Introduction. The societal burden of cognitive impairments in China has prompted researchers to develop clinical prediction models aimed at making risk assessments that enable preventative interventions. However, it is unclear which risk factors best predict future cognitive impairment and if predictive ability is consistent across different socioeconomic groups. Methods. We quantified the ability of demographics, instrumental activities of daily living, activities of daily living, cognitive tests, social factors, psychological factors, diet, exercise and sleep, chronic diseases, and three recently published prediction models predict future cognitive impairments in the general Chinese population and among male, female, rural, urban, educated, and uneducated elderly. Data were taken from the 2011 and 2014 waves of the Chinese Longitudinal Healthy Longevity Survey (CLHLS). Results. The risk factor groups with the most predictive ability in the general population were demographics (AUC, 0.78, 95% CI, 0.77-0.78), cognitive tests (AUC, 0.72, 95% CI, 0.72-0.73), and instrumental activities of daily living (AUC, 0.71, 95% CI, 0.70-0.71). Demographics, cognitive tests, instrumental activities of daily living, and all three re-created prediction models had significantly higher AUCs when making predictions among women compared to men and among the uneducated compared to the educated. Discussion. This study suggests that demographics, cognitive tests, and instrumental activities of daily living are the most useful risk factors for predicting future cognitive impairment among elderly Chinese. However, the most useful risk factors and existing models have lower predictive power among male, urban, and educated elderly. More efforts are needed to ensure that equally accurate risk assessments can be conducted across different socioeconomic groups in China.

翻译：引言。中国认知障碍的社会负担促使研究人员开发临床预测模型，旨在进行风险评估以实现预防性干预。然而，目前尚不清楚哪些风险因素能最佳预测未来认知障碍，以及其预测能力在不同社会经济群体中是否一致。方法。我们量化了人口统计学特征、工具性日常生活活动能力、日常生活活动能力、认知测试、社会因素、心理因素、饮食、运动与睡眠、慢性疾病，以及三个近期发表的预测模型对中老年中国普通人群及男性、女性、农村、城市、受过教育和未受教育老年人群未来认知障碍的预测能力。数据来自2011年和2014年两波中国老年健康影响因素跟踪调查（CLHLS）。结果。在普通人群中预测能力最强的风险因素组为人口统计学特征（AUC，0.78，95% CI，0.77-0.78）、认知测试（AUC，0.72，95% CI，0.72-0.73）和工具性日常生活活动能力（AUC，0.71，95% CI，0.70-0.71）。人口统计学特征、认知测试、工具性日常生活活动能力以及所有三个重建的预测模型在女性群体中的预测AUC显著高于男性群体，在未受教育群体中显著高于受教育群体。讨论。本研究表明，人口统计学特征、认知测试和工具性日常生活活动能力是预测中国老年人未来认知障碍最有效的风险因素。然而，最有效的风险因素和现有模型在男性、城市和受过教育老年人群中的预测能力较低。需进一步努力确保在中国不同社会经济群体中能够进行同等准确的风险评估。