OBJECTIVES: 1) Determine whether textual Clinical Practice Guidelines (CPGs) are interpreted accurately and unequivocally by targeted physicians. 2) Specify audience and perception of the CPGs.

METHOD: Comparative analysis of answers given by a panel of general practitioners to a series of questions and clinical case studies related to three textual CPGs produced and published by the French National Agency for Accreditation and Evaluation in Health (ANAES).

RESULTS: 68 to 96% of physicians are aware of the existence of the CPGs studied. Less than 50% state having read them. On average, 38% of physician interpretations of CPGs are incorrect (i.e., not in agreement with expert interpretation). Furthermore, there is disagreement among physicians responses.

CONCLUSION: This study credits the argument of disparities in practice which derive from inaccurate and discordant CPGs' interpretations. The results should prompt those responsible for producing such decision-making support to design documents that are better structured, less ambiguous, and more precise. In a model which facilitates their computerisation the expression of CPGs provides a solution that should be included upstream in the publication process.