Although numerous studies have been published, controversy still exists regarding fusion and nonsurgical treatment for symptomatic degenerative lumbar spine conditions. Definite conclusions are difficult to draw because of differences in patient inclusion criteria, fusion technique, nonoperative treatment regimen, and clinical outcome measures used to determine success. The objective of this study was to evaluate lumbar fusion and nonsurgical interventions for various degenerative spine disorders using the Oswestry Disability Index (ODI) as a primary outcome measure in a systematic review. A secondary objective was to determine whether there is a difference in clinical outcomes based on the specific diagnosis. Systematic review. Patients with low back pain of at least 12 weeks duration and older than 18 years, with prospectively collected ODI scores and at least a 12-month follow-up. ODI and Short Form-36 (SF-36). A MEDLINE, HealthSTAR, CINAHL, and Cochrane database search was done using the search strategy recommended by the Cochrane Back Review Group. Proceedings from annual meetings of various spine societies and reference lists from review articles and retrieved articles were evaluated for possible inclusion. Criteria for inclusion were prospective randomized clinical trials in patients with low back pain of at least 12 weeks duration and older than 18 years; with prospectively collected ODI scores and at least a 12-month follow-up. The methodological quality of the studies was assessed using the van Tulder criteria. Data extracted from each study included demographics, study design, diagnosis, baseline and change in ODI, and baseline and change in SF-36 Physical Composite Score (PCS). The data were pooled and analyzed based on the primary reported inclusion diagnosis: degenerative disc disease (DDD), chronic low back pain (CLBP), and spondylolisthesis; and treatment: fusion (unspecified, posterior, anterior, combined) and nonsurgical. Twenty-five studies met the inclusion criteria. The distribution of sex and smokers was similar across diagnoses and treatments. Patients with spondylolisthesis were older than patients with DDD and CLBP. Patients with spondylolisthesis had the greatest ODI improvement followed by patients with DDD and CLBP. The three fusion types produced similar amounts of improvement in ODI. Nonsurgical patients did not improve as much but had a lower baseline ODI. Improvements in the SF-36 PCS were fairly consistent across diagnostic groups and treatment types. Substantial improvement can be expected in patients treated with fusion, regardless of technique, when an established indication such as spondylolisthesis or DDD exists. CLBP patients are less disabled and experience less improvement.