Confounding would be like if you were testing out a new chemo drug vs the gold standard, and found that the new chemo drug did better than the gold standard BUT the mean age of the group given the new drug is lower than that of the gold standard group. Thus AGE would be a confounding factor, ie younger people tend to do better with chemo than older and the observed effect may not be due to the drug. In our case, maybe younger people tended to be given the new drug whereas older people were more often given the gold standard. Remember that a confounder is associated with the exposure AND the outcome, and is not in the causal pathway.
An effect modification is just that -> a change in the outcome due to the presence or absence of some factor. Think of it as synergy, i.e. the size of the effect changes according to the level of another factor, such as sex or age.
HTH
