Leukaemia and non-Hodgkin's lymphoma in children of male Sellafield radiation workers

Our objective was to investigate if there was (i) an excess risk of leukaemia/non-Hodgkin's lymphoma among children of male radiation workers at the Sellafield nuclear installation in Cumbria, northwest England; (ii) a dose-response relationship between fathers' preconceptional irradiation and their children's risk of leukaemia/non-Hodgkin's lymphoma; and (iii) whether any observed association could be explained by demographic factors. We performed a cohort study of live births, 1950-1991 in Cumbria, followed up to age 25 years or the end of 1991, comparing the risk of leukaemia/non-Hodgkin's lymphoma among all 9,859 children of male radiation workers to that among all 256,851 children of non-Sellafield fathers. Children of radiation workers had a higher risk of leukaemia/non-Hodgkin's lymphoma than other children [rate ratio (RR) = 1.9, 95% confidence interval (CI) 1.0-3.1, p = 0.05]. Adjustment for population mixing greatly reduced the excess risk in the village of Seascale, adjacent to Sellafield, but had little effect elsewhere. The risk increased significantly with father's total preconceptional external radiation dose (RR(100mSv) = 1.6, 95% CI 1.0-2.2, p = 0.05). This dose-response was not reduced by adjustment for population mixing. Although our 13 exposed cases included 10 considered previously (Gardner et al., BMJ 1990;300:423-34), we used a cohort rather than a case-control design, with wider temporal and geographic boundaries, and confirmed the statistical association between father's preconceptional irradiation and child's risk of leukaemia/non-Hodgkin's lymphoma that they reported. The possibility remains that paternal preconceptional irradiation may be a risk factor for leukaemia/non-Hodgkin's lymphoma, and this effect may not be confined to Seascale.     

Quantifying the effect of population mixing on childhood leukaemia risk: the Seascale cluster.

A statistical model was developed based on Poisson regression of incidence of childhood leukaemia and non-Hodgkin's lymphoma (NHL) in relation to population mixing among all 119 539 children born 1969-1989 to mothers living in Cumbria, north-west England, (excluding Seascale). This model was used to predict the number of cases in Seascale (the village adjacent to the Sellafield nuclear installation) children, born 1950-1989 and diagnosed before 1993. After allowing for age, the incidence of acute lymphoblastic leukaemia (ALL) and NHL was significantly higher among children born in areas with the highest levels of population mixing, relative risk (RR) = 11.7 (95% confidence interval (CI) 3.2-43) and was highest among children of incomers. The model predicted up to 3.0 (95% CI 1.3-6.0) cases of ALL/NHL in children born in Seascale compared to six observed and 2.0 (95% CI 1.0-3.4) cases in children resident, but not born, in Seascale compared to two observed. Population mixing is a significant risk factor for ALL/NHL, especially in young children, accounting for over 50% of cases in Cumbria and most cases in Seascale.     

Childhood leukaemia, nuclear sites, and population mixing

The excess of childhood leukaemia (CL) in Seascale, near the Sellafield nuclear reprocessing site in rural NW England, suggested that an epidemic of an underlying infection, to which CL is a rare response, is promoted by marked population mixing (PM) in rural areas, in which the prevalence of susceptibles is higher than average. This hypothesis has been confirmed by 12 studies in non-radiation situations. Of the five established CL excesses near nuclear sites, four are associated with significant PM; in the fifth, the Krummel power station in Germany, the subject has not been thoroughly investigated.     

Cancer mortality and morbidity among workers at the Sellafield plant of British Nuclear Fuels.

