Processingand assessing psychiatric referrals

from A was then designed to referral delays and their quality. This was to all GPs. were again audited over two three month periods. There was no significant difference between corresponding time delays in the three periods. Only one item, the presence or absence of past history, was significantly improved in the third period. These resultsand their implications are discussed.

The main components of medical audit are the audit of medical practice, outcome, the use of resources and the processing of referrals.
Medical practice is most commonly audited. Audit of outcome is difficult, time-consuming and controversial (Ellis & Sensky, 1991). The audit of the use of resources in psychiatry is complex, as hospital resources must be related to the community services and specific features of the neighbourhood (Royal College of Psychia trists, 1989). Auditing the processing of referrals is a simple task with obvious clinical relevance, but there is surprisingly little published work on this topic. We have audited the processing of referrals and the quality of these referrals from general practitioners to a sectorised mental health team in Nottingham. This was done before and after sending a referrers' guide which was designed to reduce referral delays and improve the quality of referrals.

The study
This study was undertaken in the north west sector of Nottingham, where a small mental health team served a population of 37,000. Re sources included seven in-patient beds, out patient clinics and a community mental health centre (CMHC). The consultant's secretarial sup port was hospital-based and the non medical members were supported by secretaries at the CMHC. As a matter of tradition referrals had been received at the hospital. New referrals were discussed at team meetings every Tuesday.
We studied all referrals from general practit ioners (GPs) during three periods, each of three months duration, commencing mid October 1990, April 1991 and April 1992.
A ten page referrers' guide was sent to the GPs in January 1991, the end of the first period. The guide described the service (team membership, aims and services offered) and referral pro cedures, underlined in the text. Referrers were clearly advised, by asterisks in the text, that postal delays could be avoided if referral letters were sent to the team base. They were asked to include information that was considered to be useful in assessing their referral (main symp toms, reason for referral, medication, past history).
In December 1991 a letter was sent to all the referrers asking them to include in their referral letters the services (e.g. GP counsellor, Relate etc) used by their patients prior to the referral. They were also informed again of the move of our team base.
The referrals were assessed and the following information was obtained: place where referral received, date typed on the referral letter, date when referral was discussed and date of first appointment. In the majority it was possible to obtain the date when the referral was discussed and date of first appointment. In the majority it was also possible to obtain the date when the referral was received. Enquiries were not made to ascertain the time delay between dictation, typ ing and posting from the GP. The following time delays were calculated: (a) date on referral letter -date when referral was received (Tl) (b) date on referral letter -discussion day (T2) (c) date of discussion -date of the first appointment (T3).
The quality of the referrals was assessed as described by Pullen & Yellowlees (1985).

Findings
The total number of referrals during the three periods were 27, 24 and 36 respectively; 75% of these referrals were from GPs. All, except two AUDIT referrals, both in the third period, were received at the hospital. It was possible to calculate the Tl delay in 53 referrals. T2 and T3 delays were obtained in all the referrals. There was no significant difference between the corresponding time delays in the three periods (Mann Whitney U test). Therefore, these results were collated and the means (Â±s.d.) for the three periods, Tl, T2 and T3 were calcu lated. These were 7.85 (Â±3.4)days, 11.2 (Â±4.6) days and 15.5 (Â±11.6)days respectively. Tls for the two referrals to the team base were much shorter, being two and three days respectively. The only significant finding in the quality of the referrals, in the corresponding three periods, was the mention of presence or absence of past his tory. It was mentioned in four (20%), nine (50%) and 15 (57%) of the referrals in the first, second and third periods respectively, being mentioned significantly more in the third period compared with the first (P<0.01, x2 6.624, d.f. 1). The results of the frequency of mention of the other items during the three periods were collated as there were no significant differences. The reason for the referral and the main symptoms were the most commonly mentioned items being present in 59 (92%) and 51 (80%) of the referrals. Medi cation was mentioned in 48 (75%) referrals. The presence or absence of family history was mentioned in only two (3%) referrals.

