The NHS plan refers temptingly to the concept of 'one-stop health and social care services'. Patients and users would no longer be pushed from organisational pillar to professional post. A call to one would be a call to all: seamless care at the point of delivery.

While the new care trusts look like being the government's ultimate weapon in this strategy, a number of existing initiatives already point the way.

At the centre of this redesign seems to be the co-location of primary healthcare and social care staff. The plan declares: 'In future, social services will be delivered in new settings such as GP surgeries, and social care staff will work alongside GPs and other primary and community health teams as part of a single local care network.'

There is a long history of social workers working alongside hospital doctors. The Seebohm report of 1970 recognised the place of social worker attachment, and recommended that the proposed social services departments collaborate with GPs to set up a range of joint projects.

1This message was reiterated some years later in a further report on social work support for the health services which called for social workers' accommodation to be included in new health centre designs.

2Yet co-location remains a minority arrangement dependent on the whims of both social services managers and GPs. Now that co-location is about to move into the mainstream, it is timely to reflect on what has been learned from existing arrangements.

Much of the evidence certainly suggests that co-location produces benefits. The most commonly cited advantage is the ease of communication between practice-based team members, which in turn is seen as developing effective teamwork, thereby ensuring better quality of care. Part of this enhanced quality is the social worker's accessibility to the practice's patients: things can be quickly sorted out via a single access point to a range of health and social care services. One study of five co-location projects in South Worcestershire, for example, found improved liaison, effective joint action, greater continuity of care, and social workers dealing with twice as many clients as their colleagues in area teams.

3But co-location is not without its difficulties. The most frequently mentioned is that there are too few social workers to give each practice one fulltime appointment. This leaves scope for part-time co-location, where a named social worker is shared between two or more general practices, but even this may be thwarted by lack of space in older, smaller practices.

GPs vary in enthusiasm for the idea anyway, and the temptation has been to concentrate social work support in well disposed practices, which tend to be larger and more progressive. One consequence has been the creation of a two-tier service, with some patients receiving a high level of dedicated social work support and others having to make do with a less accessible and more remote mainstream system.

Logistical difficulties can probably be overcome in time, but others are more deep-seated.

Social services may not wish to go down the co-location route: structuring social work teams into chronological or clientgroup specialisms does not fit well with general practices' desire for a generic social worker.

Departments may be unwilling to move from established geographical boundaries towards practice boundaries, especially where a practice has no easily discernible catchment area. And they may be committed to priorities other than primary care, which would be distorted by a practice-based workforce. As pointed out elsewhere, the NHS plan's emphasis on social services' role in unblocking hospital beds ignores their wider responsibilities to the whole community.

4Part of the difficulty here is that the very concept seems to be indissolubly linked to a practice-based arrangement: co-location equals a social work presence in a practice. In fact other arrangements are perfectly possible and merit equal consideration. Much of the daily professional interaction will be between social workers and community nurses, rather than with GPs, and some areas have already co-located a district nurse in a social work team with some success. A further option was proposed by the Health Advisory Service, which supported the use of community hospitals as a focal point where primary healthcare and social services staff could interact with specialist providers in the care of older people.

5Finally comes the well-rehearsed but woefully underresearched issue of professional culture. There is a long-standing assumption that it is possible to identify distinctive medical and social models of care which might act as cultural sieves to filter out the fine grain of interprofessional working. To the extent to which such 'invisible' barriers exist, they will be more difficult to dismantle than financial or organisational barriers. The NHS plan has made some bold commitments to financial and structural change, but when it comes to implementing 'one-stop health and social care' it is precisely these complex operational roles and relationships which will be critical. There will be more to effective co-location than merely putting bodies in the same building.

REFERENCES

1 The Seebohm Report. Committee on Local Authority and Allied Services, 1970. HMSO.

2 Department of Health and Social Security. Social Work Support for the Health Service. 1974.HMSO.

3 Cumella S, le Mesurier N, Tomlin H. Social Work in Practice.Worcester: the Martley Press, 1996.

4 Wistow G. The Modernised Personal Social Services: NHS handmaidens or partners in citizenship? ADSS/ Nuffield Institute for Health, 2000.

5 Health Advisory Service. Services for People Who Are Elderly. The Stationery Office, 1997.