years earlier. Ever exposure to ETS from other sources was not associated with lung cancer risk. Risks from combined exposure to spousal and workplace ETS were higher for squamous cell carcinoma and small-cell carcinoma than for adenocarcinoma, but the differences were not statistically significant.

CONCLUSIONS

Our results indicate no association between childhood exposure to ETS and lung cancer risk. We did find weak evidence of a dose-response relationship between risk of lung cancer and exposure to spousal and workplace ETS. There was no detectable risk after cessation of exposure.

Mortality from liver cancer and liver disease in haemophilic men and boys in UK given blood products contaminated with hepatitis C. UK Haemophilia Centre Directors' Organisation

BACKGROUND

Most people with haemophilia who were treated with blood products before the introduction of virus-inactivation procedures were infected with the hepatitis-C virus (HCV). However, there is little quantitative information about the long-term effects on mortality of such infection.

METHODS

We carried out a cohort study of mortality from liver cancer and liver disease in 4865 haemophilic men and boys in the UK. They were treated between 1969 and 1985 with blood products carrying a high risk of HCV infection, and were followed up from first recorded exposure to Jan 1, 1993.

FINDINGS

Based on death-certificate information, mortality was 16.7 times higher than in the general population for liver disease (95% CI 12.5-22.0; 51 deaths), and 5.6 times higher (1.8-13.0; five deaths) for liver cancer. For men and boys with severe haemophilia who were not infected with HIV-1, the cumulative risks of death from chronic or unspecified liver disease or from liver cancer in the 25 years since first recorded exposure to high HCV-risk products were 1.4% (0.7-3.0) at all ages, and 0.10% (0.01-0.7), 2.2% (0.8-6.1), and 14.3% (4.5-40.9) for those with first recorded exposure at ages under 25, 25-44, and 45 or older. For those with haemophilia and HIV-1 infection, the corresponding risks were 6.5% (4.5-9.5) at all ages, and 3.8% (2.1-6.8), 17.1% (10.0-28.5), and 18.7% (6.4-47.6) in the three age-groups. In those with severe haemophilia, age-standardised all-cause mortality was stable during 1969-84 but increased during 1985-92 in both HIV-1-infected and HIV-1-uninfected groups. Among those not infected with HIV-1, the increase in all-cause mortality resulted largely from deaths attributed to chronic or unspecified liver disease or liver cancer in men aged over 45.

INTERPRETATION

There is an emerging risk of mortality from liver disease and liver cancer in the UK haemophilia population in individuals both infected and uninfected with HIV-1, which probably results from infection with hepatitis C.

Importance of age at infection with HIV-1 for survival and development of AIDS in UK haemophilia population. UK Haemophilia Centre Directors' Organisation

BACKGROUND

Greater age at infection with HIV-1 is known to be associated with shorter time to development of AIDS, but the size of the differences between people infected in infancy and those infected in old age has not been examined in a single large population of patients with death as an endpoint. We, therefore, investigated the role of age at seroconversion in the entire UK population of haemophiliacs.

METHODS

We studied 1216 HIV-1-infected haemophilia patients in the UK who were registered with the National Haemophilia Register and were alive on Jan 1, 1985. Their estimated ages at HIV-1 seroconversion ranged from 8 months to 79 years.

FINDINGS

10 years after seroconversion 67% (95% Cl 64-69) of the population were still alive. Survival was strongly related to age at seroconversion (86% [82-90], 72% [68-76], 45% [39-51], and 12% [5-21] at 10 years among those patients who seroconverted at ages < 15, 15-34, 35-54, and > or = 55). This steep age-gradient in survival was not explained by deaths expected in the absence of HIV infection or by confounding with other factors such as haemophilia type or severity. The age-gradient was steeper for survival (ie, time from HIV-1 infection to death) than for time to diagnosis of AIDS, partly because survival after an AIDS diagnosis was poorer in older patients, and there was also a substantial increase in mortality among HIV-infected patients who did not satisfy the formal AIDS definition and this increase was greater in older patients.

