Posters · Glossary
What is a CONSORT flow diagram?
A CONSORT flow diagram is the box and arrow figure that tracks every participant in a randomised trial from screening through allocation to analysis, showing how many were lost at each step and why. It is required by the CONSORT reporting guideline, and its whole value is that the counts reconcile down the page.
A reviewer checks the flow before they read anything else, because it is the fastest way to find out whether the numbers in the results can be trusted.
Nuwan Madhusanka · Co-founder
3 min read · Published
Why the arithmetic is the point
Every arrow in the diagram is a subtraction, and a reader with a calculator can check all of them in under a minute. Screened minus excluded must equal randomised. The two arms must sum to the randomised total. Randomised minus withdrawals must equal analysed. A flow that is out by even one participant undermines every other number on the page, because it suggests nobody checked.
Fitting it on a poster
A poster flow usually gets six boxes and one aside for exclusions, which is fewer than a paper allows. The way to fit is to fold the reasons for exclusion into the aside with counts separated by commas rather than giving each its own box. Keep the main path as assessed, randomised, the two arms and analysed. If your trial has a more complex design, the honest move is to show the simplified path and put the full diagram in the paper.
The denominator that matters
The number in the analysed boxes is the denominator for every percentage in your results. If your key numbers are computed on a different base, such as everyone randomised rather than everyone analysed, say so explicitly. The commonest error on a poster is a headline percentage whose denominator does not appear anywhere in the flow.
Questions people ask
Is a CONSORT diagram required?
For a randomised trial, most journals and most conferences expect one, and the CONSORT statement asks for it. On a poster it does double duty: it satisfies the reviewer and it answers the first question a passing reader has, which is how many people this was actually about.
What if my study was not randomised?
Then CONSORT is the wrong guideline. Observational studies use STROBE, systematic reviews use PRISMA, and each has its own diagram or checklist. Using a CONSORT flow for a non randomised design implies a design you did not run, which is worse than having no diagram.
Can the boxes hold reasons as well as counts?
The exclusion aside should, since reasons are what make the numbers interpretable. The main path boxes are better kept to a label and a number, because a reader is following the arithmetic there and extra text slows that down.
What if participants moved between arms?
Say so, and be explicit about whether you analysed by intention to treat or per protocol. Crossovers are exactly the situation where a reader needs to know which denominator your results use, and a flow that hides them is misleading even when every number in it is accurate.
How much space does it need on the board?
Roughly a column, usually under methods. It is worth the space: it is one of the few blocks a reviewer will actively look for, and squeezing it to fit is how counts end up unreadable at a metre.
Make one with posters
The button opens the generator with this use case already described. Change the wording to match your own.
Create a poster with OneCraftRelated questions
- What is a PRISMA flow diagram?A PRISMA flow diagram tracks records through a systematic review: identified, screened, assessed, included, with exclusions and reasons at each stage.
- What is a research poster?A research poster is one board that carries a whole study: abstract, methods, results and conclusion, read standing up in about three minutes.
- What DPI does a poster need?A poster is read from a metre away, so about 100 pixels per inch at full size is enough. Why 300 DPI is a photographic habit that does not apply.
Written and checked by the OneCraft team. Last checked .