It has been hypothesized that nighttime bedroom measurements of extremely low frequency electromagnetic fields (ELF EMF) may represent a more accurate reflection of exposure and have greater biologic relevance than previously used 24-/48-hour measurements. Accordingly, the authors extended a pooled analysis of case-control studies on ELF EMF exposure and risk of childhood leukemia to examine nighttime residential exposures. Data from four countries (Canada, Germany, the United Kingdom, and the United States) were included in the analysis, comprising 1,842 children diagnosed with leukemia and 3,099 controls (diagnosis dates ranged from 1988 to 1996). The odds ratios for nighttime ELF EMF exposure for categories of 0.1-<0.2 microT, 0.2-<0.4 microT, and >or=0.4 microT as compared with <0.1 microT were 1.11 (95% confidence interval (CI): 0.91, 1.36), 1.37 (95% CI: 0.99, 1.90), and 1.93 (95% CI: 1.11, 3.35), respectively. The fact that these estimates were similar to those derived using 24-/48-hour geometric mean values (odds ratios of 1.09, 1.20, and 1.98, respectively) indicates that the nighttime component cannot, on its own, account for the pattern observed. These results do not support the hypotheses that nighttime measures are more appropriate; hence, the observed association between ELF EMF and childhood leukemia still lacks a plausible explanation.