Authors' reply to the comment by S abour

Le résumé fourni par la source

Recently, we published an article in the European Journal of Pain (van Wijk et al., 2013), in which the reliability and validity of a continuous pain registration procedure were investigated. The idea behind this study was that if pain is rated continuously, one can calculate the area under the curve of varying pain intensity over time, thereby obtaining a pain measure that integrates varying pain intensity with the duration of such intensities. After publishing the manuscript, a Letter to the Editor by Sabour and Ghassemi, commenting on our article, was accepted for publication. We appreciate their interest in our work and hereby respond to their comment. In their Letter, Sabour and Ghassemi provide us with a summary from a freshmen's course in statistics by listing methods to determine reliability and validity of a single diagnostic test compared to a golden standard. Their critique on our article is that we used the wrong methods, and that our title would be misleading. However, in our study, the reliability and validity of a new measurement procedure were investigated; not those of a single diagnostic test (with a dichotomous outcome) compared to a golden standard. A new continuous measure can be validated by looking at convergent, divergent and predictive/discriminative validity (Judd et al., 1991), using for instance correlations and t-tests, and tested for reproducibility using the intraclass correlation coefficient. Hence, we suspect that Sabour and Ghassemi must have misread or misinterpreted our manuscript. A small literature search demonstrates that Sabour quite often comments on original research articles, thereby not shying away from economically recycling (parts of) previously written comments. Compare, for instance, their current Letter to a previous one (Sabour and Dastjerdi, 2012). We feel that this does not stimulate scientific discussion, but rather reflects an attempt to score impact points at the expense of real scientific work.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.