• BARRETT'S OESOPHAGUS
• COELIAC DISEASE
• CLINICAL TRIALS

We refer to two recently published guidelines in this journal from the British Society of Gastroenterology (BSG) on coeliac disease1 and Barretts oesophagus.2 We describe our BSG policy on the production of guidelines, which sets a sufficiently high quality benchmark such that we have been awarded National Institute of Health and Care Excellence (NICE) accreditation for our production process.3 The full policy and advice document are available on our website.4

Most guidelines are commissioned by the BSG clinical services and standards committee (CSSC). The formulation of guidelines are driven by a Guideline Development Group (GDG), include a writing committee, members of which are recognised authorities in the field, and others representing a range of relevant expertise, as well as patient representatives and those whose everyday practice will be influenced by the guidelines. The GDG chairman is usually chosen by the relevant specialty section, in consultation with the CSSC.

All members of the GDG must complete a declaration of conflict of interests (COI) form. Whenever members have a potential COI with a particular section of the guideline, they may continue to be involved in the overall process but must withdraw their involvement from that area. In some cases, a COI may preclude membership of the GDG. Decisions with regard to these issues are made by the Chair of the GDG, in consultation with the head of the relevant BSG section and with the CSSC executive. All such decisions are documented and available for external review.

Guidelines are developed in accordance with the principles laid down by the AGREE II instrument.5 An initial proposal from the GDG chairperson is submitted to the secretary of the CSSC and the relevant specialist section. This is also shared with relevant stakeholders including professional groups, patients and carers and their views sought. The final proposal should include the overall objective of the guideline, the target population of patients, who the target users are and the main clinical questions to be addressed.

The clinical questions are usually grouped into sections. The process used is usually the patients, interventions, controls and outcomes system, in which these four critical components are predefined as precisely as possible.

The GDG develops a systematic, comprehensive, transparent and reproducible strategy to search for evidence on which management recommendations are based. The overall strategy is decided by the GDG as a whole and must be described in adequate detail (if necessary in an appendix). Libraries generated by keyword searches are stored to allow excluded references to be traced. For assessing the quality of evidence, the GRADE system6 is recommended. The cost and service implications of implementing the guideline and the potential facilitators and organisational barriers to doing so are considered. Guidelines normally include a statement as to how the implementation of the guideline will be assessed.

On completion, the guideline is reviewed by two BSG council members, two members of the CSSC and sent to all CSSC members for comments and posted on the BSG website for comments from the members. All feedback is sent to the lead author for amendments to the guideline and if these are satisfactory, the guideline is then submitted to Gut for the usual peer review process by four to eight internationally renowned authorities.

We believe that the BSG has set high quality standards for guideline development, which we encourage other organisations to follow.