Research Article Effects of Diagnostic Errors in Pattern Differentiation and Acupuncture Prescription: A Single-Blinded, Interrater Agreement Study Ingrid Jardim de Azeredo Souza Oliveira and Arthur de Sá Ferreira Laborat´ orio de Simulac ¸˜ ao Computacional e Modelagem em Reabilitac ¸˜ ao, Programa de P´ os-graduac ¸˜ ao em Ciˆ encias da Reabilitac ¸˜ ao, Centro Universit´ ario Augusto Motta (UNISUAM), 21041-010 Rio de Janeiro, RJ, Brazil Correspondence should be addressed to Arthur de S´ a Ferreira; arthur sf@ig.com.br Received 11 January 2015; Revised 14 March 2015; Accepted 15 March 2015 Academic Editor: Hongcai Shang Copyright © 2015 I. J. A. S. Oliveira and A. de S´ a Ferreira. Tis is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Tis study compared the interrater agreement for pattern diferentiation and acupoints prescription between two groups of human patients simulated with diferent diagnostic outcomes. Patients were simulated using a dataset about zangfu patterns and separated into groups ( = 30 each) according to the diagnostic outcome determined by a computational model. A questionnaire with 90 patients was delivered to 6 TCM experts (4-year minimal of clinic experience) who were asked to indicate a single pattern (among 73) and 8 acupoints (among 378). Interrater agreement was higher for pattern diferentiation than for acupuncture prescription. Interrater agreement on pattern diferentiation was slight for both groups with correct (Light’s  = 0.167, 95% CI = [0.108; 0.254]) and incorrect diagnosis (Light’s  = 0.190, 95% CI = [0.120; 0.286]). Interrater agreement on acupuncture prescription was slight for both groups of correct ( = 0.029, 95% CI = [0.015; 0.057]) and incorrect diagnosis ( = 0.040, 95% CI = [0.023; 0.058],  = 0.075). Diagnostic performance of raters yielded the following: accuracy = 60.9%, sensitivity = 21.7%, and specifcity = 100%. An overall improvement in the interrater agreement and diagnostic accuracy was observed when the data were analyzed using the internal systems instead of the pattern’s labels. 1. Introduction Diagnostic errors are difcult to recognize but are not rare in the Western practice [1]. Te difculty for dif- ferentiating between two closely related or similar diag- noses, with possibly very diferent prognosis or therapeutic options, is acknowledged as a source of error [2, 3]. Tradi- tional medicines, in particular traditional Chinese medicine (TCM), are no exceptions. Te systematic-philosophic rela- tionships between humans and nature applied by TCM experts [4] do not guarantee an error-free diagnostic process [5]; diferent patterns that require distinct therapeutic choices regarding acupuncture prescription might also be confused. In contrast with the available treatment regimens for various diseases in the Western medicine, there are no defned protocols of acupoints for patterns mainly because of both the personalized approach of TCM’s diagnostic process and the large variety of criteria for selecting acupoints [6]. In this sense, high interrater agreements, that is, the degree to which raters achieve identical results under similar assessment conditions rating the same items [7], alongside an accurate diagnosis are two important characteristics of diagnostics models. Previous studies reported a variable degree of interrater agreement on pattern diferentiation and/or therapeutic pre- scription [8–16], though they present important limitations either from the TCM or scientifc perspectives. For instance, there was a lack of calculating and/or reporting statistics of agreement [9–11, 13] or a lack of investigating the pattern diferentiation and therapeutic prescription for the same cases [10, 12–16]. Most importantly, all the above-cited studies used real human patients with a narrow range of diseases and corresponding TCM patterns, in which the true patterns were unknown and, therefore, it was not possible to assess Hindawi Publishing Corporation Evidence-Based Complementary and Alternative Medicine Volume 2015, Article ID 469675, 11 pages http://dx.doi.org/10.1155/2015/469675