The search for stable prognostic models in multiple imputed data sets.

The search for stable prognostic models in multiple imputed data sets.
复制标题

DOI:
10.1186/1471-2288-10-81
复制
发表时间:
2010-09-17
影响因子:
4
通讯作者:
van der Windt DA
van der Windt DA
中科院分区:
医学3区
文献类型:
--
作者:
Vergouw D;Heymans MW;Peat GM;Kuijpers T;Croft PR;de Vet HC;van der Horst HE;van der Windt DA

文献摘要

参考文献

被引文献

相似文献

在预后研究中,模型不稳定和数据缺失可能是令人不安的因素。针对这些情况提出的处理方法有bootstrapping (B)和Multiple imputation (MI)。作者考察了这些方法对模型组成的影响。模型是在2001年1月至2003年1月期间在荷兰的一般实践中咨询的587例肩部问题患者(荷兰肩部研究)中构建的。结果测量为持续性肩失能和持续性肩痛。潜在的预测因素包括社会人口统计学变量、疼痛问题的特征、身体活动和心理社会因素。使用完整的案例分析、MI、bootstrap或MI和bootstrap来评估模型的组成和性能(校准和判别)。结果表明,由于缺失数据的处理方式不同,模型组成不同,而自举提供了所选预测模型稳定性的额外信息。在预测建模中,缺失的数据需要由人工智能处理,为了提供关于模型稳定性的信息,建议选择自举模型。
In prognostic studies model instability and missing data can be troubling factors. Proposed methods for handling these situations are bootstrapping (B) and Multiple imputation (MI). The authors examined the influence of these methods on model composition. Models were constructed using a cohort of 587 patients consulting between January 2001 and January 2003 with a shoulder problem in general practice in the Netherlands (the Dutch Shoulder Study). Outcome measures were persistent shoulder disability and persistent shoulder pain. Potential predictors included socio-demographic variables, characteristics of the pain problem, physical activity and psychosocial factors. Model composition and performance (calibration and discrimination) were assessed for models using a complete case analysis, MI, bootstrapping or both MI and bootstrapping. Results showed that model composition varied between models as a result of how missing data was handled and that bootstrapping provided additional information on the stability of the selected prognostic model. In prognostic modeling missing data needs to be handled by MI and bootstrap model selection is advised in order to provide information on model stability.
DOI: 10.1198/0003130043277
发表时间: 2004-05-01
影响因子: 1.8
作者:
Austin, PC;Tu, JV
通讯作者: Tu, JV
DOI: 10.1186/1471-2288-9-64
发表时间: 2009-09-14
影响因子: 4
作者:
Beyene, Joseph;Atenafu, Eshetu G.;Sung, Lillian
通讯作者: Sung, Lillian
DOI: 10.1007/bf02294361
发表时间: 1987-09-01
期刊: PSYCHOMETRIKA
影响因子: 3
作者:
BOZDOGAN, H
通讯作者: BOZDOGAN, H
DOI: 10.1503/cmaj.1041159
发表时间: 2005-06-07
影响因子: 14.6
作者:
Dionne, CE;Bourbonnais, R;Larocque, I
通讯作者: Larocque, I
DOI: 10.1186/1471-2288-7-33
发表时间: 2007-07-13
影响因子: 4
作者:
Heymans, Martijn W.;van Buuren, Stef;de Vet, Henrica C. W.
通讯作者: de Vet, Henrica C. W.