Background: Spinal muscular atrophy (SMA) is caused by abnormalities of the survival motor neuron (SMN) 1 gene, leading to deficiency in SMN protein and loss of spinal cord alpha motor neurons. Newer disease-modifying agents (DMA) targeting the involved genes, including nusinersen and gene replacement therapies, have improved gross motor and respiratory function, but their impact on scoliosis development has not been established. This study aimed to determine risk factors for scoliosis development in SMA, specifically genetic severity and DMA use. Methods: In this retrospective cohort study, children with SMA and minimum 2-year follow-up were included. The primary outcome was the prevalence of clinically relevant scoliosis. Secondary outcomes included SMA type, SMN2 copy number, Hammersmith Functional Motor Scale (HFMS), ambulatory status [functional mobility scale at 50m (FMS50)], DMA use, and hip displacement as risk factors. Univariate/multivariate logistic regression analyses were performed to identify dependent/independent risk factors. Results: One hundred sixty-five patients (51% female) with SMA types I-III met the inclusion criteria, with total follow-up of 9.8 years. The prevalence of scoliosis was 79%; age of onset 7.9 years. The major curve angle for the entire cohort at first assessment and final follow-up was 37 degrees (SD: 27 degrees) and 62 degrees (SD: 31 degrees) (P<0.0001), respectively. Significant risk factors for scoliosis by univariate analysis were SMA type (I/II, P=0.02), HFMS (>23, P<0.001), nonambulatory status (FMS50=1, P<0.0001), DMA treatment (P=0.02), and hip displacement (P<0.0001). Multivariate analysis revealed that HFMS >23 (P=0.02) and DMA (P=0.05) treatment were independent (protective) risk factors. Conclusions: The development of scoliosis in SMA is high, with risk factors associated with proxy measures of disease severity, including SMA type, nonambulatory status, hip displacement, and most notably, gross motor function (by HFMS). DMA use and HFMS >23 were associated with a decreased risk of scoliosis development. Identified risk factors can be used in the development of surveillance programs for early detection of scoliosis in SMA. Level of Evidence: Level III.