Infusion Journal is supported by:
Get notified when a new issue of Infusion Journal has been published.
The Art of Data
Understanding Violin Plots in Medical Research
Michelle Simpson, PharmD, BCSCP, MWC | Editor-in-Chief, Infusion Journal
In medical research, clear and effective data visualization is essential. Figures and graphs highlight the most important findings in publications, presentations, and posters. Readers rely on visual representations to quickly grasp trends and make comparisons. Traditional box plots and histogram bar graphs have long been the standard for summarizing data, but journals are adopting visualization policies that encourage authors to combine the capabilities of both and use graphics that depict the data in figures providing as much granularity as possible. One increasingly common yet often misunderstood tool is the violin plot.1 What makes it special is that it shows the full distribution of the data in addition to the summary data. In this issue, the research article titled “Preference for Inebilizumab Home Infusion Among People with Neuromyelitis Optica Spectrum Disorder” incorporated violin plots to illustrate changes in health survey scores over three time points. To understand what is being represented, Infusion Journal’s readers need to know how to interpret these figures effectively in order to comprehend the results of the research.
What Is a Violin Plot?
A violin plot is a hybrid of a box plot and a density plot providing a richer understanding of data distributions. The width of the “violin” at any given point reflects the density of data at that value, offering a more complete picture for quickly approximating where the data is centered and how it is spread. Similar to a bar graph, a violin plot shows the shape of the dataset to help visualize the distribution of the data. It is a density estimate rotated 90 degrees and then mirrored, making it symmetric and appearing as a smoothed-over histogram built around center lines rather than stacked on baselines (Figure 1). Other than this difference in display pattern, curves in a violin plot follow the same construction and interpretation. Peaks, valleys, and tails of each group’s density can be compared to see where groups are similar or different. The width represents higher and lower probability of the point density. Datasets containing more than one peak distribution are what give this type of graph its distinctive violin shape and name.
When to Use a Violin Plot
Not every dataset benefits from a violin plot, but there are specific scenarios where this visualization tool shines:
1. When Distribution Matters – If the spread of data points is as important as the median or mean, a violin plot provides a more informative view than a simple box plot.
2. When Comparing Multiple Groups – Side-by-side violin plots are effective in visualizing how distributions differ across categories, such as patient populations or treatment groups.
3. When Sample Sizes Are Large – With enough data points, violin plots provide a smooth estimation of density, making trends clearer.
4. When Outliers and Variability Need Emphasis – Unlike bar graphs or simple line plots, violin plots do not obscure underlying variability, ensuring that extreme values and clustering effects remain visible.
Violin Plots and Research Posters: A Visual Advantage
The ability to convey complex data efficiently is not only critical for journal publications but also for research abstracts and posters. In academic conferences and scientific presentations, researchers are challenged to distill extensive findings into digestible, visually compelling figures. Choosing the right type of graph can significantly impact how audiences engage with the data.
Violin plots are particularly valuable for posters where comparisons across multiple groups or time points are key takeaways. They allow viewers to quickly see not just whether there is a statistical difference but how distributions compare in shape and spread. For example, in a clinical study assessing patient-reported outcomes at baseline, mid-treatment, and post-treatment, a violin plot can reveal whether most patients improved steadily or whether certain subgroups responded differently. This level of insight is often lost in simple bar charts or line graphs that focus only on averages.
Moreover, in poster presentations where space is limited, violin plots consolidate multiple layers of information into a single, intuitive figure. A well-constructed violin plot can eliminate the need for separate histograms or density plots, making the visual narrative of the study both efficient and engaging.
Expanding the Use of Violin Plots in Medical Research
Despite their advantages, violin plots remain underutilized in medical literature. Some researchers hesitate to use them due to their relative unfamiliarity, while others default to more traditional graphs out of comfort and experience. However, as medical research continues to embrace more sophisticated statistical methods and larger datasets, it is imperative to adopt visualization tools that match this complexity.
Violin plots are particularly well-suited for patient-reported outcomes, biomarker distributions, and longitudinal studies, where understanding variability is crucial. Their ability to reveal underlying patterns makes them invaluable in fields such as epidemiology, clinical trials, and health economics.
Data visualization is not merely a technical choice—it is a communication strategy. As researchers, we must select figures that best serve our audience’s understanding. The violin plot, while less common than bar graphs or box plots, offers an unparalleled ability to display both summary statistics and distributional nuances. As seen in this issue, this tool can enhance the interpretation of studies tracking change over time. Furthermore, incorporating violin plots into research posters can provide a competitive edge in scientific communication, allowing for richer storytelling with data.
By embracing violin plots where appropriate, we enhance not only the clarity of individual studies but also the broader dialogue in medical research. As visualization techniques evolve, so too should our approach to data publication, ensuring that our findings are both rigorous and accessible to diverse audiences.
If you have a patient case or idea for writing a case report or questions about submitting a manuscript to Infusion Journal, contact: infusionjournal@nhia.org.
Learn more about the journal and review information on manuscript submission.
1. Hintze JL, Nelson RD (1998) Violin Plots: A Box Plot-Density Trace Synergism. The American Statistician 52:181-184.
