ObjectivesTo systematically collect and analyse diverse definitions of 'evidence' in both health and social sciences, and help users to correctly use the term 'evidence' and rethink what is the definition of 'evidence' in scientific research.DesignScoping review.MethodsDefinitions of evidence in the health sciences and social sciences were included. We have excluded the definition of evidence applied in the legal field, abstracts without full text, documents not published in either Chinese or English and so on. We established a multidisciplinary working group and systematically searched five electronic databases including Medline, Web of Science, EBSCO, the Chinese Social Sciences Citation Index and the Chinese Science Citation Database from their inception to 26 February 2022. We also searched websites and reviewed the reference lists of the identified studies. Six reviewers working in pairs, independently, selected studies according to the inclusion and exclusion criteria, and extracted information. Any differences were discussed in pairs, and if there was disagreement, it was resolved via discussion or with the help of a third reviewer. Reviewers extracted document characteristics, the original content for the definitions of 'evidence', assessed definitions as either intensional or extensional, and any citations for the given definition.ResultsForty-nine documents were finally included after screening, and 68 definitions were obtained. After excluding duplicates, a total of 54 different definitions of 'evidence' were identified. There were 42 intensional definitions and 12 extensional definitions. The top three definiens were 'information', 'fact' and 'research/study'. The definition of 'evidence' differed between health and social sciences. The term 'research' appeared most frequently in the definitions.ConclusionsThe definition of 'evidence' has gradually attracted the attention of many scholars and decision-makers in health and social sciences. Nevertheless, there is no widely recognised and accepted definition in scientific research. Given the wide use of the term, we need to think about whether, or under what circumstances, a standardised, clear, meaningful and widely applicable definition of 'evidence' might be helpful.

Background Without adequate reporting of research, valuable time and resources are wasted. In the same vein, adequate reporting of practice guidelines to optimise patient care is equally important. Our study examines the quality of reporting of published WHO guidelines, over time, using the RIGHT (Reporting Items for Practice Guidelines in HealThcare) reporting checklist. Methods We examined English-language guidelines approved by the WHO Guidelines Review Committee from inception of the committee in 2007 until 31 December 2017. Pairs of independent, trained reviewers assessed the reporting quality of these guidelines. Descriptive data were summarised with frequencies and percentages. Results We included 182 eligible guidelines. Overall, 25 out of the 34 RIGHT items were reported in 75% or more of the WHO guidelines. The reporting rates improved over time. Further, 90% of the guidelines reported document type in the title. The identification of evidence, the rationale for recommendations and the review process were reported in more than 80% of guidelines. The certainty of the evidence using the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) system was assessed in 81% of the guidelines assessed. While 82% of guidelines reported funding sources, only 25% mentioned the role of funders. Conclusions WHO guidelines provide adequate reporting of many of the RIGHT items and reporting has improved over time. WHO guidelines compare favourably to guidelines produced by other organisations. However, reporting can be further improved in a number of areas.