Use case
A clinician can see exactly how a study defined its outcome
A clinician deciding whether a study applies to their patients needs the exact definitions behind its conclusion: what counted as treatment response, which patients were left out, which thresholds were used. A companion repository shows them in the code that ran, with the data dictionary, and later releases show whether the finding held on new patients.
Clinical conclusions turn on definitions a paper summarizes in a sentence
Claim: hypothesis
Clinical conclusions turn on definitions that a paper summarizes in a sentence. Time in range for glucose, for example, depends on the range, the sensor, the minimum data required and how gaps are handled. Seeing the definition as code, linked to a data dictionary and a standard vocabulary, tells a clinician whether the study's patients look like theirs.
What would show this claim is wrong
- Clinicians shown the code definitions make the same decisions about applying a study as clinicians shown the methods section.
Guideline groups read definitions straight from the code
Claim: hypothesis
Guideline groups and clinical decision support teams could read definitions straight from companion repositories when they weigh studies, and cite the release they checked. Health systems and professional societies would pay for the reading time, the same way they pay for evidence review now.
What would show this claim is wrong
- Guideline groups don't use companion repositories even when every study they weigh has one.
Compare one study's outcome definition with your own
Claim: hypothesis
Take one study that changed your practice and has a companion repository. Find its outcome definition in the code and compare it with how you'd define the same outcome for your patients.
What would show this claim is wrong
- The code is too hard for a clinician to read, even with a data dictionary, so the definition stays hidden.
Related
- Companion Executable Code Repositories
- Versioned releases keep the evidence current while the paper stays fixed
- A result anyone can rerun counts for more than a famous name