TY - JOUR

T1 - Erratum to

T2 - Quantitative evaluation of the clinical efficacy of attention bias modification treatment for anxiety disorders (Depression and Anxiety, 32, 6, (383-391), 10.1002/da.22344)

AU - Linetzky, Marian

AU - Pergamin-Hight, Lee

AU - Pine, Daniel S.

AU - Bar-Haim, Yair

N1 - Publisher Copyright:
© 2017 Wiley Periodicals, Inc.

PY - 2018/1

Y1 - 2018/1

N2 - An error occurred in Table 1 of the published article. Specifically, the 95% confidence intervals depicted in this table are all of the same size. We now tracked down the origin of this error, which resulted from our estimation of the variance of d based on the four studies that provided information allowing calculation of variance and then entered the mean of these variances for all studies when producing our Table 1. Instead, a more accurate way should have been to first estimate the magnitude of the correlation (r) between pre- and posttreatment values and then use this estimate to calculate the variance of d for each study individually (see formula). When reports on pre–post correlations were missing, we used correlations from other studies with the same type of population and measure. Alternatively, if such proxy was not available, we used the mean r (rather than mean variance) based on the four studies that reported the needed data. Vd = (1/n + d2/2n)2(1 - r) Standard difference in means within groups for clinician-ratings and self-reported measures of anxiety (Table presented.) Importantly, the recalculations indicate that the magnitude and significance of the relevant effects changed only very slightly and do not call for any change in results interpretation. Next please find the corrected Table 1. The mean of each study remains the same but the 95% CIs now reflect the specific estimate for each study as described above. In terms of the reported effects, for self-reported measures in the attention bias modification treatment (ABMT) condition, d = 0.83, P <.001 (previously d = 0.89, P <.001) relative to the control condition, d = 0.48, P <.001 (previously d = 0.49, P <.001), Q = 3.01, P =.08 (previously Q = 3.49, P =.06). For clinician-ratings in the ABMT condition, d = 0.94, P <.001 (previously d = 1.00, P <.001) relative to control, d = 0.44, P <.001 (previously d = 0.42, P <.001), Q = 8.5, P <.001 (previously Q = 7.25, P <.01). We apologize for the error.

AB - An error occurred in Table 1 of the published article. Specifically, the 95% confidence intervals depicted in this table are all of the same size. We now tracked down the origin of this error, which resulted from our estimation of the variance of d based on the four studies that provided information allowing calculation of variance and then entered the mean of these variances for all studies when producing our Table 1. Instead, a more accurate way should have been to first estimate the magnitude of the correlation (r) between pre- and posttreatment values and then use this estimate to calculate the variance of d for each study individually (see formula). When reports on pre–post correlations were missing, we used correlations from other studies with the same type of population and measure. Alternatively, if such proxy was not available, we used the mean r (rather than mean variance) based on the four studies that reported the needed data. Vd = (1/n + d2/2n)2(1 - r) Standard difference in means within groups for clinician-ratings and self-reported measures of anxiety (Table presented.) Importantly, the recalculations indicate that the magnitude and significance of the relevant effects changed only very slightly and do not call for any change in results interpretation. Next please find the corrected Table 1. The mean of each study remains the same but the 95% CIs now reflect the specific estimate for each study as described above. In terms of the reported effects, for self-reported measures in the attention bias modification treatment (ABMT) condition, d = 0.83, P <.001 (previously d = 0.89, P <.001) relative to the control condition, d = 0.48, P <.001 (previously d = 0.49, P <.001), Q = 3.01, P =.08 (previously Q = 3.49, P =.06). For clinician-ratings in the ABMT condition, d = 0.94, P <.001 (previously d = 1.00, P <.001) relative to control, d = 0.44, P <.001 (previously d = 0.42, P <.001), Q = 8.5, P <.001 (previously Q = 7.25, P <.01). We apologize for the error.

UR - http://www.scopus.com/inward/record.url?scp=85036540280&partnerID=8YFLogxK

U2 - 10.1002/da.22690

DO - 10.1002/da.22690

M3 - תגובה / דיון

AN - SCOPUS:85036540280

VL - 35

SP - 111

EP - 112

JO - Depression and Anxiety

JF - Depression and Anxiety

SN - 1091-4269

IS - 1

ER -