The mortality of all 14,282 workers employed at the Sellafield plant of British Nuclear Fuels between 1947 and 1975 was studied up to the end of 1988 and cancer incidence was examined from 1971 to 1986. This updates a previous report on mortality only up to the end of 1983. Ninety-nine per cent of the workers were traced satisfactorily. Cancer mortality was 4% less than that of England and Wales [standardised mortality ratio (SMR) = 96; 95% confidence interval (CI) = 90,103] and the same as that of Cumbria (SMR = 100: Cl = 94,107). Cancer incidence was 10% less than that of England and Wales [standardised registration ratio (SRR) = 90; Cl = 83.97] and 18% less than that of Northern Region (SRR = 82; Cl = 75.88). Cancer mortality rates were significantly in excess of national rates for cancers of the pleura (nine observed, 2.6 expected; P = 0.001), thyroid (six observed, 1.8 expected; P = 0.01) and ill defined and secondary sites (53 observed, 39.2 expected; P = 0.02). There were significant deficits of cancers of the liver and gall bladder, larynx and lung. Among radiation workers there were significant positive correlations between accumulated radiation dose and mortality from cancers of ill-defined and secondary sites (10 year lag: P = 0.01) and for leukaemia (2 year lag: P = 0.009), but not for cancers of the pleura and thyroid cancer. Previous findings of such associations with multiple myeloma and bladder cancer were less strong. There was a significant excess of incident cases of cancer of the oesophagus (P = 0.01), but this was not associated with accumulated radiation dose. For cancers other than leukaemia, the dose-response risk estimates were below those of the adult atomic bomb survivors, but the 90% confidence interval included risks of zero and of 2-3 times higher. For leukaemia (12 deaths, excluding CLL), under an excess relative risk model, the risk estimate derived for the Sellafield workers was about four times higher than that for the adult atomic bomb survivors with a confidence interval ranging from a half to nearly 20 times that of the atomic bomb survivors. Overall, however, there was no excess of leukaemia among the workers compared with national rates.     

A case-control study of childhood leukaemia and paternal occupational contact level in rural Sweden

In a national case-control study in Sweden, we investigated whether in rural areas (where susceptible individuals are more prevalent than in urban areas) leukaemia risk was higher among the young children of fathers with many work contacts, as the infective hypothesis has predicted. A total of 1935 cases diagnosed in 1958-1998 together with 7736 age-matched (within 1 year) population controls (of whom 970 and 3880 respectively were aged 0-4) were linked to paternal occupational details as recorded in the census closest to the year of birth. Applying the two classifications of occupational contact level used in a study of rural Scotland, the odds ratios for children aged 0-4 years in the highest contact category (which includes teachers) in the most rural Swedish counties were 3.47 (95% CI 1.54, 7.85) and 1.59 (1.07, 2.38) respectively, relative to the medium and low (reference) category; no such excess was found in urban or intermediate counties. There was also a significant positive trend at ages 0-4 in the rural counties across the three levels of increasing occupational contact (P for trend 0.02 and 0.03, respectively), but again not in the urban or intermediate counties. No such effect or trend was found at ages 5-14 in any of the three county groupings. The findings confirm those of a recent study in rural Scotland, and also suggest that unusual population mixing (as occurred in Scotland as a result of the North Sea oil industry) is not a necessary requirement for the effect, since comparable mixing has not been a feature of rural Sweden.     

Paternal preconceptional irradiation, population mixing and solid tumorsin the children of radiation workers (England)

Objective: To investigate whether there was an increased incidence of solid tumors among offspring of male radiation workers at the Sellafield nuclear installation in Cumbria, northwest England and whether paternal preconceptional irradiation was associated with the risk of solid tumors. Methods: A cohort study of 266,710 live births in Cumbria, 1950–1991, followed up to age 25 years on the end of 1991. Results: Children of radiation workers had a non-significantly increased risk of solid tumors (RR = 1.5, 95% CI: 0.9–2.4, p = 0.09), determined largely by an increased risk of cancers excluding leukemias, lymphomas, brain, spinal and gender-specific tumors (RR = 1.9, 95% CI: 1.0–3.3, p = 0.05), which was partly explained by differing patterns of parental migration (adjusted RR = 1.7, 95% CI: 0.8–3.2, p = 0.50). Within children of radiation workers there was no evidence of an increased risk with increasing paternal preconception dose of external radiation (hazard ratio per 100 mSv for all solid tumors = 0.6, 95% CI: 0.1–1.8, p = 0.52). Conclusions: Any observed excess of solid tumors in children of radiation workers may be partly explained by population mixing. Fathers' occupational exposure to radiation before conception was not found to be risk factor for solid tumors in their children.     

