Evaluating the Chinese versions of delirium assessment scales: a diagnostic systematic review

The purpose of this study is to examine the validity, reliability and methodological quality of delirium scales that have been translated and adapted in China using quality assessment tools.

A comprehensive search was conducted in PubMed, Embase, Web of Science, China Knowledge Network (CNKI), VIP database, Wanfang database, and China Biomedical Literature Database. The search covered the period from the establishment of the database until September 1, 2023. Two researchers independently screened the literature and extracted data. Studies were included if they focused on the translation of a delirium scale from English into simplified or traditional Chinese, with a study population aged ≥ 18 years and full text available. The risk of bias was assessed through the QUADAS-2 instrument. Level of evidence recommendation is completed with the GRADE and performed with GRADE GPT. Due to high heterogeneity across studies, a random-effects model was applied to calculate diagnostic accuracy indicators (sensitivity, specificity, and area under the curve index). This study has registered in the prospero.

Thirteen studies were included, of which 2 were case-control studies and 11 were cross-sectional studies. These studies involved 13 adult delirium assessment tools, and were all translated following the Brislin or ISPOR principle. The results of the methodological quality assessment showed that 3D-CAM, 4AT, CAM-ICU, CAM-ICU-7, and S-PTD had higher quality ratings, with 4AT being the highest quality. 4AT, CAM, 3D-CAM, CAM-CR, CAM-ICU, CAM-ICU-7, and Nu-DESC were recommended at a level B. The Cronbach's coefficient of most studies is over 0.8, and the inter-rater reliability of most studies is near or over 0.9, indicating good internal consistency and stability. Besides, a significant inverse correlation was found between these 13 Chinese-adapted delirium scales and their reference tests, with most studies were over 0.7, and especially 3 studies offered S-CVI and I-CVI value that were all larger than 0.9, indicating a good discriminate validity and content validity, though different cut-off points were recommended by different scales. Moreover, the sensitivity and specificity of these studies were mainly larger than 0.9, which proved the good diagnostic accuracy of these included scales. The pooled sensitivity of 7 Chinese delirium adaption scales that provided statistical data is 0.93 (95% CI: 0.89-0.96), and the pooled specificity is 0.94 (95% CI: 0.94-0.96), and the AUC is 0.98 (95% CI: 0.96-0.99).

The research on the Chinese adaptation of the delirium scale in China is relatively abundant and of acceptable quality. Taking into account factors of methodological quality and diagnostic accuracy, Chinese-adapted delirium assessment scales such as 3D-CAM, 4AT, CAM, CAM-ICU, CAM-ICU-7, and NuDESC appear to be suitable alternatives to the original English delirium scales and are recommended for use in primary care settings in China. Future research and continuous optimization are needed to improve the scientific rigor and accuracy of these tools, which will help advance the field. No Patient or Public Contribution is considered.