Objective: Substance use treatment is often performed inside locked wards. We investigate the effects of adopting a policy of open-door treatment for a substance use treatment and dual diagnosis ward. Methods: This is a prospective open-label study investigating 3-month study periods before opening (P1), immediately after (P2), and 1 year after the first period (P3). Data on committed patients, coercion (seclusion, forced medication, absconding events with subsequent police search), violence, and substance use was collected daily. We applied generalised estimating equation models. Results: The mean daily number of patients with ongoing commitment changed from 2.64 (P1) to 2.12 (P2) to 0.96 (P3), corresponding to a reduction of relative risk (RR) for having an ongoing commitment by 20% in P2 (RR 0.80; 95% CI 0.66-0.98) and 67% in P3 (RR 0.33; 95% CI 0.25-0.42). The mean daily number of coercive events was 0.29, 0.13, and 0.05, corresponding to a risk for undergoing coercive measures reduced by 56% (RR 0.44; 95% CI 0.22-0.90) and 85% (RR 0.15; 95% CI 0.05-0.45). Substance use, violence or ward atmosphere did not differ significantly. Conclusions: Our results support findings from general psychiatric wards of reduced coercion after adopting a primarily open-door policy. However, coercive events were rare during all periods. The widespread practice of restricting the freedom of inpatients with substance use disorders by locking ward doors is highly questionable.