Translational statistics and dynamic nomograms
MetadataShow full item record
This item's downloads: 789 (view details)
Translational Medicine, within biomedical and public health research domains, is defined as the convergence of basic and clinical research with the aim to transfer knowledge on the benefits and risks of therapies. The concept of Translational Statistics is proposed to facilitate the integration of biostatistics within clinical research to enhance communication of statistical research findings in an accurate and accessible manner to diverse audiences (e.g. policy makers, patients and the media). The use of appropriate visualisation is central to all areas of statistical research. Providing meaningful graphical representations of data is necessary to identify features about the population from which the data were sampled and may throw up an unsuspected view of the data such as a pattern or unusual observations. Informative graphical representations of statistical models play an important translational role. Static nomograms have been used to visualise statistical models. In this study, we propose the use of dynamic nomograms as a visualisation and translational tool to further aid the communication of the results of a statistical analysis to a non-statistical audience. A visualisation tool for time-to-event data is presented which contains a collection of useful graphical summaries, in particular, the Mean Residual Life function. It includes the classical survival summaries as well as the dynamic prediction for survival function and the mean residual life function as the two attractive alternatives. In theory, most regression-type models presented in the literature could have an accompanying web address to direct the reader to the corresponding dynamic nomogram allowing them to 'interact' with the model to gain insight into the effect of each explanatory variable on the primary response.