An Overview of Introductory and Advanced Survival Analysis Methods in Clinical Applications: Where Have we Come so far?

The time-to-event relationship for survival<br /> modeling is considered when designing a study in clinical<br /> trials. However, because time-to-event data are mostly not<br /> normally distributed, survival analysis uses non-parametric<br /> data processing and analysis methods, mainly Kaplan-Meier<br /> (KM) estimation models and Cox proportional hazards (CPH)<br /> regression models. At the same time, the log-rank test can be<br /> applied to compare curves from different groups. However,<br /> resorting to conventional survival analysis when fundamental<br /> assumptions, such as the Cox PH assumption, are not met can<br /> seriously affect the results, rendering them flawed.<br /> Consequently, it is necessary to examine and report more<br /> sophisticated statistical methods related to the processing of<br /> survival data, but at the same time, able to adequately respond<br /> to the contemporary real problems of clinical applications. On<br /> the other hand, the frequent misinterpretation of survival<br /> analysis methodology, combined with the fact that it is a<br /> complex statistical tool for clinicians, necessitates a better<br /> understanding of the basic principles underlying this analysis<br /> to effectively interpret medical studies in making treatment<br /> decisions. In this review, we first consider the basic models and<br /> mechanisms behind survival analysis. Then, due to common<br /> errors arising from the inappropriate application of<br /> conventional models, we revise more demanding statistical<br /> extensions of survival models related to data manipulation to<br /> avoid wrong results. By providing a structured review of the<br /> most representative statistical methods and tests covering<br /> contemporary survival analysis, we hope this review will assist<br /> in solving problems that arise in clinical applications