They note though that the follow up population that had caught covid after testing didn't differ significantly from those that didn't. That seems to add credib

They note though that the follow up population that had caught covid after testing didn't differ significantly from those that didn't. That seems to add credibility here to a causal mechanism as opposed to a background correlation of risk of infection they are missing.