Surviving the Curve: Tips for Handling Data Display

Abstract

Handling data displays for both new and seasoned editors can sometimes be tricky. There are standards or expectations for certain figure types. While not all editors are responsible for, or have the ability to, substantially edit figures and tables, most editors can make suggestions or pose queries to help ensure that data displays are accurate, complete, and clear. Herein is a brief primer on one of the most common figure types in the medical literature, including a short checklist that might be useful before acceptance, during figure creation, or incorporated into a journal’s instructions for authors. We’ve all seen the figures—the 2 (or more) lines snaking upward (or downward) across a plot, called Kaplan-Meier or survival curves. So, why are they called “survival” curves? In this context, survival could be literal (mortality in study participants over the course of treatment) or it could mean the occurrence of a particular outcome over time, but not necessarily death (e.g., disease recurrence, a particular symptom). The main goal of this type of data display is to show differences in survival/event occurrence between groups (e.g., between those assigned to a new agent vs. placebo in a clinical trial).1 Survival curves are often used in reporting major outcomes in studies; they are big-picture data displays. These plots are good for showing overall change over time, but not specific differences at discrete points.2 The curves themselves aren’t smooth as the term implies; they are a series of horizontal steps. In fact, some journals provide guidance to […]