Background: Lumbar fusion for degenerative indications is associated with a great degree of practice variation. We summarize the current evidence on the comparative safety and efficacy of lumbar fusion, decompression alone, or non-operative care for degenerative indications. Methods: Literature search of electronic bibliographic databases was conducted. Comparative studies reporting validated measures of safety or efficacy were included. Treatments effects were calculated through DerSimonian and Laird random effects models. Results: We retrieved 62 studies (17 randomized controlled, 15 prospective, 15 retrospective, and 15 registries), enrolling a total 302,347 adult patients. Disability, pain, and patient satisfaction following fusion, decompression alone, or non-operative care were dependent on surgical indications and study methodology. Relative to decompression alone, the risk of reoperation following fusion was increased for spinal stenosis (relative risk [RR] 1.17, 95% CI 1.06 to 1.30, p<0.004) and decreased for spondylolisthesis (RR 0.71, 95% CI 0.59 to 0.84, p<0.001). In all indications, complications were more frequent following fusion (RR 1.88, 95% CI 1.37 to 2.58, p<0.001). Mortality and treatment modality were not associated. Conclusions: Improvements were greatest in patients undergoing fusion for spondylolisthesis while complications limited the role of fusion for spinal stenosis. The relative safety and efficacy of fusion for chronic low back pain suggested careful patient selection is required.