STUDY DESIGN: Inter- and intra-rater variability study. OBJECTIVE: On the basis of a Scoliosis Research Society effort, this study seeks to determine whether the new adult spinal deformity (ASD) classification system is clear and reliable. SUMMARY OF BACKGROUND DATA: A classification of adult ASD can serve several purposes, including consistent characterization of a clinical entity, a basis for comparing different treatments, and recommended treatments. Although pediatric scoliosis classifications are well established, an ASD classification is still being developed. A previous classification developed by Schwab et al has met with clinical relevance but did not include pelvic parameters, which have shown substantial correlation with health-related quality of life measures in recent studies. METHODS: Initiated by the Scoliosis Research Society Adult Deformity Committee, this study revised a previously published classification to include pelvic parameters. Modifier cutoffs were determined using health-related quality of life analysis from a multicenter database of adult deformity patients. Nine readers graded 21 premarked cases twice each, approximately 1 week apart. Inter- and intra-rater variability and agreement were determined for curve type and each modifier separately. Fleiss' kappa was used for reliability measures, with values of 0.00 to 0.20 considered slight, 0.21 to 0.40 fair, 0.41 to 0.60 moderate, 0.61 to 0.80 substantial, and 0.81 to 1.00 almost perfect agreement. RESULTS: Inter-rater kappa for curve type was 0.80 and 0.87 for the 2 readings, respectively, with modifier kappas of 0.75 and 0.86, 0.97 and 0.98, and 0.96 and 0.96 for pelvic incidence minus lumbar lordosis (PI-LL), pelvic tilt (PT), and sagittal vertical axis (SVA), respectively. By the second reading, curve type was identified by all readers consistently in 66.7%, PI-LL in 71.4%, PT in 95.2%, and SVA in 90.5% of cases. Intra-rater kappa averaged 0.94 for curve type, 0.88 for PI-LL, 0.97 for PT, and 0.97 for SVA across all readers. CONCLUSION: Data from this study show that there is excellent inter- and intra-rater reliability and inter-rater agreement for curve type and each modifier. The high degree of reliability demonstrates that applying the classification system is easy and consistent.