How would you build a traceability map linking product claims to studies and submission sections?

Why

A label claim such as 'reduced the risk of hospitalization by 30 percent' must be traceable to a pre-specified endpoint in a specific trial, an analysis in the statistical analysis plan, a table in the clinical study report in Module 5, and summaries in Module 2. When teams write sections separately, numbers drift: the clinical overview rounds differently from the study report, or a summary cites a sensitivity analysis instead of the primary one. A traceability map is a table with one row per claim and columns for the supporting study, endpoint, estimand, analysis population, result, CTD location and dataset.

It also records the strength of support: primary endpoint with multiplicity control, key secondary within the testing hierarchy, or exploratory only. Exploratory results cannot carry efficacy claims. The map also helps when the label is negotiated, because each change can be checked against evidence fast.

And it ties into Define-XML and analysis datasets, so reviewers can reproduce numbers.

How to work it through

1. List every proposed claim from the draft label and target product profile.
2. For each claim, identify the study, endpoint, analysis and population that support it.
3. Record the exact CTD section, table number and dataset location for each result.
4. Grade the support as confirmatory, supportive or exploratory.
5. Check that every number matches across label, Module 2 and Module 5.
6. Keep the map under version control and update it with each label negotiation.

Where the answer stops holding

The map is only as accurate as the people who maintain it, and late analyses often bypass it. It shows where evidence is, not whether a regulator will accept it. For large programs with many studies, a spreadsheet becomes hard to manage and a structured tool may be needed.