Scholarlink

<p>Maladaptive health anxiety has serious negative impacts on individuals and society, so there is a need to study the predictors of health anxiety to provide a basis for its precise intervention. This cross-sectional online survey collected demographic and mental health data from 643 Chinese adults, aged 18 to 59, who experienced health anxiety as measured by the Short Health Anxiety Inventory (SHAI). Among them, 324 participants (50.39%) were female. Six machine learning algorithms, including Gaussian Naive Bayes (GNB), K-nearest neighbor algorithm (KNN), logistic regression (LR), random forest (RF), multi-layer perceptron (MLP) and support vector machine (SVM), were used to identify the predictors of health anxiety. Accuracy, precision, recall, F1 score, and the area under the receiver operating characteristic curve (AUC) were used to evaluate the performance of machine learning model. The GNB model performed best in classifying health anxiety, achieving an accuracy of 0.813, precision of 0.846, recall of 0.820, F1 score of 0.827, and an AUC of 0.903. Catastrophic interpretation for specific physical diseases, depression symptoms, general anxiety symptoms, catastrophic interpretation for general bodily complaints, and age were the most important predictors for health anxiety. The GNB model may be helpful to identify high-risk individuals with health anxiety, make health prevention and intervention plans, and enhance the mental health and cognitive development of the general public. Further research is needed to validate the model's results across multicultural and multinational environments.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Catastrophic interpretations as key predictors of health anxiety symptoms: a machine learning approach

  • 期刊论文

  • 发表日期：

    2026年1月17日

查看全文
  • Xiayu Du,
  • Jiawei Hu,
  • Congrong Shi,
  • Xingyun Liu,
  • Zhihong Ren

摘要

Maladaptive health anxiety has serious negative impacts on individuals and society, so there is a need to study the predictors of health anxiety to provide a basis for its precise intervention. This cross-sectional online survey collected demographic and mental health data from 643 Chinese adults, aged 18 to 59, who experienced health anxiety as measured by the Short Health Anxiety Inventory (SHAI). Among them, 324 participants (50.39%) were female. Six machine learning algorithms, including Gaussian Naive Bayes (GNB), K-nearest neighbor algorithm (KNN), logistic regression (LR), random forest (RF), multi-layer perceptron (MLP) and support vector machine (SVM), were used to identify the predictors of health anxiety. Accuracy, precision, recall, F1 score, and the area under the receiver operating characteristic curve (AUC) were used to evaluate the performance of machine learning model. The GNB model performed best in classifying health anxiety, achieving an accuracy of 0.813, precision of 0.846, recall of 0.820, F1 score of 0.827, and an AUC of 0.903. Catastrophic interpretation for specific physical diseases, depression symptoms, general anxiety symptoms, catastrophic interpretation for general bodily complaints, and age were the most important predictors for health anxiety. The GNB model may be helpful to identify high-risk individuals with health anxiety, make health prevention and intervention plans, and enhance the mental health and cognitive development of the general public. Further research is needed to validate the model's results across multicultural and multinational environments.