BACKGROUND: Dyslipidemia is a risk factor for cardiovascular disease, with elevated low-density lipoprotein cholesterol (LDL-C) and decreased high-density lipoprotein cholesterol (HDL-C) recognized as risk factors for acute coronary events. Studies suggest an association between low cholesterol levels and poor outcomes in acute sepsis. We sought to determine the relationship between baseline cholesterol levels and long-term rates of sepsis. METHODS: We used data from the Reasons for Geographic and Racial Differences in Stroke (REGARDS) cohort, a population-based cohort of 30,239 community-dwelling adults. The primary outcome was first sepsis event, defined as hospitalization for an infection with the presence of ≥2 systemic inflammatory response syndrome criteria (abnormal temperature, heart rate, respiratory rate, white blood cell count) during the first 28 hours of hospitalization. Cox models assessed the association between quartiles of HDL-C or LDL-C and first sepsis event, adjusted for participant demographics, health behaviors, chronic medical conditions, and biomarkers. RESULTS: We included 29,690 subjects with available baseline HDL-C and LDL-C. There were 3423 hospitalizations for serious infections, with 1845 total sepsis events among 1526 individuals. Serum HDL-C quartile was not associated with long-term rates of sepsis (hazard ratio (HR) (95% CI): Q1 (HDL-C 5-40 mg/dl), 1.08 (0.91-1.28); Q2 (HDL-C 41-49 mg/dl), 1.06 (0.90-1.26); Q3 (HDL-C 50-61 mg/dl), 1.04 (0.89-1.23); Q4, reference). However, compared with the highest quartile of LDL-C, low LDL-C was associated with higher rates of sepsis (Q1 (LDL-C 3-89 mg/dl), 1.30 (1.10-1.52); Q2 (LDL-C 90-111 mg/dl), 1.24 (1.06-1.47); Q3 (LDL-C 112-135 mg/dl), 1.07 (0.91-1.26); Q4, reference). CONCLUSION: Low LDL-C was associated with higher long-terms rates of community-acquired sepsis. HDL-C level was not associated with long-term sepsis rates.