TY - JOUR
T1 - Corrigendum to “Quality of clinical practice guidelines in Japan remains low
T2 - a cross-sectional meta-epidemiological study” [Journal of Clinical Epidemiology 138 (2021) 10550] (Journal of Clinical Epidemiology (2021) 138 (22–31), (S0895435621002031), (10.1016/j.jclinepi.2021.06.025))
AU - Kataoka, Yuki
AU - Anan, Keisuke
AU - Taito, Shunsuke
AU - Tsujimoto, Yasushi
AU - Kurata, Yasuko
AU - Wada, Yoshitaka
AU - Maruta, Masaki
AU - Kanaoka, Koshiro
AU - Oide, Shiho
AU - Takahashi, Sei
AU - Nango, Eishu
N1 - Publisher Copyright:
© 2023 Elsevier Inc.
PY - 2023/10
Y1 - 2023/10
N2 - The authors regret that upon reviewing this article recently, we discovered an error in the calculation method used for the AGREE II scores, which unfortunately, was not noticed during the writing and editing stages. The authors deeply apologize for any confusion or misinterpretation this might have caused. In the original text, the calculation of the AGREE II scores were done by dividing the theoretical maximum score by the obtained score. However, the correct calculation should be scaled after subtracting the minimum possible score. This error overestimated the quality of guidelines. Actual formula [Formula presented] Correct formula [Formula presented] The mistakes appear in the Abstract, the Results section, Figure 2, and Table 3, as follows: Abstract Original: The median score was 0.54 (IQR, 0.38-0.62) for Stakeholder involvement, 0.57 (IQR, 0.51–0.66) in Rigor of development, 0.33 (IQR 0.21–0.46) in Applicability, and 0.63 (IQR 0.46–0.73) in Editorial independence. The number of guideline developers/clinical question ratio (odds ratio [OR]: 4.14, 95% confidence interval [CI]: 1.97, 8.70) and the adopted guideline development methods (OR: 3.69, 95% CI: 1.14, 12.0) were significantly related to the Rigor of development. Corrected: The median score was 0.46 (IQR, 0.27-0.56) for Stakeholder involvement, 0.53 (IQR, 0.47-0.64) in Rigor of development, 0.22 (IQR 0.07-0.36) in Applicability, and 0.56 (IQR 0.38-0.69) in Editorial independence. The number of guideline developers/clinical question ratio (odds ratio [OR]: 3.72, 95% confidence interval [CI]: 1.75, 7.90) and the adopted guideline development methods (OR: 4.06, 95% CI: 1.11, 14.9) were significantly related to the Rigor of development. 3.3 Quality of the included SR-CPGs Original: The median score was 0.81 (IQR, 0.64–0.92) for Scope and purpose, 0.54 (IQR 0.38–0.62) for Stakeholder involvement, 0.57 (IQR, 0.51–0.66) for Rigor of development, 0.82 (IQR 0.77–0.92) for Clarity of presentation, 0.33 (IQR 0.21–0.46) for Applicability, 0.63 (IQR 0.46–0.73) in Editorial independence, and 0.49 (IQR 0.39–0.57) for total quality. Corrected: The median score was 0.78 (IQR, 0.58-0.90) for Scope and purpose, 0.46 (IQR 0.27-0.56) for Stakeholder involvement, 0.53 (IQR, 0.47-0.65) for Rigor of development, 0.79 (IQR 0.73-0.90) for Clarity of presentation, 0.22 (IQR 0.07-0.36) for Applicability, 0.56 (IQR 0.38-0.69) in Editorial independence, and 0.40 (IQR 0.29-0.50) for total quality. 3.4. Factors related to high Rigor of development quality Original: Logistic regression analysis with quartiles of the higher number of guideline developers/CQ ratio was significantly related to high quality (odds ratio [OR]: 4.14, 95% confidence intervals [CI]: 1.97, 8.70). The adopted guideline development methods (GRADE, Minds 2014, or after vs. others) were also significantly related (OR: 3.69, 95% CI: 1.14, 12.0). Corrected: Logistic regression analysis with quartiles of the higher number of guideline developers/CQ ratio was significantly related to high quality (odds ratio [OR]: 3.72, 95% confidence intervals [CI]: 1.75, 7.90). The adopted guideline development methods (GRADE, Minds 2014, or after vs. others) were also significantly related (OR: 4.06, 95% CI: 1.11, 14.9). Figure 2 Original:[Formula presented] Corrected:[Formula presented] Table 3 Relationship with the high rigor of development quality of AGREE II Original: [Formula presented] Corrected: [Formula presented] Importantly, the authors would like to emphasize that this error does not alter the overall conclusions of the study. The core findings and implications remain consistent with our original interpretation, despite the mistake in the AGREE II scoring calculation. However, in the spirit of accuracy and precision that underpins our scientific community, the authors believe it is crucial to amend this error.
