Objectives: To determine the most important knowledge sources that can influence clinical practice and to cluster them in conceptual groups based on their relative importance.
Methods: Faculty members, fellows and residents of a large teaching tertiary care hospital were asked to rate the importance of different resources in their daily clinical practice and their understanding of some common terms from evidence-based medicine. The knowledge sources were distributed in a two-dimensional map using multidimensional scaling and hierarchical cluster analysis.
Results: A total of 250 of 320 recruited hospital staff returned the questionnaires. The most important resources in daily practice were English journals, text books and literature searching for faculty members, experience, text books and English journals for fellows and text books, experience and peers for residents. Regional journals were the least important resources for all study groups. About 62.7% of residents did not know the meaning of 'number needed to treat', 36.8%'confidence interval', 54.9%'confounding factor' and 44.6%'meta-analysis'. The percentages for faculty members were 41.3%, 37%, 42.2% and 39.1%. The knowledge sources were placed in four clusters in a point map derived from the multidimensional scaling process.
Conclusion: The dominance of the traditional information resources and experience-based medicine debate which is the consequence of traditional approaches to medical education may be one of the considerable barriers to the dissemination of evidence-based medicine in developing countries. The evidence-based clinical practice guidelines could be used as a useful passive-predigested source for busy clinicians to make informed decisions. A considerable Western bias may undermine the local research in developing world.