The second paragraph of the article somewhat misrepresents the study.
There wasn't a group that chose to opt out, and another group that chose not to. Everyone agreed to be in the study, and then a random half of the cohort was removed from the mailing lists.