STROBE Flow Diagram Generator for Observational Studies
Free STROBE flow diagram generator for cohort, case-control and cross-sectional studies. Enter counts and reasons, check they add up, export SVG or PNG.
Pick cohort, case-control or cross-sectional, enter your counts: the flow is drawn exactly as SVG or PNG, free
Start from an example (illustrative numbers)
Cohort flow: from potentially eligible to analysed, with a follow-up stage.
STROBE Flow Diagram Generator
Free to try ·
Your AI STROBE-style diagram will appear here
AI drawings can misplace or misspell numbers. For exact counts, use the Precise Diagram tab instead
STROBE Flow Diagram Examples
Exact renders from the generator for cohort, case-control and cross-sectional designs, all with illustrative numbers
Simple cohort flow
Exact render from the tool: six stages, one exclusion reason per stage. Illustrative numbers.
Two exposure groups
Exact render with the exposure split on from inclusion: each group has its own exclusions, and group counts add up to the stage totals. Illustrative numbers.
Multiple exclusion reasons
Exact render with 2 to 4 reasons for non-participation per stage (item 13b). Illustrative numbers.
Matched case-control study
Exact render with the case-control layout: 420 cases and 900 controls followed separately to 310 and 620 analysed, controls matched 1:2 on age and sex. Illustrative numbers.
Cross-sectional survey by exposure
Exact render with the cross-sectional layout: no follow-up stage, and the analysed stage split into two exposure groups. Illustrative numbers.
What is a STROBE flow diagram?
STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) is a reporting guideline for cohort, case-control and cross-sectional studies. It has a 22-item checklist. Item 13 is about participants: 13a asks for the numbers of individuals at each stage of the study, 13b asks for the reasons for non-participation at each stage, and 13c says "Consider use of a flow diagram". A STROBE flow diagram is that figure: a chart that follows people from the pool of potentially eligible individuals down to the group you analysed, with the people who dropped out and the reasons shown at each step.
What this generator draws
Think of it as a STROBE diagram template you fill in with your own counts. Pick the study design first: cohort, case-control or cross-sectional. Each layout reports the numbers and reasons that STROBE item 13 asks for, with a side box of exclusion reasons and counts between stages (13b). You can edit every stage label and add or remove reasons at each stage. A cohort flow can be split into two exposure groups from inclusion onward, a cross-sectional flow can split the analysed stage by exposure group, and a case-control flow always runs cases and controls in two separate columns, with an optional matching note.
How cohort, case-control and cross-sectional flow diagrams differ
- Cohort study: one flow from potentially eligible, examined for eligibility and confirmed eligible to included in the study, completing follow-up and analysed. The follow-up box counts people still observed at the end, and the exclusion box before it lists losses to follow-up. If you compare exposure groups, the counts can be given per group from inclusion onward.
- Case-control study: cases and controls are selected separately (STROBE item 6a asks for the sources and methods of case ascertainment and control selection), so the diagram has two columns. Each column counts its own people: cases ascertained or controls selected, confirmed eligible, included in the study (for example, interviewed) and analysed, with its own reasons for non-participation. There is no follow-up box. The diagram draws one group of controls; if you used more than one control source, name the source in the column heading or in the exclusion reasons, or draw a separate diagram for each. The STROBE explanation paper advises describing the flow of participants separately for case and control groups, and item 6b asks matched studies to give the matching criteria and the number of controls per case, which is what the matching note is for.
- Cross-sectional study: everyone is assessed at one point in time, so there is no follow-up box. The flow usually runs from people sampled or invited, to confirmed eligible, to those who took part (responded), to those analysed, with non-response and its reasons at each step. The STROBE cross-sectional checklist asks for item 13 information separately for exposed and unexposed groups where that applies, so the analysed stage can be split into two groups. The split starts at the analysed stage because exposure is usually known only for people who took part.
How to make a STROBE flow diagram
- Choose the study design at the top of the tool: Cohort, Case-control or Cross-sectional. Each design keeps its own numbers, so you can switch without losing your edits.
