Background and objective <p>Nusinersen was the first disease-modifying treatment approved for spinal muscular atrophy (SMA). However, long-term results of broad unselected populations—particularly adolescents and adults—remain limited. We aimed to evaluate nusinersen long-term persistence and effectiveness in a population-based cohort.</p> Methods <p>We conducted a population-based, ambispective observational study of all SMA patients in the Valencian community (Spain) between 2017 and 2022, with follow-up until December 2025 or censoring (due to death, clinical trial, or treatment switch). Demographic, clinical, and motor outcomes using revised SMA Functional Composite Score (SMA-FCR) were collected. Patients were classified as responders or non-responders. Nusinersen discontinuation risks and motor trajectories were evaluated using Bayesian linear and mixed linear models.</p> Results <p>Of 72 patients included, 18 were &lt; 12 years old (all treated with nusinersen) and 54 were ≥ 12 years (28 treated; 26 untreated) at the baseline visit. After a median follow-up of 4.6 years until censoring, all children were found responders, compared with 68% of those ≥ 12 years. Discontinuation rates were 11% in children compared to 75% in the older cohort. In patients ≥ 12 years, reasons for discontinuation included: treatment burden (71%), and loss (53%) or lack of benefit (43%). Lower baseline SMA-FCR (expEstimate = 0.84 [0.718,0.93],prob:1) and older age (expEstimate = 1.028 [1.011,1.055],prob:1) independently predicted higher discontinuation risk. Sustained treatment was associated with SMA-FCR increase, while untreated and discontinued patients showed slight deterioration.</p> Discussion <p>Nusinersen persistence was high in children but declined significantly after age 12 due to treatment burden and limited efficacy although a 25% of adolescents and younger adults with higher baseline function experienced sustained benefit.</p>