Provided a larger sample size, doesn’t an increasing rate of an outcome suggest existence a confounding variable that the smaller previously tested sample faile

Provided a larger sample size, doesn’t an increasing rate of an outcome suggest existence a confounding variable that the smaller previously tested sample failed to identify, and (not absolutely- but more likely than not) would apply increasingly at the larger scale? Not a statistician here- and honestly asking.