SQL update that seems to require GROUP BY

I have an update I'd like to make to a table, something like

UPDATE mytable
LEFT JOIN worker ON first=worker_first_name && last=worker_last_name
GROUP BY CONCAT(first, '%%', last)
SET taint = COUNT(pkey) > 1;

But of course the table_reference in UPDATE does not allow GROUP BYs. How can I set th开发者_如何学Cis column? It's just checking if the first/last name combination occurs at most once in the other table.

UPDATE mytable
INNER JOIN (
  select first, last, COUNT(pkey) > 1 Result
  FROM mytable
  LEFT JOIN worker ON first=worker_first_name AND last=worker_last_name
  GROUP BY first, last) OTHER
    on OTHER.first = mytable.first
   and OTHER.last = mytable.last
SET taint = OTHER.Result;

Something like:

UPDATE mytable
SET taint = ((select count(pkey) from worker where worker_first_name = first and worker_last_name = last) > 1)

But I'm guessing performance might leave something to be desired. There might be a more performant way.