This was a cross-sectional study that enrolled 160 men aged 50 and above who were sexually active (sexual intercourse in the preceding 6 months) from a large primary care treatment centre. The subjects of interest were elderly aged 65 and above, and men aged 50-65 were used for comparison. The overall response rate was 66.9%. The men who participated were generally more affluent. Standard of living was measured by the presence of maid and housing type. Erectile function (EF) score was significantly higher in those who hired maids (P ¼ 0.02). Housing type was not associated with erectile dysfunction (ED). In Model A (included both monthly income and education), significant non-parametric correlations were found between monthly income and EF, intercourse satisfaction (IS), orgasmic function (OF) and sexual desire (SD) domains. After statistical adjustments, only EF (Po0.01) and IS (P ¼ 0.04) remained positively associated with monthly income. OF was negatively associated with age (Po0.01) and diabetes (P ¼ 0.04), whereas SD was negatively associated with age (Po0.01) in the multivariate analysis. Overall satisfaction was not significantly associated with any factor. In Model B (excluded monthly income from analysis), education attainment was positively associated with OF (P ¼ 0.04), but was not significant after adjustment for multiple testing. In the final multivariate model, only monthly income (Po0.01) and age (Po0.01), but not education (P ¼ 0.47), remained significantly associated with EF. This study suggests the influence of social determinants on EF and that this influence can extend into late life.