Article contents
Us et abus des Échelles d’évaluation en psychopathologie
Published online by Cambridge University Press: 28 April 2020
Résumé
La quantification de la psychopathologie est devenue indispensable en recherche clinique car, même pour les données de laboratoire, la signification statistique n’acquiert de sens que si, par des corrélations avec des données cliniques quantifiées, on lui trouve une signification clinique.
Mais il existe déjà trop d’échelles d’évaluation dont la validité n’a pas été suffisamment éprouvée. Il est grand temps que psychiatres et psychologues cliniciens contribuent à des recherches non plus avec des échelles mais sur ces échelles.
Après un bref historique, l’auteur aborde systématiquement des aspects de la validité, de la fidélité et de la sensibilité des échelles d’évaluation qui lui paraissent importantes et/ou sous-estimées.
Summary
The quantification of psychopathology with the help of psychiatric rating scales has reached the point of no-return: there is no research in psychiatry without a search for its clinical significance, and no research on this clinical significance without rating scales.
But there is a dramatic need for an increase in the number and quality of projects not only with rating scales but also on rating scales. At present, there exist too many such scales and not enough research on their validity vs. limitations; this research could be the field of a more fruitful collaboration between psychiatrists and clinical psychologists.
After a brief survey of the historical background of the quantification of psychopathology (from psychophysics to Wundt, Galton, Kraepelin and the era of psychopharmacology) and of recent literature on the available scales, the author comments on various aspects of quantitative psychopathology which have direct implications on the quality of the measures: self-rating vs. observer ratings, global or specific vs. comprehensive scales, back-translation, use of video in the detection of semantic or conceptual mistranslations, questions and response modalities, incidence of the validation sample on the validity of a scale, orthogonality of the measurements (exclusion of items on anxiety from a depression scale, weighting of item scores), cutoff scores, semi-structured interview, number and criteria of steps of severity, transcultural differences in the pathology threshold, incidence of a time-blind evaluation on the sensitivity of the measurement of change, factor analysis of pre- vs. posttreatment scores etc.
Keywords
- Type
- Editorial
- Information
- Copyright
- Copyright © European Psychiatric Association 1987
References
- 4
- Cited by
Comments
No Comments have been published for this article.