It is curious that Seymour and colleagues ^{Reference Seymour, Negi, Flemons, Impey, Thomas and Witrylak1} feel we confuse community clinics with what they refer to as community-oriented mental health services. We regard community clinics as an important part of community-oriented older people's mental health services.

It is good to look at a range of models for delivering services to older people and we should routinely ask patients and carers for their views on all the services they receive. ^{Reference Clark, Benbow, Moreland, Scott and Jolley2} Enthusiasm for early referral of patients with memory problems and transfer of ‘graduates’ to old age services have changed expectations, increased the number of people seen for initial assessment and posed quandaries regarding how to achieve responsible and responsive follow-up of patients with enduring or relapsing illnesses. At the same time there is increasing demand to work in teams and to liaise with a variety of colleagues, services and agencies. These are not alternatives; rather, old age psychiatrists have to find a balance in how they allocate their time, given competing priorities. Different individuals (and services) will reach different resolutions of these dilemmas.

Nevertheless, it is the case that many services experience high numbers of patients who do not attend appointments when out-patient clinics are hospital-based: transport problems and demands on time allocation from carers argue against their efficacy for older people. So does the absence of immediate appreciation of what life is actually like for the individual and family carers in the home. Much of the information gained on a home visit is non-verbal and lost when people are seen in a clinic.

Perhaps part of the problem here is that different people have different understandings of how teams operate and what the role of a doctor is in relation to other team members. The multiplicity of teams developing in some areas has advantages in making more services available to particular groups of patients and their families, but also introduces disadvantages, such as: access criteria may be rigid and problematic for patients who do not fit into neat categories; demands on the time of staff working with several teams may be considerable and may have impact on their commitment to other service areas; each team will need to devote time to liaise with others, which may detract from direct care.

We are in the process of preparing a proposal to review work patterns within old age psychiatry services following on from our earlier studies of workload and stress. ^{Reference Benbow and Jolley3} This should add further substance to the discussion.