Keywords Agonist treatment, benzodiazepine absti- nence, benzodiazepine withdrawal, high-dose benzodiaz- epine dependence, maintenance treatment, substitution treatment

We are grateful for this opportunity to respond tothe commentaries on our paper [1]. In particular, weappreciate the interesting account of Peter Tyrer on thedevelopmentof therecommendationsmadebytheRoyalCollage of Psychiatrists for the use of benzodiazepines(BZD) [2]. It reminds us that the results of consensusmeetings and the derived clinical recommendations (orguidelines) need to be interpreted with some care. Hiscautionary comments on who to enrol in future ran-domized trials comparing substitution with withdrawalof BZD are important, in order to avoid premature re-strictionsof apossibletargetgroup.Of course,ourgroupagrees fully with his assessment that ‘we are now in aposition of equipoise whereby randomized controlledtrials of such procedures would be fully justified andboth the advantages and disadvantages of a substitutionpolicy exposed for all to view’.In the eyes of Dr Soyka, however, as there are noempirical data supporting our somewhat ‘astonishing’view on BZD substitution, our contribution is solelyan ‘academic’ discussion [3]. He does not consider thewealthof studieswhichfoundthatBZDwithdrawaltreat-mentisnotverysuccessfulinthelongterm,notonlybutevenmoresointhepatientgroup,forwhichweproposedsubstitution treatment (see references in [1]). Therefore,he thinks that the focus of research should rather beupon means to encourage treatment acceptance (absti-nence?) among the—probably undertreated—group ofpatients who have become dependent upon BZD andupon modulators of its receptors.While we do not contest the meaningfulness of thesepointsmadebyDrSoyka,westillthinkthatitiscriticaltoexplore alternatives to withdrawal treatment for thosewho have failed ‘here and now’. Further, we did notsimply propose to evaluate BZD with a long half-life,but ‘a slow-onset and long-acting benzodiazepine’.Whethersuchanendeavourshouldbecalledsubstitutionor maintenance is not important. What is important isthat our proposition might be a viable way for those whowere not able to profit from withdrawal, might attractmore patients and result in better outcomes than simplyfruitlessly repeating withdrawal treatment again andagain.Toqualifysuchaprocedureasaformof ‘therapeu-tic pessimism’ and ‘unconditional surrender’ by simplyreplacing one BZD by another is exactly what has beencarried out in Germany and elsewhere with regardto heroin dependence. In the former, until the early1990s, both the political and medical establishmentswere extraordinarily hostile towards any approach toaddiction that was not rigidly orientated towards absti-nence as a process as well as a goal, with well-knowndeleterious consequences [4].