Jacobs Bart K M, Maseko Tafadzwa, Lynen Lutgarde, Henriquez-Trujillo Aquiles Rodrigo, Buyze Jozefien
Department of Clinical Sciences, Institute of Tropical Medicine, Nationalestraat 155, 2000, Antwerp, Belgium.
Faculty of Medicine and Health Sciences, University of Antwerp, Campus Drie Eiken, Building S, Universiteitsplein 1, 2610, Wilrijk, Antwerp, Belgium.
BMC Med Res Methodol. 2025 Jun 3;25(1):152. doi: 10.1186/s12874-025-02591-5.
There is still demand for algorithms that can be used at the point of care, especially when dealing with events that do not present with a single obvious clinical indicator. The Spiegelhalter-Knill-Jones (SKJ) method is an approach for the development of a clinical score that focuses on the effect size of predictors, which is more relevant in settings where events may be rare or data is scarce. However, it does require predictors to be binary or dichotomised.
We developed an extension of the Spiegelhalter-Knill-Jones method that can include continuous variables and added additional features that make it more useful in a variety of settings. We illustrated our method on two historical datasets dealing with viral failure in HIV patients in Cambodia. We used area under the curve (AUC) and risk classification improvement (RCI) as metrics to evaluate the performance of resulting predictions scores and risk classifications.
All new features worked as intended. Scoring systems developed with the new method outperformed an earlier application of a classic version of SKJ method on the training dataset, while no significant difference was found on any of the performance measures in the test dataset.
This extension provides a useful tool for clinical decision-making that is much more flexible than the original version of SKJ, and can be applied in a variety of settings.
对于可在医疗现场使用的算法仍有需求，尤其是在处理没有单一明显临床指标的情况时。斯皮格尔哈特-尼尔-琼斯（SKJ）方法是一种开发临床评分的方法，该方法侧重于预测因素的效应大小，这在事件可能罕见或数据稀缺的情况下更为相关。然而，它确实要求预测因素为二元或二分变量。
我们开发了斯皮格尔哈特-尼尔-琼斯方法的扩展版本，该版本可以纳入连续变量，并添加了使其在各种情况下更有用的其他特征。我们在柬埔寨两个关于HIV患者病毒治疗失败的历史数据集上展示了我们的方法。我们使用曲线下面积（AUC）和风险分类改善（RCI）作为指标来评估所得预测分数和风险分类的性能。
所有新特征都按预期发挥了作用。使用新方法开发的评分系统在训练数据集上的表现优于早期应用的经典SKJ方法，而在测试数据集的任何性能指标上均未发现显著差异。
此扩展为临床决策提供了一个有用的工具，它比原始版本的SKJ灵活得多，并且可以应用于各种情况。
