My own research on this awhile back lead me to believe the risk of transfer is tiny, given 2 basic things.
The first has already been discussed (good application practices). Using areas where transfer is unlikely but without also jeopardising efficacy for yourself, is a good start. Just being mindful and not reckless, really.
Secondly, even if transfer should occur, the risk of uptake is remote. Theres a few factors which lead to this conclusion.
Transdermal delivery is notorious for its relatively low penetration. Suppose we are generous and grant that, for any single application, with the very best delivery system ("carrier"), 50% of the applied active makes it through the skin layers and into the blood.
Keep in mind that our skin serves as a protective barrier.
Lets be further generous and say that, any active which didnt get through the skin and into the blood is still sitting on the surface of your skin. So, 50% of the initial dose.
That 50% sitting on your skin is not getting through your skin without any further chemical help. Remember, the initial carrier has now evaporated and itself dispersed. Youd have to add further applications of DMSO/hand-sanitiser/etc to try and elicit further absorption.
Essentially, for simplicity sake, you have "raw" active powder residue on your skin. If some of this residue was to make it onto the skin of someone else, what would the likelihood be of that residue, with no carrier, getting through their protective skin layers?
One study suggests "zilch". Logic suggests zilch. Imagine rubbing 50mg of steroid raw across your forearm with your finger. If a company sold you powder and presented this application technique what would you think?
Given the dermal related facts, and the good application practices already discussed, transfer and uptake risks are tiny.