The influence of occupation on health--some results from a study in the UK oil industry

This study describes the patterns of mortality of 35,000 male employees, with a minimum of 1 year's continuous service, who worked at 8 refineries in the UK in the period 1.1.50-31.10.75. The trace rate of those involved was 99.8%. Overall the ratio of observed to expected deaths was 0.84 (O = 4406, E = 5259.9, P < 0.00001). The numbers of deaths for many of the chronic degenerative diseases were lower than 'expected'. The number of observed deaths from all cancers was appreciably less than expected (O = 1147, E = 1286.4, O/E = 0.89, P = 0.00006). Lung cancer was particularly reduced (O = 416, E = 532.7, O/E = 0.78, P < 0.00001); there was no excess of leukaemia (in workers including some exposed to benzene). Other comparable studies have suggested an excess of gastrointestinal cancer. In the present study deaths from oesophageal, stomach, intestinal and rectal cancer were slightly raised for all workers (O = 346, E = 328.6, O/E = 1.05, P < 0.4); this was particularly noticeable for those joining before 1950 with long service and with increased latent interval. There were also excesses based on small numbers of deaths from nasal cancer (O = 7, E = 3.1, O/E = 2.24, P < 0.05), and melanoma (O = 14, E = 6.5, O/E = 2.16, P 0.01).     

Incidence of childhood leukaemia in the vicinity of nuclear sites in France, 1990-1998

Overall, 670 cases (O) of childhood leukaemia were diagnosed within 20 km of the 29 French nuclear installations between 1990 and 1998 compared to an expected number (E) of 729.09 cases (O/E=0.92, 95% confidence interval (CI)=[0.85-0.99]). Each of the four areas defined around the sites showed non significant deficits of cases (0-5 km: O=65, O/E=0.87, CI=[0.67-1.10]; 5-10 km: O=165, O/E=0.95, CI=[0.81-1.10]; 10-15 km: O=220, O/E=0.88, CI=[0.77-1.00]; 15-20 km: O=220, O/E=0.96, CI=[0.84-1.10]). There was no evidence of a trend in standardised incidence ratio with distance from the sites for all children or for any of the three age groups studied. Similar results were obtained when the start-up year of the electricity-generating nuclear sites and their electric nuclear power were taken into account. No evidence was found of a generally increased risk of childhood leukaemia around the 29 French nuclear sites under study during 1990-1998.     

Population mixing and leukaemia in young people around the La Hague nuclear waste reprocessing plant

In order to investigate for an association between population mixing and the occurrence of leukaemia in young people (less than 25 years), a geographical study was conducted, for the years 1979 to 1998, in Nord Cotentin (France). This area experienced between the years 1978 and 1992 a major influx of workers for the construction of a nuclear power station and a new nuclear waste reprocessing unit. A population mixing index was defined on the basis of the number of workers born outside the French department of 'La Manche' and living in each 'commune', the basic geographical unit under study. The analyses were done with indirect standardisation and Poisson regression model allowing or not for extra-Poisson variation. Urban 'communes' were considered as the reference population. The Incidence Rate Ratio was 2.7 in rural 'communes' belonging to the highest tertile of population mixing (95% Bayesian credible interval, 95%BCI=1.2-5.9). A positive trend was observed among rural strata with increasing population mixing index (IRR for trend=1.4, 95%BCI=1.1-1.8). The risk became stronger for Acute Lymphoblastic Leukaemia in children 1-6 years old in the highest tertile of population mixing (IRR=5.5, 95%BCI=1.4-23.3). These findings provide further support for a possible infective basis of childhood leukaemia.     

Leukaemia and non-Hodgkin's lymphoma in young persons resident in small areas of West Cumbria in relation to paternal preconceptional irradiation.

The results of a previous study suggested that an association between childhood leukaemia and the radiation dose received occupationally by a father before the conception of his child might provide the explanation for the marked excess of childhood leukaemia and non-Hodgkin''s lymphoma in the village of Seascale, West Cumbria. The present study identifies other small areas (electoral wards) in West Cumbria where excess cases of leukaemia and non-Hodgkin''s lymphoma in young people have occurred and determines whether a recorded dose of radiation was received occupationally by the father before the conception of each of the affected individuals. Forty-one cases of leukaemia and non-Hodgkin''s lymphoma were diagnosed during 1968-85 in young people under 25 years of age resident in the 49 electoral wards lying within the boundary of West Cumbria and the adjacent ward of Broughton. Raised incidence rate ratios (two-sided P<0.01) were found for acute lymphoblastic leukaemia among those aged 0-14 years (concentrated among those aged 0-4 years) in Seascale ward and among those aged 0-24 years (also concentrated among those aged 0-4 years) in Egremont North ward, for acute myeloid leukaemia among those aged 0-14 years in Sandwith ward, for all leukaemias among those aged 0-14 years in Broughton ward (South Lakeland) and for non-Hodgkin''s lymphoma among those aged 0-14 years in Seascale ward. For West Cumbria as a whole, incidence rates were not usual. Apart from Seascale, for none of these electoral wards has a father of an affected child been linked definitely to an occupational dose of radiation recorded before the conception of the child. Particularly striking are the excesses of acute lymphoblastic leukaemia cases among young children living in the wards of Seascale and Egremont North, situated 11 km apart. The cases in Egremont North are not associated with recorded doses of radiation received occupationally by fathers before the conception of the affected children, even though the total numbers of children associated with such doses born in Seascale and Egremont North wards are similar. This finding is further evidence against a causal role for paternal preconceptional radiation exposure in the cases of childhood leukaemia in Seascale.     

