• AGREE instrument;
  • clinical practice guideline;
  • thyroid cancers;
  • thyroid nodules


Objectives  We assessed the quality of a sample of clinical guidelines for thyroid nodules and thyroid cancers, using the Appraisal of Guidelines Research and Evaluation (AGREE) instrument. We also evaluated the reliability and validity of the AGREE instrument and summarized the key recommendations of the appraised guidelines.

Methods  Twenty-six clinical researchers and endocrinologists who had been trained in the principles of developing clinical guidelines and using the AGREE instrument participated in the study and appraised the guidelines. Clinical guidelines selected via a systematic search were assessed, each by eight participants. We compared the AGREE domain scores of the guidelines, and compared the participants' scores before and after group discussions. We used Cronbach's alpha, and intraclass correlation coefficients to assess the reliability, and the Spearman's rho to assess the correlation between the overall assessment and other variables.

Results  Seven guidelines were included in the study. ‘Scope and purpose’ and ‘clarity and presentation’ achieved the highest domain scores. ‘Applicability’ received the lowest domain scores and reliability coefficients. ‘Rigor of development’ and ‘clarity and presentation’ obtained the highest correlations with overall assessment scores. There was a significant relationship between the overall assessment score and the numbers of algorithms, tables and figures in the guidelines.

Conclusions  We identified three clinical guidelines that obtained high overall assessment scores and were recommended for use in practice. Our findings have important implications for those developing clinical guidelines, especially as clarity and presentation significantly influenced the participants' assessment of the guidelines. The developers should ensure that the recommendations are presented clearly and unambiguously, and flowcharts, algorithms and other tools are developed to help the users in applying the recommendations into practice. The optimal number of appraisers for each guideline is four. Further work is needed to improve the ‘applicability’ domain of the AGREE.