AB - The authors regret that upon reviewing this article recently, we discovered an error in the calculation method used for the AGREE II scores, which unfortunately, was not noticed during the writing and editing stages. The authors deeply apologize for any confusion or misinterpretation this might have caused. In the original text, the calculation of the AGREE II scores were done by dividing the theoretical maximum score by the obtained score. However, the correct calculation should be scaled after subtracting the minimum possible score. This error overestimated the quality of guidelines. Actual formula [Formula presented] Correct formula [Formula presented] The mistakes appear in the Abstract, the Results section, Figure 2, and Table 3, as follows: Abstract Original: The median score was 0.54 (IQR, 0.38-0.62) for Stakeholder involvement, 0.57 (IQR, 0.51–0.66) in Rigor of development, 0.33 (IQR 0.21–0.46) in Applicability, and 0.63 (IQR 0.46–0.73) in Editorial independence. The number of guideline developers/clinical question ratio (odds ratio [OR]: 4.14, 95% confidence interval [CI]: 1.97, 8.70) and the adopted guideline development methods (OR: 3.69, 95% CI: 1.14, 12.0) were significantly related to the Rigor of development. Corrected: The median score was 0.46 (IQR, 0.27-0.56) for Stakeholder involvement, 0.53 (IQR, 0.47-0.64) in Rigor of development, 0.22 (IQR 0.07-0.36) in Applicability, and 0.56 (IQR 0.38-0.69) in Editorial independence. The number of guideline developers/clinical question ratio (odds ratio [OR]: 3.72, 95% confidence interval [CI]: 1.75, 7.90) and the adopted guideline development methods (OR: 4.06, 95% CI: 1.11, 14.9) were significantly related to the Rigor of development. 3.3 Quality of the included SR-CPGs Original: The median score was 0.81 (IQR, 0.64–0.92) for Scope and purpose, 0.54 (IQR 0.38–0.62) for Stakeholder involvement, 0.57 (IQR, 0.51–0.66) for Rigor of development, 0.82 (IQR 0.77–0.92) for Clarity of presentation, 0.33 (IQR 0.21–0.46) for Applicability, 0.63 (IQR 0.46–0.73) in Editorial independence, and 0.49 (IQR 0.39–0.57) for total quality. Corrected: The median score was 0.78 (IQR, 0.58-0.90) for Scope and purpose, 0.46 (IQR 0.27-0.56) for Stakeholder involvement, 0.53 (IQR, 0.47-0.65) for Rigor of development, 0.79 (IQR 0.73-0.90) for Clarity of presentation, 0.22 (IQR 0.07-0.36) for Applicability, 0.56 (IQR 0.38-0.69) in Editorial independence, and 0.40 (IQR 0.29-0.50) for total quality. 3.4. Factors related to high Rigor of development quality Original: Logistic regression analysis with quartiles of the higher number of guideline developers/CQ ratio was significantly related to high quality (odds ratio [OR]: 4.14, 95% confidence intervals [CI]: 1.97, 8.70). The adopted guideline development methods (GRADE, Minds 2014, or after vs. others) were also significantly related (OR: 3.69, 95% CI: 1.14, 12.0). Corrected: Logistic regression analysis with quartiles of the higher number of guideline developers/CQ ratio was significantly related to high quality (odds ratio [OR]: 3.72, 95% confidence intervals [CI]: 1.75, 7.90). The adopted guideline development methods (GRADE, Minds 2014, or after vs. others) were also significantly related (OR: 4.06, 95% CI: 1.11, 14.9). Figure 2 Original:[Formula presented] Corrected:[Formula presented] Table 3 Relationship with the high rigor of development quality of AGREE II Original: [Formula presented] Corrected: [Formula presented] Importantly, the authors would like to emphasize that this error does not alter the overall conclusions of the study. The core findings and implications remain consistent with our original interpretation, despite the mistake in the AGREE II scoring calculation. However, in the spirit of accuracy and precision that underpins our scientific community, the authors believe it is crucial to amend this error.
UR - http://www.scopus.com/inward/record.url?scp=85171374735&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85171374735&partnerID=8YFLogxK
U2 - 10.1016/j.jclinepi.2023.08.014
DO - 10.1016/j.jclinepi.2023.08.014
M3 - Comment/debate
C2 - 37714794
AN - SCOPUS:85171374735
SN - 0895-4356
VL - 162
SP - 191
EP - 192
JO - Journal of Clinical Epidemiology
JF - Journal of Clinical Epidemiology
ER -