Subsequent primary cancers in relation to treatment of ovarian cancer

The incidence of subsequent primary cancers was assessed in relation to treatment for a cohort of 7,203 patients from the Birmingham and West Midlands Cancer Registry diagnosed between 1957 and 1976. The total of 213 cancers observed one or more years after treatment for ovarian cancer (mean follow-up = 6.5 person-years) represented a significant excess (observed (O) = 213, expected (E) = 140.07, relative risk (RR) = 1.5, 95% CI 1.3-1.7, P less than 0.001). Among patients whose treatment included chemotherapy (CT), with or without radiotherapy (RT), the risk of acute and non-lymphocytic leukaemia (A + NLL) was significantly increased (O = 5, E = 0.18, RR = 27.8, 95% CI 9.0-64.8, P less than 0.001). The relative risks of A + NLL following RT without CT (RR = 4.5) and after other treatments (RR = 2.9) were not significantly in excess of 1.0. Significant excesses of subsequent cancers were observed at several sites: breast (RR = 1.7, 95% CI 1.3-2.2), lung (RR = 2.0, 95% CI 1.3-3.4), colon and rectum (RR = 1.6, 95% CI 1.1-2.3), urinary system (RR = 1.9, 95% CI 0.9-3.7), nervous system (RR = 3.3, 95% CI 1.2-7.3) and connective tissue (RR = 6.7, 95% CI 1.8-17.1) but the relationship with type of treatment was not so clearly defined as that for leukaemia. Although the treatment groups were broad and based on routinely collected data, they can enhance the use of cohort analyses for exploratory and monitoring purposes.     

Paternal occupational contact level and childhood leukaemia in rural Scotland: a case-control study

In a national Scottish study of 809 cases of leukaemia and non-Hodgkins lymphoma diagnosed in 1950-89 among children aged 0-4 years who were born in Scotland, together with 2363 matched population controls, we investigated one aspect of the infective hypothesis. This concerns whether in rural areas (where the prevalence of susceptible individuals is likely to be higher) the risk is greater among the young children of men whose work involves contacts with many different people, particularly children, as noted in certain childhood infections. A positive trend was found in rural areas across 3 levels of increasing paternal occupational contact (as recorded at birth) by each of 2 previously defined classifications; no such effect was found in urban areas. The rural trend was more marked in that part of the study period with greater population mixing, but the difference from the period with less mixing was not itself significant, leaving open whether these rural findings reflect the extreme isolation of much of rural Scotland, or the effects in such areas of a degree of population mixing. In marked contrast, among the 850 cases and 2492 controls aged 5-14, those in rural areas in the higher population mixing period showed a significantly decreasing trend with increasing paternal occupational contact level. This would be consistent with immunity produced either by earlier infection at ages 0-4 years, or directly by low doses of the infective agent that were largely immunizing at these older ages. The findings overall provide further support for infection underlying childhood leukaemia and for the role of adults.     

Clustering of childhood leukaemia in Hong Kong: association with the childhood peak and common acute lymphoblastic leukaemia and with population mixing.

Incidence data of childhood leukaemia (CL) in Hong Kong (1984-90) have been analysed for evidence of variation between small areas. All cases (n=261) were classified by morphological cell type, with the majority (n=205) being acute lymphoblastic leukaemia (ALL), and haematological review has permitted immunophenotypic classification for 73% of these. The data have been examined for evidence of spatial clustering within small census areas (TPUs) and for association with population mixing, with attention focused on those subgroups (especially the childhood peak of ALL--taken here to be diagnoses in children from 24 months up to the seventh birthday--and common ALL) which, it has been hypothesized, may be caused by unusual patterns of exposure and response to common infections. For the whole of Hong Kong, there was evidence of spatial clustering of ALL at ages 0-4 years (P = 0.09) and in the childhood peak (P<0.05). When these analyses were restricted to TPUs where extreme population mixing may have occurred, overall incidence was elevated and significant evidence of clustering was found for ALL (P<0.007) at these ages and for the common ALL in the childhood peak (P = 0.032). Replication of the analyses for subsets of leukaemia that were not dominated by the childhood peak of ALL found no evidence of clustering. This is the first investigation of an association between population mixing and childhood leukaemia in Asia and the first to include clustering and to consider particular subsets. The results are supportive of the ''infectious'' aetiology hypothesis for subsets of childhood leukaemia, specifically common ALL in the childhood peak.     

