Abstract Background Reducing substance use in youth is a global health priority. We compared two cohorts from the same 12-week residential substance use disorder (SUD) facility over a 10 year period: Cohort A (2008–2009) and Cohort B (2018–2020). The essential components of the program remained the same with the primary treatment being dialectical behaviour therapy (DBT) plus residential milieu. Methods Young people in the current Cohort B (N = 100) versus historical Cohort A (N = 102) had a similar ratio of males (74 vs. 70%) but were slightly older (mean 20.6 vs. 19.5 years). Linear mixed models were used to model outcome measures (global psychiatric symptoms, substance use severity, and quality of life) longitudinally up to 12 months later. Results Baseline to end-of-treatment comparisons showed that the current Cohort B had overall higher levels of global psychiatric symptoms (d = 0.70), but both groups reduced psychiatric symptoms (Cohort A: d = 1.05; Cohort B: d = 0.61), and had comparable increases in confidence to resist substance use (d = 0.95). Longitudinal data from the current Cohort B showed significant decreases in substance use severity from baseline to 6-month follow-up (d = 1.83), which were sustained at 12-month follow-up (d = 0.94), and increases in quality of life from baseline to end-of-treatment (d = 0.83). Conclusions We demonstrate how DBT plus milieu residential care for young people continues to show positive effects in a 10-year comparison. However, youth seeking treatment today compared to 10 years ago evidenced higher acuity of psychiatric symptoms reinforcing the importance of continuous improvement of psychological treatments. Trial registration Australian New Zealand Clinical Trials Registry: trial IDACTRN12618000866202, retrospectively registered on 22/05/2018, .