Comment
Early and appropriate interventions avoid crisis admissions. Reducing the 'postal delay' would lead to patients being seen promptly, enabling earlier appropriate interventions to be made. We were surprised and disappointed that there was no reduction in the delay periods (TI, T2) follow ing first the posting of the referrers' guide, and subsequently the letter. Equally, there was no major improvement in the quality of referrals over time. This confirms earlier findings (Branger et al, 1992) which highlight problems in com munication between hospitals and GPs.
Why was there no impact following posting of the referrers' guide? Perhaps the referrers did not perceive that sending referrals to the base would make a difference. These suggestions for change were part of an extensive referrers' guide. Per haps they did not consider the extensive infor mation to be relevant. This could lead to the referrers not reading the entire guide and not noticing these requests. But there was no im provement following the one page letter. Perhaps a telephone contact might have made a more significant impact.
The delay in receiving referrals was long when compared with the delay in discussion. This con tributed to an increase in time elapsing before appropriate interventions could be made. Expla nations for this delay include a delay in typing the referral, a postal delay, and an intrahospital transfer delay after arrival of the referral at a major teaching hospital. It is unlikely that the first two factors could universally account for all the delays. Although the number of referrals received at the health centre were few, they were all received within three days. This makes intrahospital transfer delay the most likely explanation.
The reason for the referral and main com plaints were the most commonly mentioned items. This is not surprising as one would expect most referrers to state the main reasons for referring. Medication and past history were men tioned in over 50% of the referrals. Family history was mentioned in only two referrals.
Comparing the results of the present study with those of Yellowlees & Pullen (1985), the only significant finding was that family history was mentioned significantly more in both their 1973 (P<0.001, x2 25.15, df 1), and 1983 (P<0.001, x2 15.244, d.f. 1) samples. Why should there be such a difference between the two studies? The reasons for this are not clearly evident. Perhaps, in our study, GPs did not consider family his tories to be relevant or they assumed that the psychiatrist was already aware of seriously affected family members. Many of the referrals were computerised, and we wondered whether the fixed format led to the omission.
The components of the audit cycle are setting standards, observing practice, comparing ob served practice with standards and implement ing change. National standards are not available with which we can compare our results. This is an important area requiring audit to set national standards. We have set up a database to monitor further the trends of our observed practice. Elec tronic data interchange has been shown to be a method of improving communication between the primary carer and the hospital (Branger et al, 1992). This could be an effective way of imple menting change as there is direct communi cation between the GP and the psychiatrist, thus avoiding typing and postal delays. We are setting up a fax with a view to observing its impact on our observed practice. Further audit is required in this neglected but important area to enable comparison of referral procedure delays with other teams.

Brian Dalai
Candidates taking multiple choice question (MCQ) ex aminations are often unsure of the best strategy to use when uncertain of the answer to a question. Some authorities advocate a cautious 'never guess' strategy and others suggest a bold 'guess everything' approach.
In this study, candidates who had taken a MCQ paper were asked to go back and guess the questions initially marked 'don't know'. The resultant 'guess scores' ranged from -0.6% to 6%. It was concluded that many candidates could substantially increase their scores by adopting a bolder approach. It is proposed that the guess score is a useful measure of the effectiveness of the candidate's MCQ technique.
Candidates taking multiple choice question (MCQ) examinations are often uncertain about how many questions they should answer in order to do full justice to the knowledge that they have without risking marks by reckless guessing. The Royal College of Psychiatrists' Handbook for Inceptors recommended a relatively cautious approach, advising candidates to attempt questions when they are reasonably sure of the answer and to mark the rest 'don't know '. Harden et al (1976) have advocated the bolder strategy of attempting all questions, reasoning that a ran dom guess has a 50% chance of being correct and so any attempt based on some knowledge should increase this probability and so improve the score. Holden (1987) has suggested that can didates might usefully obtain a specimen paper, answer alternate questions using the different recommended techniques (i.e. 'cautious' v. 'bold') and compare the marks obtained to see which technique is more effective.
Fleming (1988) advised a group of medical students that, although wild guessing (guessing on the basis of total ignorance) is as likely to lose marks as gain them, educated guesses (based on some knowledge of the subject) are more likely to be right than wrong. In a subsequent MCQ ex amination the students answered an increased proportion of the questions and this increase was most marked for those who had previously been most cautious. It was also shown that when this cautious group answered more boldly they increased their scores and their performance relative to the other students.

The study
A sample MCQ paper was sent to 23 trainees on the Nottingham SHO/registrar rotational train ing scheme in psychiatry who had completed at least one year and not more than three years of psychiatric training. The paper was obtained from a course for candidates taking the MRCPsych Part I examination and comprised 50 MCQs, each with five parts, to be answered in 90 minutes. Ten of the trainees were due to take the MRCPsych Part I examination the following month, six were due to take the MRCPsych Part II and seven were between exams. Immediately af ter completing the paper, each trainee was asked to go back and attempt all the questions that had initially been marked 'don't know', using a red pen. All the trainees were invited to take the test under examination conditions, but only six were able to do so. The other test papers were sent to trainees by post, along with instructions to take the paper under conditions that approximated as closely as possible to examination conditions. Replies were received from all ten of those due to take MRCPsych Part I, three of those about to take MRCPsych Part II and four of those between exams (overall response rate 17/23=74%). The papers were initially marked with a mark gained for each correct answer, a mark lost for each incorrect answer and no marks for 'don't know' responses to obtain the score. Next, the trainees' attempts to guess the questions initially marked