Recent genetic epidemiology studies have been dominated by genome-wide association (GWA) studies using single nucleotide polymorphisms (SNPs). However, a form of structural genomic variation, termed copy number variation (CNV), is also widespread throughout the human genome, and can be highly polymorphic between individuals. Such variation has long been shown, through candidate gene studies using low-throughput molecular biology techniques, to have direct consequences on human health and variation. Many studies have now sought to extensively characterise this variation on a genome-wide scale and, increasingly, attempts are being made to identify associations between CNV and human disease. Although many of the study design issues that have been described for SNP GWA studies are also relevant for CNV GWA studies, CNV studies also present their own unique set of challenges and considerations. New microarray-based technologies are enabling more accurate mapping of CNVs, and CNV maps of the human genome are being regularly refined with increasing resolution. The study of CNV and its effects on human health and disease therefore present a dynamic and exciting challenge for researchers in the field of genetic epidemiology.