If many studies ignore the same cofounder, it doesn't matter that they are consistent because they are all biased in the same way. Consistency doesn't prove anything on its own it needs to be combined with other criteria.