Search strategy

Details of terms used to search MEDLINE are shown in Appendix 1. Appropriate adaptations of search syntax were made when searching other databases. Searches were made of Cochrane controlled trials register (CENTRAL), MEDLINE, EMBASE, CINAHL, PSYCHLIT, and ZETOC. Searches were from database inception to July 2004. Reference lists of articles were searched to identify additional relevant reports and key journals were hand-searched. No language restrictions were applied and translations made when necessary. There were no restrictions applied to the years of publication. Expert opinion was sought.

Any new articles identified after July 2004 and up to April 2005 through reading of major journals reporting risk-scoring methods were also included.

Articles were incorporated into a Reference Manager database.

Screening of abstracts and titles

Articles were rejected on initial screening of abstracts and titles if two reviewers (ADB, PB) determined that there was no content relating to the use of risk functions or scores in the primary prevention of cardiovascular diseases. If the title or abstract could not be rejected with certainty, the article was acquired for further evaluation.

Potentially relevant articles were acquired and independently read by two reviewers (ADB, PB). Decisions on inclusion or exclusion into any of the reviews were recorded on a data inclusion/exclusion form according to specific inclusion criteria as described in each review section. Disagreements were resolved by discussion and, if necessary, in consultation with members of the project advisory panel.

Data extraction

Details of data extraction are described separately in each review section. Two reviewers (ADB, PB or members of advisory panel) extracted and checked relevant data, which was then synthesised as appropriate. Authors of studies with insufficient information were contacted.