Brief DISCERN, six questions for the evaluation of evidence-based content of health-related websites

Objective: To extract and to validate a brief version of the DISCERN which could identify mental health-related websites with good content quality. Method: The present study is based on the analysis of data issued from six previous studies which used DISCERN and a standardized tool for the evaluation of content quality (evidence-based health information) of 388 mental health-related websites. After extracting the Brief DISCERN, several psychometric properties (content validity through a Factor analysis, internal consistency by the Cronbach's alpha index, predictive validity through the diagnostic tests, concurrent validity by the strength of association between the Brief DISCERN and the original DISCERN scores) were investigated to ascertain its general applicability. Results: A Brief DISCERN composed of two factors and six items was extracted from the original 16 items version of the DISCERN. Cronbach's alpha coefficients were more than acceptable for the complete questionnaire (alpha 0.74) and for the two distinct domains: treatments information (alpha = 0.87) and reliability (alpha 0.83). Sensibility and specificity of the Brief DISCERN cut-off score >= 16 in the detection of good content quality websites were 0.357 and 0.945, respectively. Its predictive positive and negative values were 0.98 and 0.83, respectively. A statistically significant linear correlation was found between the total scores of the Brief DISCERN and those of the original DISCERN (r = 0.84 and p < 0.0005). Conclusion: The Brief DISCERN seems to be a reliable and valid instrument able to discriminate between websites with good and poor content quality. Practice implications: The Brief DISCERN is a simple tool which could facilitate the identification of good information on the web by patients and general consumers. (C) 2009 Elsevier Ireland Ltd. All rights reserved.