Understanding prognostic models: The example of the PHASES score for unruptured intracranial aneurysms

IF 1.5 4区医学 Q4 CLINICAL NEUROLOGY

Neurochirurgie Pub Date : 2025-04-05 DOI:10.1016/j.neuchi.2025.101666

François Zhu , Tim E. Darsaut , Jean Raymond

{"title":"Understanding prognostic models: The example of the PHASES score for unruptured intracranial aneurysms","authors":"François Zhu , Tim E. Darsaut , Jean Raymond","doi":"10.1016/j.neuchi.2025.101666","DOIUrl":null,"url":null,"abstract":"<div><h3>Background</h3><div>Prognostic studies may inform individuals about the future course of their disease and help clinical decision making, but problems abound.</div></div><div><h3>Methods</h3><div>We summarize a study on the natural history of unruptured intracranial aneurysms (UIAs) and review the various steps in the construction of prognostic models. We emphasize the fundamental inductive problems of prognostic studies that attempt to use the backward road from the extension of patients suffering an event to create a new intensional definition of classes of patients at risk.</div></div><div><h3>Results</h3><div>The first step in a prognostic model is to identify candidate baseline variables to be entered into the model, according to background knowledge, previous studies, and statistical associations with the event of interest. This is a multivariate task. The modeler already knows the outcomes the model is supposed to ‘predict’, so that multiple models are tested against the data until a satisfactory fit is obtained. The variables used to construct the model should not be added in an <em>ad hoc</em> fashion to fit heterogeneous studies. They should be selected in such a fashion as to be exportable outside the study to new patients. An infinite number of models can fit the same data. Thus, the most important step is to validate the prognostic value of the model in patients that were not used to construct the model. In the case of UIAs, this has never been done.</div></div><div><h3>Conclusion</h3><div>Prognostic studies present multiple problems. Unvalidated models should not be used in clinical practice.</div></div>","PeriodicalId":51141,"journal":{"name":"Neurochirurgie","volume":"71 4","pages":"Article 101666"},"PeriodicalIF":1.5000,"publicationDate":"2025-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Neurochirurgie","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0028377025000396","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"CLINICAL NEUROLOGY","Score":null,"Total":0}

引用次数: 0

Abstract

Background

Prognostic studies may inform individuals about the future course of their disease and help clinical decision making, but problems abound.

Methods

We summarize a study on the natural history of unruptured intracranial aneurysms (UIAs) and review the various steps in the construction of prognostic models. We emphasize the fundamental inductive problems of prognostic studies that attempt to use the backward road from the extension of patients suffering an event to create a new intensional definition of classes of patients at risk.

Results

The first step in a prognostic model is to identify candidate baseline variables to be entered into the model, according to background knowledge, previous studies, and statistical associations with the event of interest. This is a multivariate task. The modeler already knows the outcomes the model is supposed to ‘predict’, so that multiple models are tested against the data until a satisfactory fit is obtained. The variables used to construct the model should not be added in an ad hoc fashion to fit heterogeneous studies. They should be selected in such a fashion as to be exportable outside the study to new patients. An infinite number of models can fit the same data. Thus, the most important step is to validate the prognostic value of the model in patients that were not used to construct the model. In the case of UIAs, this has never been done.

Conclusion

Prognostic studies present multiple problems. Unvalidated models should not be used in clinical practice.

查看原文本刊更多论文

方法我们总结了一项关于未破裂颅内动脉瘤（UIAs）自然史的研究，并回顾了构建预后模型的各个步骤。我们强调了预后研究中的基本归纳问题，这些研究试图利用从罹患某种疾病的患者扩展开来的逆向途径，对高危患者的类别进行新的内涵定义。结果预后模型的第一步是根据背景知识、先前的研究以及与相关事件的统计学关联，确定要输入模型的候选基线变量。这是一项多变量任务。建模者已经知道模型要 "预测 "的结果，因此要根据数据对多个模型进行测试，直到得到满意的拟合结果。用于构建模型的变量不应临时添加，以适应不同的研究。在选择变量时，应使其能够在研究之外输出到新的患者身上。有无数个模型可以拟合相同的数据。因此，最重要的一步是在未用于构建模型的患者中验证模型的预后价值。结论预后研究存在多种问题。未经验证的模型不应在临床实践中使用。

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

Neurochirurgie 医学-临床神经学

CiteScore

2.70

自引率

6.20%

发文量

100

审稿时长

29 days

期刊介绍： Neurochirurgie publishes articles on treatment, teaching and research, neurosurgery training and the professional aspects of our discipline, and also the history and progress of neurosurgery. It focuses on pathologies of the head, spine and central and peripheral nervous systems and their vascularization. All aspects of the specialty are dealt with: trauma, tumor, degenerative disease, infection, vascular pathology, and radiosurgery, and pediatrics. Transversal studies are also welcome: neuroanatomy, neurophysiology, neurology, neuropediatrics, psychiatry, neuropsychology, physical medicine and neurologic rehabilitation, neuro-anesthesia, neurologic intensive care, neuroradiology, functional exploration, neuropathology, neuro-ophthalmology, otoneurology, maxillofacial surgery, neuro-endocrinology and spine surgery. Technical and methodological aspects are also taken onboard: diagnostic and therapeutic techniques, methods for assessing results, epidemiology, surgical, interventional and radiological techniques, simulations and pathophysiological hypotheses, and educational tools. The editorial board may refuse submissions that fail to meet the journal''s aims and scope; such studies will not be peer-reviewed, and the editor in chief will promptly inform the corresponding author, so as not to delay submission to a more suitable journal. With a view to attracting an international audience of both readers and writers, Neurochirurgie especially welcomes articles in English, and gives priority to original studies. Other kinds of article - reviews, case reports, technical notes and meta-analyses - are equally published. Every year, a special edition is dedicated to the topic selected by the French Society of Neurosurgery for its annual report.