PDF(1574 KB)
PDF(1574 KB)
PDF(1574 KB)
限制性立方样条在临床研究数据分析中的应用
({{custom_author.role_cn}}), {{javascript:window.custom_author_cn_index++;}}Application of restricted cubic splines in clinical research data analysis
({{custom_author.role_en}}), {{javascript:window.custom_author_en_index++;}}临床研究中，时常需要构建回归模型分析自变量与因变量之间的关系。然而，回归模型大多要求自变量与因变量呈线性关联。当自变量与因变量不满足上述条件时，将连续型变量转化为分类变量会导致损失部分信息并可能引进新的偏倚。因此，这种情况下就需要构建样条回归直接拟合自变量与因变量之间的非线性关系，这种分析常用的方法就是限制性立方样条。本文从概念入手，就开展临床研究数据分析时，如何应用限制性立方样条拟合自变量与因变量之间的非线性关系进行阐述，以期为临床医务人员开展临床研究提供参考。
In clinical research, it is often necessary to construct regression models to analyze the relationship between independent variables and dependent variables. However, most regression models require a linear association of the independent variables with the dependent variable. When the independent variable and the dependent variable do not satisfy these conditions, transforming continuous variables into categorical variables will result in the loss of some information and may introduce new biases. Therefore, in this case, it is necessary to construct a spline regression to directly fit the nonlinear relationship between the independent variable and the dependent variable, and a common method for this analysis is the restricted cubic spline. This article starts with the concept of how to apply a restricted cubic spline to fit the nonlinear relationship between the independent variable and the dependent variable when analyzing clinical research data, so as to provide reference for clinical medical personnel to conduct clinical research.
限制性立方样条 / 非线性 / R语言 {{custom_keyword}} /
restricted cubic spline / nonlinear / the R programming language {{custom_keyword}} /
表1 RCS节点位置推荐 |
| 节点数量 | 节点1 | 节点2 | 节点3 | 节点4 | 节点5 | 节点6 | 节点7 |
|---|---|---|---|---|---|---|---|
| 3 | 0.100 0 | 0.500 0 | 0.900 0 | ||||
| 4 | 0.050 0 | 0.350 0 | 0.650 0 | 0.950 0 | |||
| 5 | 0.050 0 | 0.275 0 | 0.500 0 | 0.725 0 | 0.950 0 | ||
| 6 | 0.050 0 | 0.230 0 | 0.410 0 | 0.590 0 | 0.770 0 | 0.950 0 | |
| 7 | 0.025 0 | 0.183 3 | 0.341 7 | 0.500 0 | 0.658 3 | 0.816 7 | 0.975 0 |
| [1] |
The semi-parametric regression achieved via penalized spline smoothing can be expressed in a linear mixed models framework. This allows such models to be fitted using standard mixed models software routines with which many biostatisticians are familiar. Moreover, the analysis of complex correlated data structures that are a hallmark of biostatistics, and which are typically analysed using mixed models, can now incorporate directly smoothing of the relationship between an outcome and covariates. In this paper we provide an introduction to both linear mixed models and penalized spline smoothing, and describe the connection between the two. This is illustrated with three examples, the first using birth data from the U.K., the second relating mammographic density to age in a study of female twin-pairs and the third modelling the relationship between age and bronchial hyperresponsiveness in families. The models are fitted in R (a clone of S-plus) and using Markov chain Monte Carlo (MCMC) implemented in the package WinBUGS.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [2] |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [3] |
巩浩雯, 熊殷, 刘玉秀, 等. 几种非线性混杂变量校正方法的性能比较[J]. 中国卫生统计, 2023, 40(3): 326-330.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [4] |
罗剑锋, 金欢, 李宝月, 等. 限制性立方样条在非线性回归中的应用研究[J]. 中国卫生统计, 2010, 27(3): 229-232.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [5] |
王晓晓, 陶立元, 李楠, 等. 限制性立方样条在非线性关联分析中的应用[J]. 中华儿科杂志, 2020, 58(8): 652.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [6] |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| [7] |
Sleep duration, psychiatric disorders and dementias are closely interconnected in older adults. However, the underlying genetic mechanisms and brain structural changes are unknown. Using data from the UK Biobank for participants primarily of European ancestry aged 38-73 years, including 94% white people, we identified a nonlinear association between sleep, with approximately 7 h as the optimal sleep duration, and genetic and cognitive factors, brain structure, and mental health as key measures. The brain regions most significantly underlying this interconnection included the precentral cortex, the lateral orbitofrontal cortex and the hippocampus. Longitudinal analysis revealed that both insufficient and excessive sleep duration were significantly associated with a decline in cognition on follow up. Furthermore, mediation analysis and structural equation modeling identified a unified model incorporating polygenic risk score (PRS), sleep, brain structure, cognition and mental health. This indicates that possible genetic mechanisms and brain structural changes may underlie the nonlinear relationship between sleep duration and cognition and mental health.© 2022. The Author(s), under exclusive licence to Springer Nature America, Inc.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
| {{custom_ref.label}} |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
PDF(1574 KB)
沪公网安备31010102003426号 技术支持：北京玛格泰克科技发展有限公司
/
| 〈 |
|
〉 |
