The problem with trials registries is that they're only one piece of the solution, but many more things need to work for them to be effective. Many trials are

The problem with trials registries is that they're only one piece of the solution, but many more things need to work for them to be effective. Many trials are registered, required by law to be published, but still not published, because there is no enforcement: https://fdaaa.trialstracker.net/ https://fdaaa.trialstracker.net/ https://eu.trialstracker.net/ https://eu.trialstracker.net/ Or trials alre registered, but then published with a different main result than the one that was registered as the main result (so-called outcome switching). This should be noted in peer review, but often it's not. And if you point it out to the journal later they should publish a correction, but they don't.