Epidemiology

Alluvial diagram example: participant-level data without counts

Example alluvial diagram built from one row per participant (no count column): smoking status at baseline, year 1 and year 2.

Open this example in the editorStart from scratch

Current (23)Former (12)Never (15)Current (14)Current (14)Former (21)Former (21)Never (15)Never (15)Current (17)Former (18)Never (15)BaselineYear 1Year 2
Smoking status of 50 participants at three visits, counted from participant-level rows. Illustrative data.

About this example

You don’t need to aggregate your data first. When each row is one participant, set the value column to “None” and the diagram counts the rows for you.

This works well for cohort data exported straight from a spreadsheet or statistics package, and it avoids mistakes from counting transitions by hand.

The data

BaselineYear 1Year 2
CurrentCurrentCurrent
CurrentCurrentCurrent
CurrentCurrentCurrent
CurrentCurrentCurrent
CurrentCurrentCurrent
CurrentCurrentCurrent
CurrentCurrentCurrent
CurrentCurrentCurrent
CurrentCurrentCurrent
CurrentCurrentCurrent
CurrentCurrentCurrent
CurrentCurrentCurrent
CurrentFormerFormer
CurrentFormerFormer
CurrentFormerFormer

…and 35 more rows. Open the example to see all of them.

Open the example, replace the data with your own, and export the figure as SVG, PDF or a 600 dpi PNG.

More examples