How to Create a STROBE Participant Flow Diagram for an Observational Study

A participant flow diagram can make the selection and retention of participants in an observational study much easier to understand. It can show how the study population was identified, who was eligible, who was excluded, which participants entered the study, how many completed follow-up, and how many contributed to the final analysis.

Unlike CONSORT for randomized trials, STROBE does not prescribe one universal flow-diagram template. Instead, STROBE item 13 asks authors to report the number of individuals at each relevant stage of the study, give reasons for non-participation at each stage, and consider using a flow diagram.

Create a STROBE-supportive participant flow diagram with the Blazon Scientific Study Flow Diagram Builder. The tool supports cohort, case-control and cross-sectional studies and lets you adapt labels to the way participants or records actually moved through your study.

What does STROBE item 13 require?

STROBE item 13 concerns the reporting of participants in the Results section. Authors should report the number of individuals at relevant stages of the study. Examples given by STROBE include people who were:

• potentially eligible;
• examined or assessed for eligibility;
• confirmed eligible;
• included in the study;
• completing follow-up; and
• analysed.

Authors should also give reasons for non-participation at each stage and consider presenting this information in a flow diagram.

There is no single “STROBE flow diagram”

This is an important difference between STROBE and reporting guidance for many randomized trials. Observational studies can have very different sampling and data-collection structures, so a diagram should reflect the actual study design rather than forcing every study into the same sequence of boxes.

A retrospective database study, for example, may begin with records rather than recruited participants. A prospective cohort study may need to show follow-up losses. A case-control study may need separate pathways for identifying cases and selecting controls.

Flow diagrams for cohort studies

A cohort-study diagram often begins with a source population or potentially eligible population, followed by eligibility assessment, exclusions and the final cohort. If there are exposure groups, these may then be shown in separate branches.

For a prospective cohort, later stages can show the number completing follow-up and the number included in the analysis. STROBE also asks cohort-study reports to describe methods of follow-up and, when applicable, explain how loss to follow-up was addressed.

A typical structure might be:

1. potentially eligible population;
2. assessed for eligibility;
3. excluded, with reasons;
4. included in the cohort;
5. exposure or comparison groups, if relevant;
6. completed follow-up; and
7. analysed.

Flow diagrams for case-control studies

Case-control studies often benefit from a diagram that makes the sources and selection of cases and controls explicit. STROBE asks authors to give the eligibility criteria and the sources and methods of case ascertainment and control selection. The rationale for the choice of cases and controls should also be reported.

A useful diagram may therefore show separate pathways for:

• potential cases;
• case eligibility and exclusions;
• included cases;
• the source population for controls;
• control eligibility and exclusions; and
• included controls and final analysed samples.

For matched case-control studies, the manuscript should also report the matching criteria and the number of controls per case where applicable. The flow diagram can help clarify these numbers, although not every methodological detail needs to be placed inside the figure.

Flow diagrams for cross-sectional studies

For cross-sectional studies, a diagram may show the sampling frame, selected or invited individuals, eligibility assessment, exclusions or non-response, final participants, and the sample included in analysis.

When complex sampling is used, the diagram should complement—not replace—the Methods section, where the sampling strategy and relevant analytical methods should be described.

What exclusion reasons should be shown?

Reasons should be specific enough to help readers understand how the analytical population was formed. Depending on the study, relevant reasons might include:

• not meeting eligibility criteria;
• declining participation or non-response;
• duplicate or ineligible records;
• missing essential exposure, outcome or covariate information;
• absence of required follow-up;
• withdrawal; or
• other prespecified exclusions.

Be careful when describing exclusions caused by missing data. STROBE separately asks authors to indicate the number of participants with missing data for variables of interest and to explain how missing data were addressed analytically.

How to check an observational-study flow diagram

Before publication, check that:

• the diagram follows the actual sampling and inclusion process used in the study;
• numbers at successive stages reconcile where they are expected to do so;
• reasons for non-participation or exclusion are reported clearly;
• the final analysed sample agrees with the number reported in the Results and tables;
• loss to follow-up is distinguished from exclusions for other reasons; and
• the figure does not imply a prospective recruitment process if the study actually used retrospective records or routinely collected data.

The Blazon Scientific Study Flow Diagram Builder includes live count checks while keeping the labels editable so the diagram can reflect the actual design rather than a rigid template.

How to create a STROBE-supportive participant flow diagram

1. Identify the true starting population, sampling frame or record set.
2. List each meaningful eligibility, inclusion or selection stage.
3. Record the number of individuals or records at each stage.
4. Record reasons for non-participation and exclusion.
5. Add follow-up and analysis stages when they are relevant to the design.
6. Check that the diagram is consistent with the Methods, Results and tables.
7. Review the complete STROBE checklist for the applicable study design.

Ready to build the figure? Open the CONSORT & STROBE Study Flow Diagram Builder.

References and further guidance

• STROBE checklists for observational studies.
• Combined STROBE checklist for cohort, case-control and cross-sectional studies.

A participant flow diagram supports transparent reporting but is only one part of STROBE. Authors should use the full checklist appropriate to their study design.