Surveillance creep

Monitoring you set up for one small safety reason keeps growing until you are watching far more than the problem ever required.

Your dad missed his blood thinner twice in one week, so you set up a pill box that pings your phone when a dose goes unopened. It works. Then his doctor mentions falls, so you add a wearable. Then you add the door sensor, because what if he wanders. On a Tuesday night you catch yourself scrolling his step count at 11pm, and you notice he has stopped telling you about his day, because you already know it.

This is monitoring expanding past the minimum needed for coordination or safety. The tell is that you cannot say which decision a given stream feeds. Ask of each one: what would I do differently tomorrow if this number changed? Turn off anything with no answer. Set a retention window, a week or a month, so history stops piling up. Then tell him what you kept and why, and let him say no to one of them. What changes is that the pill alert keeps doing its job while the relationship stops feeling like an audit, and the data that could leak or be subpoenaed shrinks to what you actually use. It is a false alarm when the added stream is answering a real, named risk his clinician raised, and he chose it.