2 ms·
The methodology in this study seems really sketchy to me. So we start with one indicator of bone cancer that is well-established: bone lesions. This indicator
by xg15 6y ago
The methodology in this study seems really sketchy to me.
So we start with one indicator of bone cancer that is well-established: bone lesions. This indicator shows a significant rise from medieval to current times.
Now we use another, less established indicator (ct scan results) and find... a number consistent with the first result.
Finally, we make some "extrapolations" about the rate of undetectable cancer based on our observations and lots of assumptions.
In the end, we show that our extrapolated(!) cancer rate is higher than expected and does not show as much of a difference between medieval and modern times.
None of this explained the significant rise in the first indicator. Also, even if you liberally pad your data with extrapolations, wouldn't you have to apply the same formula to the indicators of cancer in modern-day individuals?
By the way, the modern-day cancer rate is still ~3x higher than even the extrapolated rate (15% vs 50%) - yet, the article goes one step further than the study and frames this as a finding that the cancer rate didn't grow at all.