The epidemiology of leukemia

Occupational and environmental exposures and myelodysplasia: a case-control study

A case-control study of newly diagnosed myelodysplastic syndrome patients investigated lifetime exposures through occupation, environment or hobby by questionnaire, structured and semi-structured interview. The exposure histories of 400 individually matched pairs were compared. Increased or possibly increased odds ratios were observed for radiation (2.05, 95% confidence interval 1.16-3.76), halogenated organics (1.57, 0.97-2.57), metals (1.40, 0.99-2.00), several specific radiation exposures and individual chemicals and for childlessness (1.46, 1.01-2.11). Since myelodysplasia generally carries a poor prognosis, whether or not individuals convert to leukaemia or to other cancer, these findings add to previous reports of exposures implicated in the aetiology of leukaemia and add to the case for minimizing exposures to radiation and halogenated organics.

Family pesticide use and childhood brain cancer

The relationship between family pesticide use and childhood brain cancer was examined in a case-control study. Telephone interviews were conducted from June 1989 through March 1990 with the natural mothers of 45 childhood brain cancer cases, 85 friend controls, and 108 cancer controls. In comparisons to friend controls, significant positive associations were observed for use of pesticides to control nuisance pests in the home, no-pest-strips in the home, pesticides to control termites, Kwell shampoo, flea collars on pets, diazinon in the garden or orchard, and herbicides to control weeds in the yard. In comparisons to cancer controls, significant positive associations were observed for use of pesticide bombs in the home, pesticides to control termites, flea collars on pets, insecticides in the garden or orchard, carbaryl in the garden or orchard, and herbicides to control weeds in the yard. In general, positive associations in comparisons to one control group were supported by elevated odds ratios in comparisons to the other control group. Several potentially important associations were identified in this study. However, small sample sizes, potential recall bias, multiple comparisons, and lack of detailed exposure verification require further research to confirm these findings.

[Interprovincial variations in mortality from cancer: factorial and cluster analyses]

Cancer mortality rates among males, for sixteen sites in the fifty Spanish provinces were analyzed using two data reduction techniques, factor and cluster analysis. Five independent underlying factors were identified by factor analysis which appear to influence cancer mortality patterns. Cluster analysis aggregated provinces into different groups using these factor scores as the clustering criteria. Each of the groups thus defined a distinctive profile of site-specific cancer mortality rates.

Cigarette smoking and leukemia

Cigarette smoking has not been consistently associated with the subsequent development of leukemia. However, many of the earlier epidemiologic studies of leukemia have not considered specific histologic subtypes separately. The association between cigarette smoking and adult acute nonlymphocytic leukemia was examined in a case-control study of 114 patients and 133 controls. Cigarette smoking was associated with a significantly increased risk of acute myelocytic leukemia (relative risk estimate = 1.78, 95% confidence interval = 1.01 to 3.15) along with a significant (P less than 0.001) dose-response based on the total number of years of cigarette smoking. Since this is a preliminary study, more analyses of other epidemiologic studies are needed before it can be concluded that there is a causal association.

Childhood leukaemia in the West Berkshire and Basingstoke and North Hampshire District Health Authorities in relation to nuclear establishments in the vicinity

During the years 1972-85, 89 children aged 0-14 were registered with leukaemia in the West Berkshire and Basingstoke and North Hampshire District Health Authorities. Two nuclear establishments are located within the health authorities, and a third is situated nearby. Fifty of the 143 electoral wards in the two district health authorities lie wholly within, or have at least half their area lying within, a circle of radius 10 km around the establishments. In those 50 electoral wards 41 children aged 0-14 were registered with leukaemia, 28.6 registrations being expected on the basis of leukaemia registration rates in England and Wales (incidence ratio = 1.4, p less than 0.05). This excess was confined to children aged 0-4, among whom there were 29 registrations of leukaemia, 14.4 being expected (incidence ratio = 2.0, p less than 0.001). In the remaining 93 electoral wards there was a small and non-significant increase in the number of registrations of leukaemia at age 0-14 (48 observed, 40.8 expected; incidence ratio = 1.2). There was no obvious trend in the incidence of childhood leukaemia over the 14 years and the overall occurrence of the malignancy in the 143 electoral wards was consistent with a random distribution. In the surrounding Oxford and Wessex Regional Health Authorities the number of registrations of leukaemia at age 0-14 was virtually identical with that expected on the basis of registration rates in England and Wales (362 observed, 372.5 expected; incidence ratio = 1.0). These data indicate that in the two district health authorities studied there was an excess incidence of childhood leukaemia during 1972-85 in the vicinity of the nuclear establishments. In the West Berkshire and Basingstoke and North Hampshire District Health Authorities an average of 60,000 children aged 0-14 lived within a 10 km radius of a nuclear establishment each year. The normal expectation of leukaemia in these children was two cases a year, whereas the recorded incidence was three cases per year, representing one extra case of leukaemia each year among these 60,000 children.

The epidemiology of selected chronic childhood health conditions

The chronic illnesses of childhood as a whole show similar trends in rates of incidence and prevalence; within this general class, subgroups of the illnesses have distinct epidemiological characteristics and different histories of response to improvements in medical technology. This article selects 11 disorders as representative of the large class of childhood chronic illnesses. It reviews trends in the contribution of these illnesses to the overall morbidity and mortality rates among children. The current survival patterns for children with these illnesses are detailed in light of the probable demands generated for health care providers by similar patterns of survival. It is likely that these children will place heavy demands on the health care system for the foreseeable future. Epidemiological trends and improvements in survival patterns associated with these illnesses raise perplexing questions of resource allocation. New knowledge is needed to address emerging problems.