About this example
Longitudinal clinical studies often report how many patients improve, stay the same or worsen between visits. A table of transition counts is hard to read; an alluvial diagram shows the same numbers as flows, so readers can follow each baseline group through follow-up.
Coloring the flows by baseline severity keeps each group traceable across stages. Showing dropouts as their own category makes attrition visible instead of silently removing those patients, which reviewers usually ask for.
The data
| Baseline | Month 6 | Month 12 | Patients |
|---|---|---|---|
| Severe | Severe | Severe | 18 |
| Severe | Moderate | Moderate | 22 |
| Severe | Moderate | Mild | 14 |
| Severe | Mild | Remission | 6 |
| Severe | Dropped out | Dropped out | 9 |
| Moderate | Moderate | Moderate | 20 |
| Moderate | Mild | Mild | 31 |
| Moderate | Mild | Remission | 24 |
| Moderate | Severe | Severe | 7 |
| Moderate | Dropped out | Dropped out | 8 |
| Mild | Mild | Mild | 19 |
| Mild | Remission | Remission | 36 |
| Mild | Moderate | Mild | 5 |
| Mild | Dropped out | Dropped out | 4 |
Open the example, replace the data with your own, and export the figure as SVG, PDF or a 600 dpi PNG.