Abstract Objective To determine the inter-rater reliability of identifying differences and types of differences in lumbar degenerative findings comparing supine and upright MRI. Materials and methods Fifty-nine participants, low back pain patients (LBP) with or without leg pain and no-LBP individuals were consecutively enrolled to receive supine and upright MRI of the lumbar spine. Three raters independently evaluated the MRIs for degenerative spinal pathologies and compared for differences. Presence/absence of degenerative findings were recorded for all supine and upright images, and then differences from the supine to the upright positions were classified into no-change, appeared, disappeared, worsened, or improved at each individual disc level. Reliability and agreement were calculated using Gwet’s agreement coefficients (AC₁ or AC₂) and absolute agreement. Results Inter-rater reliability of evaluating differences in eight degenerative lumbar findings comparing the supine and upright MRI position, ranged from 0.929 to 0.996 according to Gwet’s agreement coefficients (AC₂). The total number of positive MRI findings in the supine position ranged from 270 to 453, with an average of 366 per rater. Observed differences from supine to upright MRI ranged from 18 to 80, with an average of 56 per rater. Conclusion Inter-rater reliability was found overall acceptable for classification of differences in eight types of degenerative pathology observed with supine and upright MRI of the lumbar spine. Results were primarily driven by high numbers and high reliability of rating negative findings, whereas agreement regarding positive findings and positive positional differences was lower.