I suspect the reason children are the hold up here is less around current data than around where we are in this pandemic and patterns of thought/risk assessment

I suspect the reason children are the hold up here is less around current data than around where we are in this pandemic and patterns of thought/risk assessment. Here's an analogy that maybe will work with the HN crowd. Say we've got a service that, for 75% of the API endpoints, often gets a severe error. 25% of the endpoints only very rarely get the same error. My team discovers that by applying one simple change to the 75% of the affected endpoints, their error rate drops dramatically. We can't, for whatever reason, apply the same fix to those 25% yet, but we're confident that within a relatively short amount of time we'll be able to. Do we declare that the service is now working well and move on to other items? Maybe -- in a real setting it'll of course depend on what other priorities the team's juggling at the moment. Maybe this is "good enough" to move on. Still, I think we can all relate to the emotional pull to say "hold on, we still don't actually know why those 25% of our endpoints weren't affected as often. Something could easily change there so that we start seeing similar error rates. We have a fix on the way for this -- can we wait just a bit longer so we can roll this out?" I think that's where a lot of people are. Again, one could certainly argue that we're close enough, but OTOH I think the reluctance to declare this "over" yet makes a lot of sense too.