INTERPRETATION

Age at infection with HIV-1 is a more important determinant of survival than has previously been appreciated. Age should, therefore, be considered in future studies of disease progression, and studies that compare people infected at different ages should provide insight into the biology of the immune response to HIV-1.

Ionizing radiation: future etiologic research and preventive strategies

Estimates of cancer risks following exposure to ionizing radiation traditionally have been based on the experience of populations exposed to substantial (and known) doses delivered over short periods of time. Examples include survivors of the atomic bombings at Hiroshima and Nagasaki, and persons treated with radiation for benign or malignant disease. Continued follow-up of these populations is important to determine the long-term effects of exposure in childhood, to characterize temporal patterns of excess risk for different types of cancer, and to understand better the interactions between radiation and other host and environmental factors. Most population exposure to radiation occurs at very low dose rates. For low linear energy transfer (LET) radiations, it often has been assumed that cancer risks per unit dose are lower following protracted exposure than following acute exposure. Studies of nuclear workers chronically exposed over a working lifetime provide data that can be used to test this hypothesis, and preliminary indications are that the risks per unit dose for most cancers other than leukemia are similar to those for acute exposure. However, these results are subject to considerable uncertainty, and further information on this question is needed. Residential radon is the major source of population exposure to high-LET radiation. Current estimates of the risk of lung cancer due to residential exposure to radon and radon daughters are based on the experience of miners exposed to much higher concentrations. Data indicate that lung cancer risk among miners is inversely associated with exposure rate, and also is influenced by the presence of other lung carcinogens such as arsenic in the mine environment. Further study of populations of radon-exposed miners would be informative, particularly those exposed at below-average levels. More direct evidence on the effects of residential exposure to radon also is desirable but might be difficult to come by, as risks associated with radon levels found in most homes might be too low to be quantified accurately in epidemiological studies.

Mortality before and after HIV infection in the complete UK population of haemophiliacs. UK Haemophilia Centre Directors' Organisation

During 1977-91, 6,278 males diagnosed with haemophilia were living in the UK. During 1979-86, 1,227 were infected with the human immunodeficiency virus (HIV-1) as a result of transfusion therapy (median estimated seroconversion date, October 1982). Among 2,448 with severe haemophilia, the annual death rate was stable at 8 per 1,000 during 1977-84; during 1985-92 death rates remained at 8 per 1,000 among HIV-seronegative patients but rose steeply in seropositive patients, reaching 81 per 1,000 in 1991-92. Among 3,830 with mild or moderate haemophilia, the pattern was similar, with an initial death rate of 4 per 1,000 in 1977-84, rising to 85 per 1,000 in 1991-92 in seropositive patients. During 1985-92, there were 403 deaths in HIV seropositive patients, whereas 60 would have been predicted from rates in seronegatives, suggesting that 85% of the deaths in seropositive patients were due to HIV infection. Most of the excess deaths were certified as due to AIDS or to conditions recognized as being associated with AIDS.

Leukemia mortality after X-ray treatment for ankylosing spondylitis

Leukemia mortality has been studied in 14,767 adult ankylosing spondylitis patients diagnosed between 1935 and 1957 in the United Kingdom, of whom 13,914 patients received X-ray treatment. By 1 January 1992, there were 60 leukemia deaths among the irradiated patients, almost treble that expected from national rates. Leukemia mortality was not increased among unirradiated patients. Among those irradiated, the ratio of observed to expected deaths for leukemia other than chronic lymphocytic leukemia was greatest in the period 1-5 years after the first treatment (ratio = 11.01, 95% confidence interval 5.26-20.98) and decreased to 1.87 (95% confidence interval 0.94-3.36) in the 25+ year period. There was no significant variation in this ratio with sex or age at first treatment. The ratio for chronic lymphocytic leukemia was slightly but not significantly raised (ratio = 1.44, 95% confidence interval 0.62-2.79). Most irradiated patients received all their exposure within a year. Based on a 1 in 15 random sample, the mean total marrow dose was 4.38 Gy. Doses were nonuniform, with heaviest doses to the lower spine. The risk for nonchronic lymphocytic leukemia was adequately described by a linear-exponential model that allowed for cell sterilization in heavily exposed parts of the marrow and time since exposure. Ten years after first exposure, the linear component of excess relative risk was 12.37 per Gy (95% confidence interval 2.25-52.07), and it was estimated that cell sterilization reduced the excess relative risk by 47% at 1 Gy (95% confidence interval 17%-79%). The average predicted relative risk in the period 1-25 years after exposure to a uniform dose of 1 Gy was 7.00.

