Darekar Amanda, Carlsson Martin, Quinn Sheila, Ntanios Fady, Mangan Erin, Arumi Daniel, Scholfield David
Statistics, Pfizer Ltd., Walton Oaks, Tadworth, UK.
Statistics, Pfizer Inc., New York, NY, USA.
Contemp Clin Trials. 2016 Nov;51:44-49. doi: 10.1016/j.cct.2016.09.005. Epub 2016 Sep 27.
The ability to set realistic expectations of treatment response in patients with overactive bladder (OAB) can have an impact on patient engagement and adherence to study medication. In order to help set treatment expectations for OAB, a Physician Predictive Tool has been developed based on predictive modelling. Models have been developed utilizing data from eight Phase 3 and 4 fesoterodine clinical trials and these models enable the prediction of individual treatment response in subjects with OAB, based on various baseline characteristics. The data utilized and covariates that were hypothesized to influence treatment response are described. The model selection and development process are also outlined, and the final model and some example results utilizing this model are presented. Finally, we discuss the potential benefits and limitations of such a predictive tool.
对膀胱过度活动症(OAB)患者的治疗反应设定现实期望的能力,可能会对患者参与度和对研究药物的依从性产生影响。为了帮助设定OAB的治疗期望,基于预测模型开发了一种医生预测工具。利用来自八项非索罗定3期和4期临床试验的数据开发了模型,这些模型能够根据各种基线特征预测OAB患者的个体治疗反应。描述了所使用的数据以及假设会影响治疗反应的协变量。还概述了模型选择和开发过程,并展示了最终模型以及使用该模型的一些示例结果。最后,我们讨论了这种预测工具的潜在益处和局限性。