Can a party function without a Cleric or Paladin? was asked in 2014. It has answers from 2015 and 2016, whereby the latter answers adds nothing fundamental to already given ones.
Another example: Does a reach weapon allow you to threaten squares 10 feet away or not? - asked in 2014, has been recently answered one more time.
It it OK by the SE standards, or should I flag such answers?
It's cool to add good answers to old questions. We even have badges for it! And answers are always held to the same basic standards no matter how old the question is.
But if what you're focused on is answers which mostly re-state existing answers, sometimes a different way of explaining something is useful--everybody's got a slightly different way of looking at things and multiple perspectives can be helpful. If you think the information isn't presented in a usefully new way, please vote accordingly: voting is how the useful stuff floats to the top and the less-useful stuff settles to the bottom.
For that specific question, I suspect the fundamental issue is that the accepted answer from 2014 is openly speculative, and basically says "I think the answer is 'no', but we'll need to wait and see." The currently second highest voted answer, posted a month and a half later, also pretty much says the same thing.
In the mean time, a lot of people have had time to playtest this since 5e was published, and many seem to have come to the personal conclusion that the answer is actually "yes". However, since the top answers still don't reflect this accumulated knowledge, several people have posted their own anecdotal answers (which are OK here, if based on actual play experience) to supply contrasting evidence.
Honestly, I'd say the system is working as designed here. Which is not to say that it's working ideally, but this is what's supposed to happen in the SE system, if people disagree with the accepted and/or top-voted answer(s). I would expect the influx of new answers to taper out once either:
the author of the accepted answer edits their answer to incorporate the more recent alternative viewpoints, or
one of the contrasting answers gets enough upvotes to show up at second place in the default sorting order.
Note that, currently, the answers at third, fourth and fifth place all express variations of "yes, a 5e party can function without a dedicated healer," mostly backed by specific play experience. That includes the most recently posted answer (out of the ones currently visible, anyway; it's possible that there have been more recent answers that have been deleted) that has risen to fourth place in less than five months (pretty well for such an old question).
Anyway, as for what you can and should do, BESW's answer pretty much says it all: upvote the good answers that you agree with, downvote those that you think are clearly wrong, unreadable, useless or superfluous. Over time, the best answers will float to the top, and the bad ones will sink to the bottom. If you wanted to speed up the process, you could always consider awarding a bounty to one of the answers that you like, which would have the side effect of increasing the question's visibility while the bounty is active, and thus attracting more voters to it.
What you should not need to do is flag any of the answers, as long as they're on-topic at this site (and not, say, spam or rude or questions mistakenly posted as answers).