- Pick an example to start from, or edit the default. All example numbers are illustrative and the image says so unless you switch that label off.
- Type the count for each stage and rename the stages to match your own wording. For a case-control study, fill in the cases column and the controls column, and add a matching note if controls were matched to cases.
- Under each stage, list the reasons people were excluded before it and how many for each reason. Add or remove reason rows as needed.
- Optional, cohort and cross-sectional: turn on the exposure split. In a cohort it starts at "Included in the study"; in a cross-sectional study it splits the analysed stage. The tool prefills a rough half-and-half split that you can overwrite.
- Fix any red messages. The diagram is only drawn when every stage equals the previous stage minus its exclusions, so a misleading figure is never exported.
- Download SVG for a vector figure you can edit, or PNG for slides and documents. Both are free, with no sign-up.
How the numbers are checked
Counts must be whole numbers of zero or more. Each stage must equal the previous stage minus that stage’s exclusions, and the message shows the sum, for example "Included in the study (820) should be 850 − 30 = 820". In a case-control diagram the same check runs separately in the cases column and the controls column, and each message names the column. With the exposure split on, the two group counts at the split stage must add up to the previous stage minus the exclusions before it. After that, each group count must equal that group’s previous count minus that group’s own exclusions, and can never be larger than before. The tool checks arithmetic only. It cannot know whether your eligibility criteria or reasons are right, so check the final figure against your data and the full STROBE checklist.
STROBE vs CONSORT vs PRISMA: which flow diagram do I need?
CONSORT is the reporting guideline for randomised trials, and its flow diagram follows participants through enrolment, allocation, follow-up and analysis. STROBE covers observational studies (cohort, case-control and cross-sectional designs), where nobody is randomised, so the diagram follows people through eligibility, inclusion, follow-up and analysis instead. PRISMA is for systematic reviews and meta-analyses, and its diagram follows records and studies, not people. Pick the guideline that matches your study design.
Is a flow diagram required by STROBE?
No. In the STROBE explanation and elaboration paper, item 13c reads "Consider use of a flow diagram", and the authors write: "While we recommend the use of a flow diagram, particularly for complex observational studies, we do not propose a specific format for the diagram." It is a recommendation, and what a journal asks for is set by that journal, so check its instructions for authors. A diagram from this tool reports the numbers and reasons that STROBE item 13 asks for, but it does not make a paper compliant with the whole guideline.
Frequently Asked Questions
Related flow diagrams
Reporting a randomised trial? Use the CONSORT diagram generator. Reporting a systematic review or meta-analysis? Use the PRISMA flow diagram generator. To lay out the causal assumptions behind an observational study, try the causal diagram (DAG) maker.
Sources
Checked October 2026.
- Vandenbroucke JP, von Elm E, Altman DG, et al. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): Explanation and Elaboration. PLoS Med 2007;4(10):e297. Items 6a, 6b, 13a, 13b and 13c (including the advice to describe the flow of participants separately for case and control groups) and the 22-item checklist. journals.plos.org
- STROBE checklists for case-control studies and for cross-sectional studies: item 6 (case ascertainment and control selection; matching criteria and number of controls per case), item 13 (13a, 13b, 13c), and the note that starred items are given separately for cases and controls, or for exposed and unexposed groups. strobe-statement.org/checklists
- EQUATOR Network, STROBE reporting guideline page: designs covered (cohort, case-control and cross-sectional studies), checklists and the original statement by von Elm et al., 2007. equator-network.org
- STROBE Statement website (the guideline’s home, linked from the explanation and elaboration paper). strobe-statement.org
Related Research Tools
ResearchCONSORT Diagram Generator
Draw the participant flow of a randomised controlled trial, the counterpart of STROBE for trials.
ResearchPRISMA Flow Diagram Generator
Draw the study-selection flow for a systematic review or meta-analysis.
ResearchCausal Diagram (DAG) Maker
Draw the causal assumptions behind an observational study, with exposure, outcome and confounders.