Department of Epidemiology and Biostatistics, West China School of Public Health and West China Fourth Hospital, Sichuan University, No.17 Section 3, Renmin South Road, Chengdu, 610041, Sichuan, China.
BMC Geriatr. 2021 May 17;21(1):311. doi: 10.1186/s12877-021-02223-9.
Instrumental Activities of Daily Living (IADL) disability is a common health burden in aging populations. The identification of high-risk individuals is essential for timely targeted interventions. Although predictors for IADL disability have been well described, studies constructing prediction tools for IADL disability among older adults were not adequately explored. Our study aims to develop and validate a web-based dynamic nomogram for individualized IADL disability prediction in older adults.
Data were obtained from the China Health and Retirement Longitudinal Study (CHARLS). We included 4791 respondents aged 60 years and over, without IADL disability at baseline in the 2011 to 2013 cohort (training cohort) and 371 respondents in the 2013 to 2015 cohort (validation cohort). Here, we defined IADL disability as needing any help in any items of the Lawton and Brody's scale. A web-based dynamic nomogram was built based on a logistic regression model in the training cohort. We validated the nomogram internally with 1000 bootstrap resamples and externally in the validation cohort. The discrimination and calibration ability of the nomogram was assessed using the concordance index (C-index) and calibration plots, respectively.
The nomogram incorporated ten predictors, including age, education level, social activity frequency, drinking frequency, smoking frequency, comorbidity condition, self-report health condition, gait speed, cognitive function, and depressive symptoms. The C-index values in the training and validation cohort were 0.715 (bootstrap-corrected C-index = 0.702) and 0.737, respectively. The internal and external calibration plots for predictions of IADL disability were in excellent agreement. An online web server was built ( https://lilizhang.shinyapps.io/DynNomapp/ ) to facilitate the use of the nomogram.
We developed a dynamic nomogram to evaluate the risk of IADL disability precisely and expediently. The application of this nomogram would be helpful for health care physicians in decision-making.
日常生活活动能力（IADL）障碍是老年人群中常见的健康负担。识别高危人群对于及时进行有针对性的干预至关重要。尽管已经很好地描述了 IADL 障碍的预测因素，但研究构建老年人 IADL 障碍预测工具的研究还不够充分。我们的研究旨在开发和验证一种基于网络的动态列线图，用于预测老年人的个体化 IADL 障碍。
数据来自中国健康与退休纵向研究（CHARLS）。我们纳入了 2011 年至 2013 年队列（训练队列）中无 IADL 障碍的 4791 名 60 岁及以上的受访者和 2013 年至 2015 年队列（验证队列）中的 371 名受访者。在这里，我们将 IADL 障碍定义为在 Lawton 和 Brody 量表的任何一项中需要任何帮助。基于训练队列中的逻辑回归模型构建了基于网络的动态列线图。我们通过 1000 次 bootstrap 重采样对该列线图进行了内部验证，并在验证队列中进行了外部验证。分别使用一致性指数（C 指数）和校准图评估了列线图的区分度和校准能力。
该列线图纳入了 10 个预测因素，包括年龄、教育程度、社会活动频率、饮酒频率、吸烟频率、合并症情况、自我报告的健康状况、步态速度、认知功能和抑郁症状。训练和验证队列中的 C 指数值分别为 0.715（bootstrap 校正的 C 指数=0.702）和 0.737。IADL 障碍预测的内部和外部校准图显示出极好的一致性。我们构建了一个在线网络服务器（https://lilizhang.shinyapps.io/DynNomapp/），以方便列线图的使用。
我们开发了一种动态列线图，可以准确、便捷地评估 IADL 障碍的风险。该列线图的应用将有助于医疗保健医生做出决策。