Radon exposure and cancers other than lung cancer in Swedish iron miners

Data are presented on the risks of cancers other than lung cancer in a cohort of iron miners from northern Sweden occupationally exposed to elevated levels of the radioactive gas radon. Compared with rates for the four northernmost counties of Sweden, mortality was increased for all cancers other than lung cancer (ratio of observed to expected deaths 1.21, 95% confidence interval 1.03-1.41), stomach cancer (ratio of observed to expected deaths 1.45, 95% confidence interval 1.04-1.98), and rectal cancer (ratio of observed to expected deaths 1.94, 95% confidence interval 1.03-3.31). Despite these overall increases, mortality was not significantly associated with cumulative exposure to radon, either for all cancers other than lung cancer or for any site of cancer other than lung cancer individually. However, the data from this cohort on its own have limited power; and for several sites of cancer the data in this study would be consistent with a radon-related increase. Further study of cancers other than lung cancer in populations exposed to radon is required.

Recent results from the study of West Bohemian uranium miners exposed to radon and its progeny

A brief description is given of the study of West Bohemian uranium miners, and recent and ongoing efforts to improve the quality of the data are summarized. Three recent analyses of the data from the cohort have led to rather different estimates of the excess relative risk of mortality from lung cancer per working-level month. The reasons for these different estimates are described, and it is concluded that estimates of lung cancer risk are strongly influenced by the quality of the exposure estimates, especially by the omission of some exposures accumulated during employment at other uranium mines, following the closure of most of the shafts at the original two mines. The most recent analysis has shown that, in common with other cohorts of radon-exposed miners, the excess relative risk of lung cancer per working-level month is modified by age and time since exposure. An inverse effect of exposure rate was also demonstrated, but it affected only men at very high concentrations and appears to be related to the time pattern of exposure. In addition, the risk was found to differ between the two main mines, possibly due to the influence of arsenic in the dust of the mines.

Radon and cancers other than lung cancer in underground miners: a collaborative analysis of 11 studies

BACKGROUND

Exposure to the radioactive gas radon and its progeny (222Rn and its radioactive decay products) has recently been linked to a variety of cancers other than lung cancer in geographic correlation studies of domestic radon exposure and in individual cohorts of occupationally exposed miners.

PURPOSE

This study was designed to characterize further the risks for cancers other than lung cancer (i.e., non-lung cancers) from atmospheric radon.

METHODS

Mortality from non-lung cancer was examined in a collaborative analysis of data from 11 cohorts of underground miners in which radon-related excesses of lung cancer had been established. The study included 64,209 men who were employed in the mines for 6.4 years on average, received average cumulative exposures of 155 working-level months (WLM), and were followed for 16.9 years on average.

