Background Context Chronic low back pain (cLBP) is the leading global cause of disability, yet the anatomical, psychological, and socioeconomic determinants of cLBP are typically studied in isolation, with limited large-scale integration of imaging data. Purpose To determine the relative contributions of anatomical, demographic, anthropometric, socioeconomic, and psychological factors to diagnosed cLBP and to spinal pathology burden. Study Design Retrospective cross-sectional analysis of a single-center imaging cohort. Patient Sample 1,480 adults undergoing lumbar spine MRI at a single academic center, March 2014-March 2024. Outcome Measures Primary: physician-diagnosed cLBP ([≥]2 low back pain diagnoses [≥]6 months apart). Secondary: anatomical pathology burden, a composite count of stenosis, degenerative disc disease, facet arthropathy, and spondylolisthesis on lumbar MRI reports. Methods Pathologies were extracted from radiology reports using a large language model (GPT-4), benchmarked against a regex (keyword search) pipeline. Multivariable logistic regression assessed cLBP associations across all five domains; leave-one-domain-out analysis quantified each domain's contribution ({Delta}AIC). Parallel linear regression modeled pathology burden. Results The LLM pipeline achieved F1=0.98 versus 0.92 for regex. cLBP was most strongly associated with depression (OR=2.69, p