Influence of morphology on survival for non-Hodgkin lymphoma in Europe and the United States

We explored the influence of morphology on geographic differences in 5-year survival for non-Hodgkin lymphoma (NHL) diagnosed in 1990-1994 and followed for 5years: 16,955 cases from 27 EUROCARE-3 cancer registries, and 22,713 cases from 9 US SEER registries. Overall 5-year relative survival was 56.1% in EUROCARE west, 47.1% in EUROCARE east and 56.3% in SEER. Relative excess risk (RER) of death was 1.05 (95% confidence interval (CI) 1.01-1.10) in EUROCARE west, 1.52 (95% CI 1.44-1.60) in EUROCARE east (SEER reference). Excess risk of death was significantly above reference (diffuse B lymphoma) for Burkitt's and NOS lymphoma; not different for lymphoblastic and other T-cell; significantly below reference (in the order of decreasing relative excess risk) for NHL NOS, mantle cell/centrocytic, lymphoplasmacytic, follicular, small lymphocytic/chronic lymphocytic leukaemia, other specified NHL and cutaneous morphologies. Interpretation of marked variation in survival with morphology is complicated by classification inconsistencies. The completeness and standardisation of cancer registry morphology data needs to be improved.     

Occurrence of non-gastric cancer in the digestive tract after remote partial gastrectomy: analysis of an Amsterdam cohort

If peptic ulcer surgery favors the formation of carcinogenic N-nitroso compounds in the gastric remnant, an increased risk of cancer at sites in the gastrointestinal tract distant from the stomach might be predicted. To estimate the risk of carcinomas in the digestive tract, other than the stomach, occurring after partial gastrectomy, we analyzed an Amsterdam cohort of 2,633 post-gastrectomy patients operated on for benign disease between 1931 and 1960. Mortality in the study population was compared with the general Dutch population through person-year analysis. An excess mortality of biliary tract cancer (O/E:2.64; CL:1.32-4.72; p less than 0.01) and pancreatic cancer (O/E:1.65; CL:1.06-2.44; p less than 0.05) was found in males more than 5 years after surgery; females showed only an increased risk of pancreatic cancer in the first 5 years postoperatively (O/E:15.33; CL:1.85-55.43; p less than 0.01), probably due to misdiagnosis. All other non-gastric sites of the digestive tract carried no increased risk for cancer. In males, mortality due to colorectal cancer more than 5 years post-operatively was significantly decreased (O/E:0.58; CL 0.34-0.92; p less than 0.01). The excess mortality of biliary-tract and pancreatic cancer in males, which increases with the duration of post-operative interval, is consistent with a dose-response phenomenon. This study therefore supports the hypothesis that carcinogens are not only locally activated in the gastric remnant, but are hepatically excreted and initiate cancer in the biliary tree and pancreatic duct. Further exploration of this mechanism of carcinogenesis is warranted, since it may also explain the pathogenesis of pancreatic and biliary cancers in patients without gastrectomy.     

Second Primary Cancers Following Non-Hodgkin's Lymphoma in Japan: Increased Risk of Hepatocellular Carcinoma

We evaluated the risk of development of second primary cancers, with particular reference to subsequent hepatocellular carcinoma (HCC), in 592 patients diagnosed as non-Hodgkin's lymphoma (NHL), at Osaka Medical Center for Cancer and Cardiovascular Diseases. During 1978–1994, 2,163 person-years of observation were accrued, and 27 of the patients developed a second primary cancer, yielding an observed-to-expected ratio (O/E) of 1.53 [95% confidence interval (CI) = 1.01–2.23]. Significant excess risk was noted for primary liver cancer (PLC; O/E=4.36, 95% CI=1.99–8.28; O =9) and non-lymphocytic leukemia (O/E=26.17, 95% CI=5.26–76.46; O=3). The excess risk of PLC was relatively constant within the first 10 years after the NHL diagnosis. Patients who received chemotherapy as the NHL treatment had a significantly increased risk of PLC (O/E=5.91, 95% CI =2.70–11.23; O=9). Their clinical reports indicated that all nine patients with PLC were diagnosed as HCC, and eight of them had clinical and/or histologic evidence of cirrhosis at the time of HCC diagnosis. None of the nine patients had a history of blood transfusion between the first NHL treatment and the diagnosis of HCC. These findings suggested that Japanese NHL patients might have an increased risk of developing HCC, and they indicated the importance of medical surveillance for liver malignancies, as well as subsequent leukemias. Possible explanations for the excess risk of subsequent HCC are discussed.     

