Guideline 31: Report on a Recommended Referral Document

Report on a Recommended Referral Document
Section 2: Development of the referral document

2.1 Literature review

The SIGN working group undertook a detailed review of the literature relating to the referral document. A number of computerised bibliographic databases were searched, including MedLine, Embase, Applied Social Sciences, Current Contents, DHSS-DATA, Health Planning and Administration, and the Social Science Citation Index. Searches using the "MeSH" subject headings of 'Referral and Consultation' were conducted, with search strategies similar to the following:

FAMILY PRACTITIONER$ or FAMILY PRACTICE or GENERAL PRACTITIONER$ or GENERAL PRACTICE or GP
REFERRAL or LETTER$ or CORRESPONDENCE or COMMUNICATION$
1 and 2

Discussion with colleagues in Scotland and in other parts of the UK revealed a small number of additional references, all of which were collected and included in the review.

This search strategy revealed around one hundred articles relating to referral communication. All abstracts were scrutinised and around 60 papers were found to be of direct relevance and so appraised by members of the working group, using a structured framework for appraisal, data abstraction and subsequent synthesis of the information. (For further details, please contact the SIGN Secretariat.)

By the nature of the subject, no evidence exists from randomised controlled trials as to whether improvements in referral communication change patient outcomes. Therefore the majority of the evidence comes from non-experimental descriptive studies and consensus methods in which referral letters are judged by a panel of experts and scored accordingly.

2.2 National consensus conference

In order to compensate for the absence of high quality published evidence relating to the referral document, the SIGN working group decided to hold a national consensus conference to allow health care professionals from across Scotland to discuss the subject in detail. Delegates were asked to complete a questionnaire before and after attending the conference. The results provided the working group with further evidence of what should be contained within the referral letter and how letters may be used in the future.

2.3 Formulating recommendations

From the literature review and analysis of the results of the national consensus conference, the working group was able to derive recommendations on essential information for inclusion in referral letters. However, the working group did not consider it possible to specify the grade of recommendation for individual pieces of information, as the referral letter should be regarded as one entity, requiring all elements to be present, for effective information transfer and seamless patient care.

The recommendations are presented in section 4 as a suggested template for referral communication. Where a piece of information is described as "essential" it should appear in both elective (routine) and emergency referral letters. Other pieces of information should always appear in elective referral letters, but may be omitted in letters relating to emergency referrals.