Aims: To identify variables associated with response to first appointments at a specialist alcohol treatment clinic, and to determine the effect on patients' attendance of a brief telephone prompt prior to the appointment. Design, participants and intervention: Multivariate analysis of patient characteristics from referral letters and aspects of the quality of the referral process was carried out. Consecutive non-repliers (n~100), non-attenders (n~100) and attenders (n~100) were studied. A further 100 patients were matched and allocated to two groups, one of which received a telephone prompt prior to their appointment. Clinic attendance was noted. Findings: Results showed that patients' older age, shorter travelling distance to the clinic, shorter waiting time and administrative delay, faster response by the patient to the appointment invitation, and morning appointments were associated with positive responses. Patients not receiving a telephone prompt were less likely to attend. Conclusions: Reductions in waiting time for first appointment, locally based services and prompting patients by telephone would appear to hold promise as strategies for reducing non-attendance in alcohol dependency. Planning would need to take account of whether services could adequately cope with additional demand. Further research to assess the influence of patients' readiness to change prior to appointments is warranted.