Each week, we publish lay abstracts of new articles from our prestigious portfolio of journals in statistics. The aim is to highlight the latest research to a broader audience in an accessible format.
The article featured today is from Statistics in Medicine with the full article now available to read here.
The survival-incorporated median vs the median in the survivors or in the always-survivors: What are we measuring? and Why? Statistics in Medicine. 2023; 1–12. doi: 10.1002/sim.9922
, , .In clinical practice, treatment decisions are often based on both survival and other clinical outcomes, such as Quality of Life (QoL) scores or cognitive outcomes. For example, regarding the clinical benefit of a treatment in patients with severe prostate cancer, treatment decisions depend on (1) whether the treatment improves survival and (2) whether the treatment improves Quality of Life. When subjects die before the follow-up assessment, the clinical outcomes become undefined. Many authors refer to this setting as “truncation by death” to distinguish it from settings where the outcome is simply missing. In such settings, treating outcomes as “missing” or “censored” due to death can be misleading for treatment effect evaluation.
The authors advocate not always treating death as a mechanism through which clinical outcomes are missing or censored, but rather as part of the outcome measure. They propose summarizing the clinical benefit of a treatment by combining death and the clinical outcome into a ranked composite outcome, ranking death worse than the clinical outcome. Then, the median of the composite outcome, the survival-incorporated median, is proposed to inform clinical practice.
The authors discuss the survival-incorporated median and compare it with other clinical outcome measures: the median in the survivors and the median in the always survivors. In some settings, even if one treatment regimen improves both survival and clinical outcomes, those clinical outcome measures could still lead to misleading treatment recommendations. However, in such settings, it is safe to use the survival-incorporated median. The authors also apply the survival-incorporated median to compare two treatment regimens for prostate cancer patients. The application illustrates the easy and efficient usage of the survival-incorporated median.
The authors characterize the survival-incorporated median. They advocate that the survival-incorporated median be broadly used by researchers and practitioners in clinical studies as a useful estimand, summarizing clinical outcomes in the presence of death to inform clinical practice.
More Details