Objective Higher lipoprotein(a) [Lp(a)] has been linked with peripheral arterial disease (PAD). Also, elevated Lp(a) serum levels have been observed in women and African Americans (AAs). It remains uncertain if sex and ethnicity modify the association between Lp(a) and PAD. Methods Lp(a) mass concentration was measured with a latex-enhanced turbidimetric immunoassay, from blood collected at baseline clinic visits after a 12-hour fast, in a multiethnic cohort. Also at baseline, the ankle-brachial index was measured. PAD was defined as an ankle-brachial index <1.0. Multivariable logistic regression was used to determine sex and ethnic differences in associations of log-transformed Lp(a) and the presence of PAD. Results In 4618 participants, the mean age was 62 ± 10 years; Lp(a) mean was 30 ± 32 mg/dL and median (interquartile range) was 18 (8-40 mg/dL); 48% were male; 36% were European American, 29% were AA, 23% were Hispanic American (HA), and 12% were Chinese American; and 11% had PAD. Across all ethnic groups, serum Lp(a) was higher among women compared with men and highest among AAs compared with other ethnicities. After adjustments for traditional cardiovascular disease risk factors (age, sex, ethnicity, hypertension, diabetes, smoking, total cholesterol, and high-density lipoprotein cholesterol) as well as interleukin-6, fibrinogen, D-dimer, and homocysteine levels, one log unit increase in Lp(a) was associated with greater odds for PAD (odds ratio [OR], 1.12; 95% confidence interval [CI], 1.01-1.25). In fully adjusted models, significant gender∗ln[Lp(a)] and ethnicity∗ln[Lp(a)] interactions were observed (P =.08 for both). The association between higher Lp(a) and PAD was strongest in HA men (OR, 1.73; 95% CI, 1.07-2.80) and HA women (OR, 1.49; 95% CI, 1.07-2.08). Nonsignificant associations were observed for European American, AA, and Chinese American men and women. Conclusions We observed a significant and independent association between elevated Lp(a) and PAD only among HA women and men, despite higher serum Lp(a) levels among AAs. Future studies are needed to determine the role that lowering of Lp(a) may have on the burden of PAD in HAs.