What is meta-analysis, and does the Lubin/Boice meta-analysis prove that residential radon levels cause lung cancer?
Meta-analysis is a statistical attempt to analyze the results of several different studies to assess the presence or absence of a trend or to summarize results. Lubin and Boice conclude that the results of their meta-analysis are consistent with the current miner-based estimates of lung cancer risk from radon which place the number of radon-related deaths at approximately 15,000 per year in the United States. Because meta-analysis has several inherent limitations (such as the inability to adequately explore the consistency of results within and between studies and to control for confounding factors) meta-analysis is NOT able to PROVE that residential radon causes lung cancer, but it does provide additional GOOD SUGGESTIVE EVIDENCE. It is one more link in the “chain of evidence” connecting residential radon exposure to increased lung cancer risk. Since the investigators performing a meta-analysis do not have access to the raw data on the individual study subjects, the analysis is based
