Paperonium

Covid-19 · BMJ · 2022

The summary: 98.0% regained taste, yet its likely range stops at 95.5%. The supplement’s reaches 99.5%.

The summary and resultsends below its own best guessThe paper’s supplement tableholds the best guess98.0%92.2 to 95.5%should be inside99.5%92.2%92%94%96%98%100%
The summaryand resultsends below itsown best guessThe paper’ssupplement tableholds thebest guess98.0%92.2 to 95.5%should be inside99.5%92.2%92%96%100%

Share who regained lost taste by 180 days: best guess and likely range

The paper says

The paper puts together 18 studies of covid-19 patients who lost smell or taste. Its summary and results say 98.0% of those who lost taste had it back by 180 days, with a range of likely values of 92.2% to 95.5%.

Its own data show

That range stops below the 98.0% it belongs to. The paper’s supplement (extra material published with it) prints the same result as 98.0%, with a range of 92.2% to 99.5%, which does contain it.

A range of likely values is built around the best guess, so it must contain it. The 98.0% and the paper’s conclusions stand, but the summary’s range is impossible, and a later review that reuses it to combine studies would get it wrong. A 2022 correction to the paper left it unchanged.

“The summary” means the paper’s abstract: the short summary printed at its top, which is the part most readers see.

Read the paper →Full technical write-up →

Prognosis and persistence of smell and taste dysfunction in patients with covid-19: meta-analysis with parametric cure modelling of recovery curves · BMJ (2022) · doi:10.1136/bmj-2021-069503