There is vast evidence for the superiority of agonist treatments (methadone, buprenorphine) over a withdrawal approach in opioid-dependent populations. Little research, however, has been conducted on the same approach for the treatment of high-dose benzodiazepine (BZD) dependence. Even large-scale reviews and meta-analyses discussing treatment strategies for benzodiazepine-dependent patients focus solely upon approaches that aim at achieving abstinence, namely on complete BZD withdrawal. While the types of interventions differ (e.g. gradual benzodiazepine taper with a long or a short half-life benzodiazepine, switching to non-benzodiazepine anxiolytics or prescribing adjunctive medications such as antidepressants or anticonvulsants on an in- or out-patient basis), the common aim of treatment still is total abstinence from benzodiazepines. However, the majority of patients suffering from high-dose BZD dependence do not succeed with long-term abstinence, irrespective of the procedure, and clinicians have been using BZD ‘substitution’ treatment in such cases for decades. Therefore, we suggest the evaluation of a substitution approach in this group, consisting of maintenance treatment with a slow-onset, long-acting BZD. Advantages of such a procedure may be improved health, less craving, fewer withdrawal complications, reduced anxiety, increased treatment retention, improvements in social functioning and less illegal activity. Cognitive impairments, the most problematic side effects of substitution treatment with benzodiazepines, could possibly be minimized by using an optimal agonist.