A retrospective review of a random sample of written complaints made by, or on behalf of, users of psychiatric services to determine: (a) the number and nature of written complaints against clinical aspects of services in a mental health trust over a 1-year period; and (b) what information complaints provide about deficiencies in the quality of care.

Out of 325 recorded complaints in 1997, 192 concerned clinical aspects of services; 89% of complainants complained once. There was a roughly equal split between complaints about technical v. interpersonal aspects of care. Complaints were far higher from in-patient than from out-patient settings. Evidence that the complaints related to psychotic symptoms was rare. All complaints were resolved locally, but 28 responses by the team were judged unsatisfactory. In 39 cases further action was taken as a result of the complaint, but no disciplinary action was taken against medical staff.

Poor communication is likely to be at the root of many complaints. Room for improvement was found with respect to responses to complaints.

A confidential questionnaire was mailed to all trainers and trainees on the UCL/North London rotation (n=127), asking about the content and ways of improving supervision.

Seventy-six per cent of trainees received regular, timetabled supervision. Ninety-four per cent of trainees felt it was a good idea, but identified improvements, including more planning, setting an agenda and flexibility. There were differences between reports from trainers and trainees regarding the content of supervision. Respondents' comments are included.

Although supervision is popular and widely practised, this survey suggests that it is still not universally practised despite College stipulations. The content of supervision could be broadened to include more non-clinical matters such as teaching and careers guidance.

The aim of this survey was to determine attitudes among consultants in different specialities towards the development of a reciprocal liaison service providing access for psychiatric patients to medical and surgical liaison services equal to the access of medical and surgical patients to psychiatric liaison services. All medical, surgical and psychiatric consultants in a district health service were surveyed, with a total response rate of 48%.

The mean number of medical and surgical patients requiring a psychiatric liaison service was 6%. The mean number of psychiatric patients requiring medical and surgical liaison services was 11%. Ratings overall for various components of the two types of liaison service were generally similar, with acute assessments and follow-up being given a high priority for both types.

As liaison services are developed, the notion of equity of access for all patients is paramount. Commissioning of such services should thereby specify the reciprocal nature of development. This survey shows that generally there is a positive attitude to the development of such a service.

This paper describes the validation of self-report of risk by patients with borderline personality disorder (BPD) as compared with the judgement of experienced psychotherapists in regular contact with them. The aim was to validate the Clinical Outcomes in Routine Evaluation System (CORE) self-report in order to be able to use it to monitor risk change for patients with BPD in psychotherapy and general psychiatric settings.

There was significant separation correlation between CORE risk sub-scales for self-harm, suicide and risk to others and therapists' estimation of significant risk v. no significant risk.

Using the cut-offs described, we suggest that the CORE questionnaire risk sub-scales can be used to assess significant risk for patients with BPD in psychotherapy, and in psychiatric and community health teams. The sub-scales should also prove valuable in allocating Care Programme Approach status.