Aims <p>Long-term functional outcomes of non-operative treatment for humeral surgical neck fractures are not well established. This study aimed to: (1) compare differences in functional outcome between the affected and contralateral shoulder in patients with non-operatively treated surgical neck fractures, (2) assess patient-reported outcome measures (PROMs), radiological outcomes, and complications, (3) identify risk factors associated with poor functional outcomes.</p> Methods <p>This retrospective cohort study included patients with isolated humeral surgical neck fractures, managed non-operatively. The primary outcome was shoulder pain, function and strength measured with the Constant Murley Score (CMS), two years or longer after trauma. CMS of affected and contralateral shoulder were compared using a Wilcoxon signed-rank test. MCID for CMS was ≥ 10. Patients completed multiple PROMs and underwent follow-up radiographs. Risk factors for poor outcomes were identified by comparing patients with satisfactory and non-satisfactory primary outcomes.</p> Results <p>This study included 68 patients, mean age 62.8 ± 12.4, 75% female. The median follow-up time was 5.3 years (IQR, 3.2–6.9). The median CMS for the affected shoulder was 78.0 (IQR, 72.1–85.0) and 83.0 (IQR, 79.0–85.0) for the contralateral shoulder. The median difference in CMS was 4.5 (IQR, 0.0–9.3; p &lt; 0.001) but clinically not relevant (MCID ≥ 10). Three radiological complications were observed. Initial medial fracture displacement at trauma was significantly associated with a poor outcome (p = 0.009).</p> Conclusion <p>There is no clinically relevant difference in functional outcome between the affected and contralateral shoulder in non-operatively managed patients with surgical neck fractures. Non-operative treatment yields acceptable functional results, good quality of life and few radiological complications. High medial displacement at trauma is a risk factor for poor functional outcomes.</p>