The methodology used is included in "Development of Clinical Practice Guidelines in the National Health System. Methodological Manual"¹⁸.

• Creation of the development group of the guideline.

In addition to the scientific coordinator, a development group of nine members was formed: 3 psychiatrists, 2 psychologists, 1 mental health nurse, 1 expert in methodology and 2 specialists from the Axencia de Avaliación de Tecnoloxías Sanitarias (avalia-t). All members of the development group submitted a declaration of interests. The possible training needs of the guideline development group were covered by a specific programme prepared by avalia-t, which consisted of six workshops. The duties of the various members who participated on the guideline are shown in the following table:

Table 1: Duties of the different members who participated on the guideline

	Coordinator	Clinicians	Methodologists and specialists of avalia-t	Documentalist	External reviewers
Development of clinical questions	+ + +	+ + +	+ +	+	-
Bibliographical search	+	-	+	+ + +	-
Evaluation and synthesis	+ +	+	+ + +	-	-
Recommendations	+ +	+ + +	+ +	-	-
Drafting	+ +	+ +	+ + +	-	-
External review					+ + +

• Development of the scope and objectives of the guideline. They were determined in conjunction with the Mental Health Authority of the Autonomous Community of Galicia. The clinical questions of the guideline followed the PICO format (Patient-Intervention-Comparison-Outcome).

• Search of the scientific information. A first search of the literature was performed without time limit in order to locate all existing CPGs in the main bibliographical databases (see the corresponding appendix in the full version of the guide) and to make a first estimate of the volume of material to be reviewed. To complete the guideline search, a general Internet search was performed using several search engines, thereby reviewing the pages of different national and international organisations, scientific societies, etc., with quality parameters and indicators and that could be of interest.

The selection of the depression guides was made from all guidelines that were based on trials or on the consensus of experts and published after the year 2000. Neither those that covered depression in specific groups (post-partum, the elderly, etc.) nor adaptations of other, existing guidelines were included. The search resulted in locating 15 clinical practice guidelines, and except for one of them (for which the complete text was not available¹⁹), they were all appraised by four technicians independently using the AGREE²⁰ instrument (“Appraisal of Guidelines Research and Evaluation”) (Appendix 9). The overall appraisal of the guidelines was established by consensus among the evaluators, always following the recommendations of the AGREE instrument. The guideline that received the highest score was prepared by the National Institute for Clinical Excellence (NICE)²¹, assessed as “highly recommended”, given that all areas earned scores above 60%. The overall assessment of most of the remaining guidelines was “recommended with modifications”, with scores of between 30 and 60%.

The NICE guideline was used as the starting point for all the questions discussed in it. When a NICE question was responded with recommendation level ‘A’, that recommendation was immediately adopted. In all other cases, the search was updated or the question was prepared again. As a general rule, specific strategies were used.

• Selection, evaluation and synthesis of the scientific information.

After reading the summaries of the articles appearing in the search, those that met the inclusion and exclusion criteria specified in the corresponding appendix were selected. Subsequently, they were given a critical reading, and data were analysed and extracted. Both the critical reading and the data extraction were selected independently by two reviewers following the recommendations of SIGN²² (Scottish Intercollegiate Guidelines Network).

• Formulation and rating of the recommendations.

The recommendations were formulated based on the “formal assessment” or “reasoned opinion” of SIGN. The scientific information has been classified and the recommendations have been rated according to the SIGN system. Controversial recommendations or recommendations in the absence of evidence were resolved by the consensus of the guideline development group, using the “appropriate use method” or “RAND/UCLA method”²³.

• External review and final recommendations.

The expert collaborators who participated in reviewing the draft of the guideline were selected at the proposal of the corresponding scientific societies. Comprehensive information of the methodological process of the CPG is available at the web page of Guiasalud, as well as at the web page of avalia-t.

• Update of the Guideline.

The updating of this guide is scheduled every three years, unless the publication of relevant scientific knowledge dictates that it should be done before that time, basically in those aspects or specific questions where the recommendations may be substantially modified.