3 ms·
But we desperately need an honest, transparent, review of coding versus reality. There was a (lancet?) submission reviewing autopsies, but that was rejected. I
by dent9876543 3y ago
But we desperately need an honest, transparent, review of coding versus reality.
There was a (lancet?) submission reviewing autopsies, but that was rejected. It came to some pretty stark conclusions, so it was either questionable outright, or perhaps getting too close to the truth. I hope the former, but I can’t help but wonder if it might be the latter.
At present, data is too opaque. And it’s not the kind of data that can be shared openly — too personally identifying, and too private. So I can see why it’s not happened. But at the same time, there’s too much politicisation to take the current figures as a reliable indicator. So something needs to happen.
- fastaguy88 3y agoIt is not clear to me how desperate the need is. It is not hard to find examples of things that, at first glance, seem clearly nonsensical. But some examples may be miscoded and corrected later, while other examples are simply misunderstandings of the common use of the code. This seems to be the primary strategy for election deniers. But even if a tree is actually a large bush, forests still exist. We can predict with reasonable accuracy how many people will die of the most common causes, because the numbers are big. That is the strength of excess death estimates. I do not think there is much disagreement over most of these statistics by people who understand the estimation process.