BACKGROUND CONTEXT: Low back pain and disability are major public health problems and may be related to paraspinal muscle abnormalities, such as a reduction in muscle size and muscle fat content. PURPOSE: The aim of this study was to examine the associations between paraspinal muscle size and fat content with lumbar spine symptoms and structure. STUDY DESIGN/SETTING: This was a community-based magnetic resonance imaging (MRI) cohort study. PATIENT SAMPLE: A total of 72 adults not selected on the basis of low back pain were included in the study. OUTCOME MEASURES: The outcomes measured were lumbar modic change and intervertebral disc height. Pain intensity and disability were measured from the Chronic Pain Grade Questionnaire at the time of MRI. METHODS: The cross-sectional area (CSA) and amount of fat in multifidus and erector spinae (high percentage defined by >50% of muscle) were measured, and their association with outcome was assessed. RESULTS: Muscle CSA was not associated with low back pain/disability or structure. High percentage of fat in multifidus was associated with an increased risk of high-intensity pain/disability (odds ratio [OR], 12.6; 95% confidence interval [CI], 2.0-78.3; p=.007) and modic change (OR, 4.3; 95% CI, 1.1-17.3; p=.04). High fat replacement of erector spinae was associated with reduced intervertebral disc height (β=-0.9 mm; 95% CI, -1.4 to -0.3; p=.002) and modic change (OR, 4.9; 95% CI, 1.1-21.9; p=.04). CONCLUSIONS: Paraspinal fat infiltration, but not muscle CSA, was associated with high-intensity pain/disability and structural abnormalities in the lumbar spine. Although cause and effect cannot be determined from this cross-sectional study, longitudinal data will help to determine whether disabling low back pain and structural abnormalities of the spine are a cause or result of fat replacement of paraspinal muscles.