Objective: Self-compassion is a positive psychological construct characterized by extending compassion toward oneself, often during periods of suffering. Whereas self-compassion has been associated with psychological outcomes, research linking self-compassion to physical health is limited. We tested the hypothesis that greater self-compassion would be associated with less subclinical cardiovascular disease (CVD) as assessed by carotid intima-media thickness (IMT). We adjusted for demographics, CVD risk factors and additionally depressive symptoms in these associations. Method: Women (N = 195; M age = 59 years) without CVD were recruited. Women completed questionnaires (Neff Self-Compassion Scale, Center for Epidemiology Studies of Depression Scale); physical measures (body mass index BMI, blood pressure BP); phlebotomy (lipids, insulin resistance); and ultrasound assessments of the carotid artery (mean, maximal IMT). Cross-sectional associations between self-compassion and IMT were assessed in linear regression models covarying for age, race/ethnicity, education, and CVD risk factors (BMI, BP, insulin resistance, lipids, medications), and additionally depressive symptoms. Results: Higher self-compassion was associated with lower mean IMT B(SE)=−.03 (.01), p = .02, adjusting for demographic factors and CVD risk factors. Associations persisted adjusting for depressive symptoms. When considering self-compassion subscale bifactors, the positive self-compassion bifactor (self-compassion), but not negative self-compassion bifactor (self-coldness), was related to lower IMT. Conclusions: Self-compassion is associated with lower subclinical CVD. Associations were not explained by standard CVD risk factors nor by depressive symptoms. Future research should consider whether enhancing self-compassion improves women's vascular health.