Shame it's behind a paywall. Obviously directly derived works or studies that replicate results are probably suitable for withdrawl too under the same criteria

Shame it's behind a paywall. Obviously directly derived works or studies that replicate results are probably suitable for withdrawl too under the same criteria. I wonder how many published papers on average should be withdrawn when events such as this happen. I suppose its related to impact assessments but I wonder how that varies from field to field.