Due to the sex differences in bone loss rate in later life [21], we additionally investigated genotypic effects separately in men and women; we found borderline evidence for a difference in the effects of rs9594759 (RANKL) on standing

balance by sex, with the effects only observed in women, and opposing directions of effect for rs3815148 (COG5) on standing balance, though we found no evidence for other differences. Genetic variants are generally not associated with typical confounders in observational epidemiology and, being fixed from conception, may be informative about the direction of causality [60]. We found no evidence of association between rs1801725 (CASR) and measures of anthropometry, physical activity levels or other demographic indicators. Proteasome inhibitor Additionally, previous investigations of CASR polymorphisms have found evidence against associations with many traits including, vitamin D levels [61], osteoarthritis [19], osteoporosis [19] or hip BMD [19] and [20], as well as no [19] or only modest [20] associations with lumbar spine BMD; however, there is some evidence that their effects on BMD may be modified by birth-weight [62]. Although

a previous smaller study of 1252 females aged between 70 and 85 years found no associations between the SNP and either grip strength or timed up and go [17], our findings based on a larger number of individuals (n = 11,239) suggest that our observed association Metformin concentration between rs1801725 and grip strength may indicate a causal role of raised serum calcium levels on poorer grip strength. The

association observed with grip strength but not with the three other physical capability phenotypes may be indicative pheromone of greater power due to the larger number of participants with available data with this trait. The inconsistent findings for the direction of effects for the BMD-raising alleles of the two SNPs considered and the associations observed for rs9594759 (RANKL) suggest further investigations are warranted in order to provide additional evidence for or against the causal role of BMD on physical capability. Previous smaller studies of older females (n = 421 [63] and 331 [64]) found no association between measures of physical performance, including grip strength, and SNP rs2234693, a variant in low LD (r2 = 0.04) with rs2941740 (ESR1). Our investigation was limited by the fact that we did not validate the genotypic effects of the SNPs on serum calcium, BMD or osteoarthritis in these studies. However, all of the SNPs chosen were robustly associated with their respective measures from large GWAS of individuals of European ancestry. The use of younger populations may help to elucidate whether associations are present at earlier stages of the life course.