Liaison-consultation meetings in general practice An audiotape analysis

Aims and method Audio-recordings were made over a period of six months of liaison-consultation meetings between general practitioners and a community mental health team in the Scottish Borders to show general trends in length of discussion and information exchange. Results Meetings were predominantly supportive, with high levels of shared information, but little educational content. Some trends in discussion time are shown. Clinical implications Audio-recording could form the basis for reviewing the function of liaison-consultation meetings.


Liaison-consultation meetings in general practice
An audiotape analysis

Rebecca J. Tipper and Ian M Pullen
Aims and method Audio-recordings were made over a period of six months of liaison-consultation meetings between general practitioners and a community mental health team in the Scottish Borders to show general trends in length of discussion and information exchange.
Results Meetings were predominantly supportive, with high levels of shared information, but little educational content. Some trends in discussion time are shown. Clinical implications Audio-recording could form the basis for reviewing the function of liaison-consultation meetings.
While written communication between psychia trists and general practitioners (GPs) has been studied extensively (Williams & Wallace, 1974: Kessel, 1984Pullen & Yellowlees. 1985), little attention has been paid to verbal communica tion. This may reflect the lack of regular verbal communication between primary and secondary care and perhaps the relative ease with which letters may be studied.
The different models of verbal communication in psychiatry include formal lectures and seminars, telephone conversations, and face-toface contact (King & Pullen, 1994). The latter can be subdivided into that occurring in different settings: outreach clinics in general practices with informal, unplanned discussions; regular liaison-consultation meetings with individual practices; and educational meetings with GPs.
Dingleton Hospital serves the scattered rural population of the Scottish Borders (population 105000) and has well-developed community mental health teams (CMHTs). Regular liaisonconsultation meetings with GPs have taken place for three decades having been developed by Maxwell Jones (Millard. 1996) as an extension of the principles of open communication and social learning that were the cornerstone of the Dingleton therapeutic community. Similar meetings initiated in West London by Burns (a former registrar at Dingleton) were studied by Midgley et al (1996). They found that 90% of discussion time was devoted to clinical matters, with 54% related to patients with psychotic illnesses.
This paper presents the results of a study to replicate their findings. The content of liaisonconsultation meetings held with the four general practices covered by one CMHT operating in central Borders was analysed. In addition, the results have been used to facilitate a review, with each of the practices, of the role and functioning of these meetings.

The study
Over a period of six months, audio-recordings were made of regular liaison-consultation meet ings between the GPs and the CMHT. The team consisted of a consultant, senior house officer, two community psychiatric nurses (CPNs) and an occupational therapist. The four rural prac tices involved were non-fundholding, compu terised, small, group practices with a total population of 22500. During the study period one three-partner practice subdivided, record ings continuing only at the two-partner practice. Meetings took place at the practices, and occurred at approximately six-weekly intervals.
A computerised print-out of the CMHT's current caseload was taken to meetings to aid recall and provide structure. Meetings lasted from 30-90 minutes.
The analysis of the audiotapes recorded the following: the name of each patient discussed: duration of the discussion; information given by GP or member of the CMHT: joint decisionmaking; and any educational discussion. All ratings were carried out by the same investigator (RT). The gender, age, length of contact with the CMHT and diagnosis (ICD-9 until February 1996, ICD-10 from April 1996) were obtained for each patient from the computerised case register.

Findings
Each patient was discussed only once per meet ing. The percentage of the total time spent discussing patients within each category is shown in Fig. 1.
Not all the patients on the caseload of the CMHT from any one practice were necessarily discussed at every meeting. The mean time per patient discussed was 111 seconds (s.d.= 124). The mean time per patient for each diagnostic category is shown in Fig. 2. Standard deviations show substantial variation in each category. These data are skewed by a discussion, at one of the meetings, of a 17-year-old woman with a diagnosis of borderline personality disorder: she was discussed for 1065 seconds. The patient's behaviour had caused great concern and, while this discussion involved exchange of informa tion, no decisions were taken, and it appeared to be an airing of frustrations and mutual support in a difficult and intractable case.
GPs shared information during 67% of patient discussions, and the CMHT in 81%. There was a trend for mean length of discussion to reduce with increased length of contact with the CMHT, but this trend was not statistically significant (see Fig. 3).
Only 7% of discussions were specifically educational and, on all occasions except one, occurred only when the consultant was present. Education related primarily to use of medication (especially antidepressants in line with the joint Colleges' consensus statement), but also to use of different services and explanations of different  (1996), where psychotic patients occupied half the discussion time. It is possible the differ ences may reflect rural/urban population var iations. Almost all discussion time was devoted to clinical matters, with only occasional forays into talk about current issues in the hospital or general practices. As in Midgley et afs study, few service users were discussed who had not yet been seen.
The reduction in discussion time with in creased length of contact with the CMHT may be a cause for concern. While new patients might be expected to generate most discussion, people with long-term mental health problems require a structured and coordinated approach from the practice and the CMHT. types of therapy. Joint decisions were taken in only 8% of discussions. Examples of these were: clinical management plans: whether to follow-up patients who defaulted from contact: organisa tion of joint meetings: referral to other services: the recording of risk assessment: and complex medication decisions.

Comment
The general atmosphere of the meetings was informal and friendly, and it is inevitable that the data cannot demonstrate the mutual support that is one of the most important features.
Depression and dysthymia take the lion's share of discussion time (32%). closely followed by schizophrenia and bipolar affective disorder (27%), then neurotic disorders (21%). This

Conclusions
Audio-recording is a simple and effective way of measuring and assessing verbal communica tion. The liaison-consultation meetings were characterised by equal input from GPs and CMHT members, and discussions were more supportive than decision-making. The analysis of the tapes drew attention to the limited educational content of the sessions and the absence of non-medical members of the pri mary care team. Audio-recording could form the basis for reviewing the functioning of liaison-consultation meetings.