Chronic cumulative exposure – categorizing residence in geothermal area and cancer incidence

Background: Residents of geothermal areas have increased incidence of non-Hodgkin’s lymphoma, breast, prostate, kidney cancers, and basal cell carcinoma of the skin. The exposure concerned gas emissions from geothermal fields and from airing geothermal hot water used for space heating, washing, bathing and showering. The aim was to describe and study exposure categorized according to duration of residence in association with cancer incidence.

Methods: The census cohort was followed from 1981 to 2014. The outcome was obtained from nation-wide cancer, death, and out-migration registries by record linkage of personal identifier. The accumulation of residence experience was obtained by record linkage with national rosters through the study period. The study group was compared with reference population according to continuous and categorized exposure metrics, and in internal analysis using Cox regression modeling adjusting for age, gender, education, housing, reproductive factors and smoking habits with different induction periods.

Results: The HR for all cancers was 1.18 (95%CI 0.97-1.43) for 1-5 years residence, 1.22 (95%CI 1.04-1.43) for 10-15 years residence, and 1.29 (95%CI 1.11-1.49) for 20-24 years residence as compared with reference area. The HR for internal analysis with five years lag for all cancers was for residence of 1 year (reference), 5 years 1.30 (95%CI 0.93-1.80), 10 years 1.68 (95%CI 1.11-2.54), 15 years 2.28 (95%CI 1.38-3.75), 20 years 2.57 (95%CI 1.47-4.50), and 24 years 4.45 (95%CI 2.85-6.96).

Conclusion: Immortal person-time is excluded from the comparison of exposed and unexposed to avoid bias. Different lag and follow-up time is important for the internal analysis.