AIMS--To compare the quality of outpatient anticoagulant control before and after the transfer of dosing responsibility to designated trained pharmacists from rotating junior medical staff. METHODS--All International Normalised Ratio (INR) values for an eight month period either side of the staff changeover were assessed for precision of therapeutic control according to described standards. Allowing for patient associated effects, observed and expected frequencies of "successful" control for the two staff groups were compared under the hypothesis of no association. RESULTS--INR results (n = 2219) for 382 patients were analysed. For patients in stable therapeutic control, there was no significant difference in performance between the two staff groups. Patients with an INR result "out" of control limits were more likely to be returned "in" to control at their next visit by the pharmacists than by the doctors. CONCLUSIONS--The quality of anticoagulant control in outpatient clinics benefits from dedicated trained staff using standard protocols.