RESULTS

For all non-lung cancers combined, mortality was close to that expected from mortality rates in the areas surrounding the mines (ratio of observed to expected deaths [O/E] = 1.01; 95% confidence interval [CI] = 0.95-1.07, based on 1179 deaths), and mortality did not increase with increasing cumulative exposure. Among 28 individual cancer categories, statistically significant increases in mortality for cancers of the stomach (O/E = 1.33; 95% CI = 1.16-1.52) and liver (O/E = 1.73; 95% CI = 1.29-2.28) and statistically significant decreases for cancers of the tongue and mouth (O/E = 0.52; 95% CI = 0.26-0.93), pharynx (O/E = 0.35; 95% CI = 0.16-0.66), and colon (O/E = 0.77; 95% CI = 0.63-0.95) were observed. For leukemia, mortality was increased in the period less than 10 years since starting work (O/E = 1.93; 95% CI = 1.19-2.95) but not subsequently. For none of these diseases was mortality significantly related to cumulative exposure. Among the remaining individual categories of non-lung cancer, mortality was related to cumulative exposure only for cancer of the pancreas (excess relative risk per WLM = 0.07%; 95% CI = 0.01-0.12) and, in the period less than 10 years since the start of employment, for other and unspecified cancers (excess relative risk per WLM = 0.22%; 95% CI = 0.08-0.37).

CONCLUSIONS

The increases in mortality from stomach and liver cancers and leukemia are unlikely to have been caused by radon, since they are unrelated to cumulative exposure. The association between cumulative exposure and pancreatic cancer seems likely to be a chance finding, while the association between cumulative exposure and other and unspecified cancers was caused by deaths certified as due to carcinomatosis (widespread disseminated cancer throughout the body) that were likely to have been due to lung cancers. This study, therefore, provides considerable evidence that high concentrations of radon in air do not cause a material risk of mortality from cancers other than lung cancer.

IMPLICATIONS

Protection standards for radon should continue to be based on consideration of the lung cancer risk alone.

Cancer mortality following X-ray treatment for ankylosing spondylitis

Mortality has been studied in 15,577 ankylosing spondylitis (AS) patients diagnosed between 1935 and 1957 in the UK, of whom 14,556 received X-ray treatment. By January 1, 1992 over half of the cohort had died. Among the irradiated patients, cancer mortality was significantly greater than expected from the national rates for England and Wales, with a ratio of observed deaths to expected (relative risk, RR) of 1.30, and significant increases individually for leukaemia, non-Hodgkin's lymphoma, multiple myeloma and cancers of the oesophagus, colon, pancreas, lung, bones, connective and soft tissue, prostate, bladder and kidney. Among the unirradiated patients, cancer mortality was lower than expected from national rates (RR = 0.79). Among irradiated patients, the RRs for leukaemia, lung cancer, and all other neoplasms all decreased significantly with increasing time since first treatment following an initial increase. By 35 years after first treatment, the radiation-related excess for lung cancer had completely disappeared, while for other neoplasms the RR remained significantly raised, although at a lower level than in earlier periods. Most irradiated patients received several courses of treatment within a 5-year period. Based on a 1 in 15 random sample, the mean total body dose received in this period was 2.64 Gy, with the heaviest dose to the vertebrae. A linear dose-response model for all neoplasms except leukaemia gave an excess RR of 0.18 Gy-1 in the period 5-24.9 years after first treatment, which decreased significantly to 0.11 Gy-1 in the period more than 25 years after first treatment. There was no evidence that a linear-quadratic model fitted the data better than a linear model. There were significant dose-response relationships individually for cancers of the lung, oesophagus, colon, pancreas, prostate, bladder and kidney.

Paraplegia and squamous cell carcinoma of the bladder in young women: findings from a case-control study

A death certificate-based case-control study was conducted on 207 women aged 25-44 who died of bladder cancer in England and Wales in the period 1971-89 and 411 controls matched on sex, year of death and age at death. An odds ratio of 12.0 (95% CI 1.5-99.7) was found for women with a history of paraplegia. Four of the six paraplegic women were reported to have had squamous cell carcinoma of the bladder compared with only 19 of the 201 non-paraplegic women. These findings suggest that squamous cell carcinomas of the bladder, especially in paraplegics, may be the result of chronic urinary tract infection.

