The Timing of Geographic Power

• Activity Number: 183
• Type: Contributed
• Date/Time: Monday, August 4, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): David Wheeler*+ and Kate Calder
• Companies: Virginia Commonwealth University and Ohio State University
• Keywords: spatial epidemiology ; latency ; cancer ; population mobility

Abstract

There is a long history in spatial epidemiology of researchers looking for spatial signal in cancer incidence to identify risk factors that are distributed heterogeneously in the environment. However, many cancer cluster studies fail to find a significant spatial signal that leads to new etiologic findings. This is likely due in part to not looking in the right place and at the appropriate time. In most studies of spatial risk of cancer, investigators search for a spatial signal at only the time of diagnosis. However, environmental exposures pertinent to cancer risk have a particular temporality and many cancers have a latency of several decades. In addition, studies in the United States typically include a mobile population. We propose that when combined with the long latency of most cancers, the levels of population mobility in the US are sufficient to obscure the spatial signal related to pertinent, historic environmental exposures. We investigate this hypothesis in a series of simulation studies designed to characterize the effect of population mobility on the ability to detect a significant area of historic environmental exposure. We vary duration and intensity of exposure.