Spatial epidemiology and pattern analysis of childhood cancers in Tehran, Iran

Abstract

Identification of cancer clusters may have an important value to the study of disease etiology in cancer surveillance. We aimed to determine the spatial pattern of childhood cancer cases (CCCs) from 2007 to 2009 in Tehran, Iran. Records of 176 childhood cancer counts (children younger than 15 years old) for 2007-2009 were obtained from Iran's Ministry of Health and Medical Education. Thereafter, they were successfully geo-coded within a geographic information system based on their residence phone number or postal code. We used two distinct techniques, namely average nearest neighbor index (ANNI) and Quadrat analyses, to measure the spatial pattern of CCCs in Tehran. The count of childhood cancers for 2007-2009 in Tehran was 117.3 per 1,000,000 children. The ANNI analysis suggested that there was a clustered pattern for the CCCs in 2007-2009. There was less than 1% likelihood that this pattern could be the result of random chance (nearest neighbor ratio = 0.73; Z-score = –6.8 standard deviations; P < 0.01). In the Quadrat analysis, the largest absolute difference between the observed and expected cumulative proportions in the frequency table was 0.2778 while the critical value of Kolmogorov-Smirnov test was 0.1649. Therefore, the Quadrat analysis confirmed that the CCCs had clustered pattern in 2007-2009 in Tehran. Both used methods suggested that childhood cancers in Tehran had clustered pattern in 2007 and 2009. We believe further research is needed to study the etiological factors, especially environmental factors, which made this cluster.