Mortality in uranium miners in west Bohemia: a long-term cohort study

A cohort of 4320 uranium miners in West Bohemia who started work at the mines during 1948 to 1959 and worked there for at least four years were followed up to the end of 1990 to determine cause specific mortality risks in relation to exposures in the mines. The miners had experienced high radon exposures, on average 219 working level months during their uranium mining careers, for which detailed measurements were available. They had also been exposed to high arsenic levels in one of the two major mines, and to dust. New follow up methods, not previously used for occupational cohorts in Czechoslovakia, were utilised. By the end of follow up 2415 (56%) of the cohort were known to have died. Overall mortality was significantly raised compared with that in the general population (relative risk (RR) = 1.56, 95% confidence interval (95% CI) 1.50-1.63), with significantly raised risks of lung cancer (RR = 5.08, 95% CI 4.71-5.47), accidents (RR = 1.59, 95% CI 1.34-1.87), homicide (RR = 5.57, 95% CI 2.66-10.21), mental disorders (RR = 5.18, 95% CI 2.83-8.70), cirrhosis (RR = 1.51, 95% CI 1.16-1.94), and non-rheumatic circulatory diseases (RR = 1.16, 95% CI 1.08-1.25). The relative risk of lung cancer was greatest four to 14 years after entry to the mines. Relative risks for homicide and accidents were raised up to 25 years from entry but not after this. Substantial significantly raised risks at 15 to 24 years after entry occurred for cirrhosis, non-rheumatic circulatory diseases,a nd pneumonia and other respiratory infections. Sizeable significantly raised risks at 25 and more years after entry, but not earlier, were present for mental disorders, tuberculosis, and non-malignant non-infectious respiratory conditions. No specific causes showed risks significantly related to age at entry to mining. Risk of lung cancer was significantly positively related to radon exposure, estimated arsenic exposure, and duration of work in the mines, but no other cause was significantly positively related to these variables. The raised risk of lung cancer in uranium miners, which is well established, is related aetiologically to radon exposure, and in the present cohort it may also in part have been due to exposure to arsenic. The raised risks of accidents, tuberculosis, and non-infectious respiratory diseases have also been seen in other uranium mining cohorts, and are likely to reflect the dangerous and dusty working conditions and the confined spaces in which work occurred. The cirrhosis and homicide deaths probably related to the lifestyle associated with mining. The raised risk of circulatory diseases does not seem to be related to radon or arsenic exposure; its causes are unclear. The use of multiple follow up methods was found to be mortality in the cohort.

Mortality in a cohort of women given X-ray therapy for metropathia haemorrhagica

Mortality to January 1, 1991, has been studied in 2,067 women in Scotland given X-ray therapy for metropathia haemorrhagica during the period 1940-1960. Average follow-up was 28 years. Overall, 1,313 deaths were observed compared with 1,297.01 expected from Scottish rates [standardized mortality ratio (SMR): 1.01]. Mortality was increased for cancers of heavily irradiated pelvic sites (SMR 5+ years after irradiation: 1.46) following mean doses to organs in the vicinity of the pelvis in the range 2.6-5.3 Gy. For these cancers the SMR was higher 30+ years after irradiation than at 5-29 years, indicating that the effects of exposure last for over 30 years, and in this period bladder cancer mortality was exceptionally high (SMR = 4.91). Mortality was also raised for leukaemia (SMR 2+ years after irradiation: 2.05), following a mean bone-marrow dose of 1.3 Gy, and for multiple myeloma (SMR 5+ years after irradiation: 2.59). For leukaemia the SMR was lower 30+ years after irradiation than at earlier periods, but remained greater than unity. For other cancers mortality was similar to Scottish rates, except for breast cancer for which mortality was low (SMR 5+ years after irradiation: 0.53), even in women aged over 50 at irradiation (SMR 5+ years after irradiation: 0.14). The deficit was principally due to a large deficit of breast cancer in women with ovarian doses of at least 5 Gy.

Paternal exposure not to blame

The "Gardner hypothesis" suggests that an excess of childhood leukaemia near a nuclear reprocessing plant is caused by paternal exposure to ionizing radiation. But the evidence shows that this explanation is wrong.

