Estimation of cut-off value for the Cyberchondria Severity Scale-12 in Chinese population

Abstract

This study aimed to establish cutoff points for the Cyberchondria Severity Scale–12 (CSS-12) to enhance its use in clinical and research settings. Data were collected via a cross-sectional web-based survey in China. Pearson correlation analyses were used to examine associations between the CSS-12, GAD-7, and PHQ-9, receiver operating characteristic (ROC) curves were used to determine optimal cutoff values, along with corresponding sensitivity and specificity, using GAD-7 and PHQ-9 scores to classify CYB presence. Moderate correlations were observed between the CSS-12 and the other measures. The ROC analyses identified a CSS-12 cutoff score of 36 for distinguishing CYB from non-CYB, with performance metrics as follows: using the GAD-7, sensitivity was 85.2%, specificity 55.3%, and area under the curve (AUC) 0.69; using the PHQ-9, sensitivity was 81.4%, specificity 51.8%, and AUC 0.67. This cutoff of 36 improves the specificity of the CSS-12, thereby facilitating more accurate identification and management of CYB.