Osteoarthritis (OA) is a common, disabling disease of the elderly. It is uncommon in individuals under 45 years of age, and prevalence is higher in females than males. OA is characterised by focal cartilage loss accompanied by osteophytes, sclerosis and often some inflammatory response. A significant genetic component to OA has been demonstrated by twin-pair, sibling risk and segregation studies. Our group previously carried out a two-stage genome-wide linkage scan and identified six chromosomal regions as potentially harbouring OA susceptibility genes. The aim of this project was to examine three of these linkage regions, one on chromosome 4 and two on chromosome 16, in more detail in order to further localise and determine the identity of the OA susceptibility gene(s) within these regions. Initially the three regions were linkage mapped using a staged approach. Linkage mapping of the region on chromosome 4q identified a region showing evidence of linkage to female hip OA with a multipoint LOD (MLS) score of 3.1. For chromosome 16 linkage mapping identified two regions with MLS of 1.7 and 1.9 respectively. The first region was restricted to families with female siblings concordant for hip OA while the second is accounted for by families with affected female siblings. The ADAMTS3 gene was identified as a strong candidate gene within the linkage region on chromosome 4 as a result of its role in procollagen processing. SNPs within the gene were identified and typed in cohorts of cases and controls. There was no evidence for an association with any of these SNPs and OA disease. Within the first linkage region on chromosome 16 the IL4R gene was identified as a strong candidate gene as a result of its proposed role in the normal response of chondrocytes to mechanotransduction. Again SNPs within the gene were identified and typed. Four SNPs show evidence of association with OA and for three of these there is evidence for a relationship between genetic variation and evidence for linkage. It appears that there is a dosage effect with at least two copies of different variant alleles being necessary to increase OA disease susceptibility.