Patterns of lung cancer mortality among uranium miners in West Bohemia with varying rates of exposure to radon and its progeny

Lung cancer mortality in a cohort of 4320 miners first employed during 1948-1959 at the Jáchymov and Horní Slavkov uranium mines in West Bohemia and followed until 1 January 1991 has been studied to gain a greater understanding of the consequences of exposure to radon and its progeny. Among men whose exposure rates never exceeded 10 working levels, excess relative risks per unit exposure were greater in younger men, and exposures received in the periods 15-24, 25-34 and 35+ years previously were found to have 47, 24 and 0% of the effect of exposures 5-14 years previously. Within this low-exposure-rate group excess relative risk increased linearly with time-weighted cumulative exposure and did not depend on exposure rate or duration of exposure. For men who spent less than 20% of their employment at the Jáchymov mine the excess relative risk per working level month was 1.36% (95% confidence interval 0.52-3.54) in the baseline category (age group 55+ and exposure received 5-14 years previously). For men who spent more than 20% of their employment at Jáchymov, the corresponding excess relative risk per working level month was higher by a factor of 1.80 (95% confidence interval 1.27-2.97). The difference may be due to the fact that men who spent more than 20% of their employment at Jáchymov were exposed to the much higher levels of arsenic in the dust at the Jáchymov mine than at other mines. When men with exposure rates above 10 working levels were included in the analysis, patterns of risk were complex and depended on both exposure rate and duration of exposure in addition to the factors mentioned above. If these findings are confirmed elsewhere, calculation of risk estimates for extrapolation to modern occupational or environmental exposures should be based on miners with exposure rates below about 10 working levels. Further investigation is desirable of the influence of dusts containing arsenic on lung cancer risk in miners exposed to radon.

Further follow up of mortality and incidence of cancer in men from the United Kingdom who participated in the United Kingdom's atmospheric nuclear weapon tests and experimental programmes

OBJECTIVES

To study the long term effects of participation in the United Kingdom's atmospheric nuclear weapon tests and experimental programmes and to test hypotheses generated by an earlier report, including the possibility that participation in tests caused small hazards of leukaemia and multiple myeloma.

DESIGN

Follow up study of mortality and cancer incidence.

SUBJECTS

21,358 servicemen and civilians from the United Kingdom who participated in the tests and a control group of 22,333 non-participants.

MAIN OUTCOME MEASURES

Numbers of deaths; standardised mortality ratios; relative risks of mortality from all causes and 27 types of cancer.

RESULTS

During seven further years of follow up the numbers of deaths observed in participants were fewer than expected from national rates for all causes, all neoplasms, leukaemia, and multiple myeloma (standardised mortality ratios 0.86, 0.85, 0.57, and 0.46); death rates were lower than in controls (relative risks 0.99, 0.96, 0.57, and 0.57; 90% confidence intervals all included 1.00). In the period more than 10 years after the initial participation in tests the relative risk of death in participants compared with controls was near unity for all causes (relative risk 0.99 (0.95 to 1.04) and all neoplasms (0.95 (0.87 to 1.04)); it was raised for bladder cancer (2.69 (1.42 to 5.20)) and reduced for cancers of the mouth, tongue, and pharynx (0.45 (0.22 to 0.93)) and for lung cancer (0.85 (0.73 to 0.99)). For leukaemia mortality was equal to that expected from national rates but greater than in controls for both the whole follow up period (1.75 (1.01 to 3.06)) and the period 2-25 years after the tests (3.38 (1.45 to 8.25)).

CONCLUSION

Participation in nuclear weapon tests had no detectable effect on expectation of life or on subsequent risk of developing cancer or other fatal diseases. The excess of leukaemia in participants compared with controls seems to be principally due to a chance deficit in the controls, but the possibility that participation in the tests may have caused a small risk of leukaemia in the early years afterwards cannot be ruled out.