Occupational and non-occupational risk factors in relation to an excess of primary liver cancer observed among residents of Brooklyn, New York

The incidence and mortality rates of primary liver cancer (PLC) among residents of Brooklyn, New York, were studied for 1976 through 1983. Standardized race and sex-specific incidence and mortality rates and rate ratios were computed and compared with overall US rates as reported by the Surveillance, Epidemiology, and End Results (SEER) program. The results indicate a significant excess of PLC among all race and sex groupings. A comparison of the degree of agreement between incidence and mortality data for Brooklyn residents showed excellent agreement among male patients with PLC and adequate agreement among female patients with PLC. Next, the effect of occupation on PLC mortality among Brooklyn residents was assessed. Specific occupations found to be at excess risk for PLC are private household workers (ratio of observed to expected cases [O/E] = 4.34; P less than 0.0001), non-domestic cleaning and food and beverage service workers (O/E = 2.59; P less than 0.0001), protective service workers (O/E = 1.78; P = 0.035), and transport equipment operatives (O/E = 1.52; P = 0.027). Since the distribution of Brooklyn workers employed in these occupations was found to be similar to the distribution of all US workers employed in these occupations, it is unlikely that these occupational risk factors can explain the observed excess of PLC among Brooklyn residents. The effects of non-occupational risk factors for PLC were then assessed using multiple regression analysis. The only non-occupational risk factor found to be associated with PLC among Brooklyn residents was cirrhosis of the liver (P = 0.0072). It is interesting that of the four occupations found to be at excess risk for PLC in this study, three have been previously shown to be at excess risk for cirrhosis mortality. Moreover, Brooklyn residents have cirrhosis mortality rates that are approximately two times higher than US rates. These facts, coupled with the findings of this study, support the hypothesis that the excess of PLC observed among Brooklyn residents might be related to an excess of cirrhosis of the liver in the same population, and thus provide support for an etiologic role of cirrhosis in the pathogenesis of PLC.     

High bladder cancer mortality in rural New England (United States): An etiologic study

An interview study of next-of-kin of 325 persons who died of bladder cancer and 673 individuals who died of other causes in Vermont and New Hampshire (United States) was conducted to assess reasons for the persistent pattern of elevated bladder cancer mortality for both genders in rural New England. There was some evidence of elevated risks for both leather and textile workers that rose to over twofold for workers who also lived near these industries and for persons with French-Canadian ancestry. Occupational exposures in the textile and leather industry may explain at least a portion of the excess bladder cancer risk in rural New England.     

Population mixing, socioeconomic status and incidence of childhood acute lymphoblastic leukaemia in England and Wales: analysis by census ward

In this population-based study of acute lymphoblastic leukaemia (ALL) diagnosed among children aged under 15 years in England and Wales during 1986-1995, we analysed incidence at census ward level in relation to a range of variables from the 1991 census, which could be relevant to theories of infectious aetiology. 'Population-mixing' measures, used as surrogates for quantity and diversity of infections entering the community, were calculated from census data on the origins and destinations of migrants in the year before the census. Incidence at ages 1-4 years tended independently to be higher in rural wards, to increase with the diversity of origin wards from which in-migrants had moved during the year before the census, and to be lower in the most deprived areas as categorised by the Carstairs index. This last association was much weaker when urban/rural status and in-migrants' diversity were allowed for. There was no evidence of association with population mixing or deprivation for ALL diagnosed at ages 0 or 5-14 years. The apparent specificity to the young childhood age group suggests that these associations are particularly marked for precursor B-cell ALL, with the disease more likely to occur when delayed exposure to infection leads to increased immunological stress, as predicted by Greaves. The association with diversity of incomers, especially in rural areas, is also consistent with the higher incidence of leukaemia predicted by Kinlen, where population mixing results in below average herd immunity to an infectious agent.