Ten things to remember about propensity scores.

IF 5.3 1区医学 Q1 ENDOCRINOLOGY & METABOLISM

European Journal of Endocrinology Pub Date : 2024-07-02 DOI:10.1093/ejendo/lvae067

Rolf H H Groenwold, Olaf M Dekkers, Saskia le Cessie

引用次数: 0

Abstract

Propensity score methods are popular to control for confounding in observational biomedical studies of risk factors or medical treatments. This paper focused on aspects of propensity score methods that often remain undiscussed, including unmeasured confounding, missing data, variable selection, statistical efficiency, estimands, the positivity assumption, and predictive performance of the propensity score model.

查看原文本刊更多论文

关于倾向分数的十件事。

倾向评分法在控制风险因素或医疗方法的观察性生物医学研究中的混杂因素方面很受欢迎。本文重点讨论了倾向得分法中经常未被讨论的方面，包括未测量混杂因素、缺失数据、变量选择、统计效率、估计值、阳性假设以及倾向得分模型的预测性能。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

European Journal of Endocrinology 医学-内分泌学与代谢

CiteScore

9.80

自引率

3.40%

发文量

354

审稿时长

1 months

期刊介绍： European Journal of Endocrinology is the official journal of the European Society of Endocrinology. Its predecessor journal is Acta Endocrinologica. The journal publishes high-quality original clinical and translational research papers and reviews in paediatric and adult endocrinology, as well as clinical practice guidelines, position statements and debates. Case reports will only be considered if they represent exceptional insights or advances in clinical endocrinology. Topics covered include, but are not limited to, Adrenal and Steroid, Bone and Mineral Metabolism, Hormones and Cancer, Pituitary and Hypothalamus, Thyroid and Reproduction. In the field of Diabetes, Obesity and Metabolism we welcome manuscripts addressing endocrine mechanisms of disease and its complications, management of obesity/diabetes in the context of other endocrine conditions, or aspects of complex disease management. Reports may encompass natural history studies, mechanistic studies, or clinical trials. Equal consideration is given to all manuscripts in English from any country.