Radon exposure and cancers other than lung cancer among uranium miners in West Bohemia

Recent observations have suggested that radon in the ambient air may cause cancers at sites other than the lung, but the evidence is indirect. We have studied site-specific cancer mortality in 4320 uranium miners in West Bohemia who have been followed-up for an average of 25 years, and in whom a four-fold radon-related excess of lung cancer has already been established. For all cancers other than lung cancer the number of deaths observed was slightly greater than that expected from national rates, but the increase was not significant statistically (ratio of observed to expected deaths [O/E] = 1.11, 95% confidence interval [CI] = 0.98-1.24) and mortality did not increase with duration of employment underground or with cumulative exposure to radon. Non-lung cancer mortality was significantly raised among men who started mining work aged under 25 but the increase was not related to cumulative radon exposure. When twenty-eight individual sites and types of cancer were examined, significantly increased risks were found for cancers of the liver (O/E = 1.67) and gallbladder and extrahepatic bile ducts (O/E = 2.26). For liver cancer, mortality did not increase with duration of employment underground or with cumulative radon exposure. For cancer of the gallbladder and extrahepatic bile ducts, mortality did not increase with duration of employment, but increased with cumulative exposure to radon. Mortality from multiple myeloma, although not significantly increased overall (O/E = 1.08), increased with cumulative exposure to radon. Mortality from leukaemia was not increased overall (O/E = 0.91) and was not related to cumulative radon exposure, but did increase with increasing duration of employment in the mines. There is no evidence in these miners that a radon-rich atmosphere increases the risk of any cancer other than lung cancer. Possible exceptions are cancer of the gallbladder and extrahepatic bile ducts and multiple myeloma but further study is needed before it can be concluded that the associations found are causal.

Factors affecting indoor radon concentrations in the United Kingdom

Data collected in a nationwide study on natural radiation exposure in UK dwellings (Wrixon et al. 1988) were re-analyzed to investigate the effects of rock type and various building and lifestyle characteristics, taken into account simultaneously, on indoor radon concentrations. A multiplicative model which takes into consideration the outdoor radon concentration is used. Indoor radon concentrations were found to be influenced by type of rock underlying the dwelling, double glazing, house type, floor level of rooms in which measurements were taken, window opening habits in the main bedroom, building materials used in the construction of the walls, floor type, and draught proofing. However, these eight factors together account for only 22% of the variation between dwellings. Estimates of the size of the effect associated with each factor are given.

Trends in childhood leukaemia in the Nordic countries in relation to fallout from atmospheric nuclear weapons testing

OBJECTIVE

To obtain further information about the risks of childhood leukaemia after exposure to ionising radiation at low doses and low dose rates before or after birth or to the father's testes shortly before conception.

DESIGN

Observational study of trends in incidence of childhood leukaemia in relation to estimated radiation exposures due to fallout from atmospheric nuclear weapons testing during the 1950s and 1960s.

SETTING

Nordic countries.

SUBJECTS

Children aged under 15 years.

MAIN OUTCOME MEASURES

Incidence rates of leukaemia by age at diagnosis, sex, country, and calendar year of diagnosis or year of birth; exposure category; relation between leukaemia and exposure for children aged 0-14 and 0-4 separately.

RESULTS

During the high fallout period the average estimated dose equivalent to the fetal red bone marrow was around 140 mu Sv and the average annual testicular dose 140 mu Sv. There was little evidence of increased incidence of leukaemia among children born in these years. Doses to the red bone marrow of a child after birth were higher, and during the high exposure period children would have been subjected to an additional dose equivalent of around 1500 mu Sv, similar to doses received by children in several parts of central and eastern Europe owing to the Chernobyl accident and about 50% greater than the annual dose equivalent to the red bone marrow of a child from natural radiation. leukaemia incidence and red marrow dose was not related overall, but rates of leukaemia in the high exposure period were slightly higher than in the surrounding medium exposure period (relative risk for ages 0-14: 1.07, 95% confidence interval 1.00 to 1.14; for ages 0-4: 1.11, 1.00 to 1.24).

CONCLUSIONS

Current predicted risks of childhood leukaemia after exposure to radiation are not greatly underestimated for low dose rate exposures.

