Logo Logo
Help
Contact
Switch language to German
Graphical presentation of patient-treatment interaction elucidated by continuous biomarkers
Graphical presentation of patient-treatment interaction elucidated by continuous biomarkers
The translation of complex statistical results into clinical practice on the roles of continuous biomarkers in patient-treatment interactions is greatly aided by clear graphical presentation. To combat the current lack of comprehensive reviews or adequate guides on graphical presentation within this topic, our study formulates guiding principles for continuous biomarker in patient treatment interaction (CBPTI) plots. In order to understand current practice, we review the development of CBPTI methodology and how CBPTI plots are currently used in clinical research. Several criteria for a good CBPTI plot are derived in this study, including general principles of visual display, appropriate quantification of statistical uncertainty, use of units presenting absolute outcome measures, correct display of benchmarks, and information content for medical decision-making. We examined a representative sample of biostatistics and clinical reports on randomized controlled trials with parallel-group design, based on papers published in four major biostatistics journals and two clinical trial methodology journals from the years 2000-2014, and six major clinical journals from 2013-2014. Each CBPTI plot found was assessed for appropriateness of its presentation and clinical utility. In the systematic review, a total of seven methodological papers and five clinical reports used CBPTI plots which we categorized into four types: distinguishing the outcome effect for each treatment group, showing outcome difference among treatment groups (by either partitioning all individuals into subpopulations or modelling the functional form of the interaction), evaluating the proportion of population impact of the biomarker, and showing the classification accuracy of the biomarker. The current practice of utilizing CBPTI plots in clinical reports suffers from several poor practices: confusing or unclear labelling in the plot, the lack of presentation of statistical uncertainty, the outcome measure scaled by relative unit instead of absolute unit, incorrect use of benchmarks, and being non-informative for medical decision-making. There is considerable scope for improvement in the graphical representation of CBPTI in clinical reports. The existing statistical toolbox is not fully translated into clinical research and also needs improvement. The current challenge is to develop instruments for high-quality graphical plots which can not only convey quantitative concepts to readers with limited statistical knowledge when sophisticated statistical algorithms are undertaken, but also facilitate medical decision-making.
Graphical presentation, patient-treatment interaction, continuous biomarker, randomized parallel-group controlled trial
Shen, Yu-Ming
2016
English
Universitätsbibliothek der Ludwig-Maximilians-Universität München
Shen, Yu-Ming (2016): Graphical presentation of patient-treatment interaction elucidated by continuous biomarkers. Dissertation, LMU München: Faculty of Medicine
[img]
Preview
PDF
Shen_Yu-Ming.pdf

1MB

Abstract

The translation of complex statistical results into clinical practice on the roles of continuous biomarkers in patient-treatment interactions is greatly aided by clear graphical presentation. To combat the current lack of comprehensive reviews or adequate guides on graphical presentation within this topic, our study formulates guiding principles for continuous biomarker in patient treatment interaction (CBPTI) plots. In order to understand current practice, we review the development of CBPTI methodology and how CBPTI plots are currently used in clinical research. Several criteria for a good CBPTI plot are derived in this study, including general principles of visual display, appropriate quantification of statistical uncertainty, use of units presenting absolute outcome measures, correct display of benchmarks, and information content for medical decision-making. We examined a representative sample of biostatistics and clinical reports on randomized controlled trials with parallel-group design, based on papers published in four major biostatistics journals and two clinical trial methodology journals from the years 2000-2014, and six major clinical journals from 2013-2014. Each CBPTI plot found was assessed for appropriateness of its presentation and clinical utility. In the systematic review, a total of seven methodological papers and five clinical reports used CBPTI plots which we categorized into four types: distinguishing the outcome effect for each treatment group, showing outcome difference among treatment groups (by either partitioning all individuals into subpopulations or modelling the functional form of the interaction), evaluating the proportion of population impact of the biomarker, and showing the classification accuracy of the biomarker. The current practice of utilizing CBPTI plots in clinical reports suffers from several poor practices: confusing or unclear labelling in the plot, the lack of presentation of statistical uncertainty, the outcome measure scaled by relative unit instead of absolute unit, incorrect use of benchmarks, and being non-informative for medical decision-making. There is considerable scope for improvement in the graphical representation of CBPTI in clinical reports. The existing statistical toolbox is not fully translated into clinical research and also needs improvement. The current challenge is to develop instruments for high-quality graphical plots which can not only convey quantitative concepts to readers with limited statistical knowledge when sophisticated statistical algorithms are undertaken, but also facilitate medical decision-making.