Objective: This study aimed to evaluate the use of second-look resection (SLR) in stage T1 bladder cancer (BC) in a population-based Swedish cohort.Materials and methods: All patients diagnosed with stage T1 BC in 2008–2009 were identified in the Swedish National Registry for Urinary Bladder Cancer. Registry data on TNM stage, grade, primary treatment and pathological reports from the SLR performed within 8 weeks of the primary transurethral resection were validated against patient charts. The endpoint was cancer-specific survival (CSS).Results: In total, 903 patients with a mean age of 74 years (range 28–99 years) were included. SLR was performed in 501 patients (55%), who had the following stages at SLR: 172 (35%) T0, 83 (17%) Ta/Tis, 210 (43%) T1 and 26 (5%) T2–4. The use of SLR varied from 18% to 77% in the six healthcare regions. Multiple adjuvant intravesical instillations were given to 420 patients (47%). SLR was associated with intravesical instillations, age younger than 74 years, discussion at multidisciplinary tumour conference, G3 tumour and treatment at high-volume hospitals. Patients undergoing SLR had a lower risk of dying from BC (hazard ratio 0.62, 95% confidence interval 0.45–0.84, p < .0022). Five-year CSS rates were as follows, in patients with the indicated tumours at SLR (p = .001): 82% in those with T1, 90% in T0, 90% in Ta/Tis and 56% in T2–4.Conclusions: There are large geographical differences in the use of SLR in stage T1 BC in Sweden, which are presumably related to local treatment traditions. Patients treated with SLR have a high rate of residual tumour but lower age, which suggests that a selection bias affects CSS.