Completeness of follow up in a cohort study of mortality using the United Kingdom National Health Service Central Registers and records held by the Department of Social Security

STUDY OBJECTIVE

The aim was to evaluate and improve the completeness of follow up in a cohort study of mortality carried out using the UK National Health Service (NHS) Central Registers.

SETTING

This was a follow up study of UK servicemen who served abroad in the 1950s and 1960s, including those who participated in the UK atmospheric nuclear weapon tests and experimental programmes.

DESIGN

Details of men for whom follow up achieved using the NHS Central Registers was unsatisfactory were submitted to the Department of Social Security (DSS) for tracing, as were details of men born before 1916 and reported by the NHS Central Registers as currently registered with a general practitioner, and a 1% sample of remaining men born in or after 1916.

MEASUREMENTS AND MAIN RESULTS

The additional follow up increased the number of deaths fully identified in the cohort by 6.5%. Mortality among those untraced on the NHS Central Registers was substantially greater than in the cohort as a whole (10.2% v 6.9%). Among those reported by the NHS Central Registers as not currently registered with a general practitioner, 2.7% were found to have died, as were 1.1% of men born before 1916 and currently reported to be registered with a general practitioner. As expected there was clear evidence that information about emigrations supplied by both the NHS Central Registers and DSS is far from complete.

CONCLUSION

Standardised mortality ratios based on follow up via the NHS Central Registers alone are likely to be somewhat low, and this should be borne in mind when interpreting the data.

Mortality among United Kingdom servicemen who served abroad in the 1950s and 1960s

The Registrar General's decennial supplements on occupational mortality provide only limited information on mortality in the armed forces in the United Kingdom. Mortality has therefore been studied among a group of 30,619 United Kingdom servicemen who served abroad in tropical or desert areas in the 1950s and 1960s, and who remained in the services for a total of at least five years. Mortality from all causes of death, all neoplasms, and all other known non-violent causes was lower than that expected from rates for all men in England and Wales, whereas mortality from accidents and violence was raised. These differences remained after adjustment for social class, affected both officers and other ranks, and had not disappeared even after the men had been followed up for at least 20 years. When mortality from 20 specific cancers and 10 other disease groups was examined there were significant excesses for cancers of the oesophagus (standardised mortality ratio (SMR) = 146; p = 0.03) and prostate (SMR = 156; p = 0.03), and significant deficits for cancers of the lung (SMR = 73; p less than 0.001), stomach (SMR = 66; p = 0.002), bladder (SMR = 53; p = 0.02), other specified neoplasms (SMR = 48; p = 0.001), coronary heart disease (SMR = 76; p less than 0.001), bronchitis, emphysema, and chronic obstructive lung disease (SMR = 42; p less than 0.001), and for five further groups of diseases unrelated to smoking or alcohol. Examination of mortality in each of the three services separately identified two specific hazards in the Royal Navy; seven deaths from mesothelioma occurred compared with less than 2.06 expected (p less than 0.005), and there was also an excess of neoplasms and of other diseases associated with alcohol (SMRs of 181 and 229; p = 0.002 and less than 0.001). Mortality from smoking related diseases other than those associated with alcohol was low in all three services, particularly among officers.

Time from infection with HIV to onset of AIDS in patients with haemophilia in the UK

The two-stage parametric regression model of Brookmeyer and Goedert has been adapted and fitted to data on the development of AIDS in haemophiliacs in the UK who are seropositive for HIV. The risk of developing AIDS by a given time following seroconversion increases with increasing age at seroconversion. It is likely that the risk increases smoothly with age, although the data have been analysed in three age categories, and it is estimated that by seven years after seroconversion 6 per cent of patients aged under 25 at seroconversion, 20 per cent of those aged 25-44 and 34 per cent of those aged 45 or more have developed AIDS. For a given age at seroconversion the annual risk of developing AIDS increases with increasing time after seroconversion, and at seven years the annual risks of developing AIDS during the next year in the three age groups are estimated to be 2 per cent for those aged less than 25 at seroconversion, and 10 and 11 per cent respectively for those aged 25-44 and 45 or more.