Ok, so this is a design decision problem. We can now go one of two ways 1. Leave it as it is. Tell patients * really loudly* it to put magnets near their ch

Ok, so this is a design decision problem. We can now go one of two ways 1. Leave it as it is. Tell patients * really loudly* it to put magnets near their chests. 2. Redesign the reed switch to use some RFID style interface so that only someone with correct password can adjust the device. My take is 1. 2. Opens up a whole world of failure possibilities that are not easy to know or predict - and every day a zero-day is possible. But 1. remains simple to understand and limited in the ways it can go wrong.