The Knox method, as well as other tests for space-time interaction, are biased when there are geographical population shifts, i.e., when there are different percent population growths in different regions. In this paper, the size of the population shift bias is investigated for the Knox test, and it is shown that it can be a considerable problem. A Monte Carlo method for constructing unbiased space-time interaction tests is then presented and illustrated on the Knox test as well as for a combined Knox test. Practical implications are discussed in terms of the interpretation of past results and the design of future studies.

BACKGROUND.Incidence patterns, trends, and spatial and/or temporal clustering of childhood brain tumors were analyzed in the population-based national cancer registry of Sweden. METHODS.Temporal trends were analyzed by a logistic regression procedure in which the average annual percentages of change in incidence rates and the corresponding 95% confidence intervals (CIs) were calculated. Spatial and/or temporal clustering were investigated by using a geographic information system and analyzed with a modified version of the Knox test and a spatial scan statistic. RESULTS.Primary brain tumors in 1223 children ages 0 -15 years were registered during 1973-1992. In 80% of cases, the tumor was classified as malignant. Conclusive histopathology was classified in 1142 cases. The age-adjusted incidence rate for all subtypes of brain tumors was 35.9 cases per million children, and for malignant brain tumors 28.6. A statistically significant increasing temporal trend was observed for the group of malignant brain tumors as a whole (P ϭ 0.0001) and the astrocytoma subgroup (P ϭ 0.0001). The annual average increases were 2.6% (95% CI ϭ 1.5-3.8) and 3.0%, respectively (95% CI ϭ 1.6 -4.4). The increase in astrocytoma cases was significantly larger for girls than for boys (P ϭ 0.021) and was most striking for girls ages 6 -15 years, with an annual average increase of 4.7%.Rates had not increased for the primitive neuroectodermal tumor (PNET)/medulloblastoma or ependymoma subgroups. The geographic distribution of astrocytoma cases was homogenous. No statistically significant space-time interaction or local clusters in space and/or time were found for astrocytomas only or when astrocytomas were grouped with PNETs/medulloblastomas and ependymomas. CONCLUSIONS.The results show statistically increased incidence rates of childhood astroglial tumors, predominantly for girls, in Sweden during the period 1973-1992, but no clustering in space or time.

Few consistent etiological factors have been identified for primary brain tumors. Inverse associations to asthma and low levels of varicella-zoster virus, immunoglobulin (Ig) levels in prevalent cases have indicted a role for the immune system in the development of glioma. Because samples from prevalent cases of glioma could be influenced by treatments such as steroids and chemotherapy, we investigated pre-diagnostic samples from three large Scandinavian cohorts. To test the hypothesis that immune response levels to these viruses are associated etiologically with glioma risk, we investigated pre-diagnostic immunoglobulin levels for cytomegalovirus (CMV), varicella-zoster virus (VZV), adenovirus (Ad), and Epstein-Barr virus (EBV) including the nuclear antigen (EBNA1) using plasma samples from 197 cases of adult glioma and 394 controls collected from population-based cohorts in Sweden and Denmark. Low VZV IgG levels were marginally significantly more common in glioma cases than the controls (odds ratio (OR) = 0.68, 95% CI 0.41–1.13) for the fourth compared with the first quartile (p = 0.06 for trend). These results were more prominent when analyzing cases with blood sampling at least 2 years before diagnosis (OR = 0.63, 95% CI 0.37–1.08) (p = 0.03). No association with glioma risk was observed for CMV, EBV, and adenovirus.

No abstract

There has been no significant increase in the incidence of acute childhood leukaemia in areas of Sweden contaminated after the Chernobyl reactor accident.