| > You can't compile a list of facts about a person's life and use that to deterministicly attribute the cause of given conditions. That is just not the case, and I shouldn't have to point out such basics on HN. > A physical exam at the start of insurance doesn't solve this, because like I said, the existing issues could have been caused by any number of problems. And yet I have to… People do a thing called "collecting data", on a large scale, and then they apply the lessons learnt from that data to calculate statistical risk. An imperfect system but, strangely enough, very effective (when not interfered with, as in the US health system). Of course, you are welcome to open a car insurance company and offer everyone the same insurance for the same price and watch as young men and previously uninsurable people flock to your service. Maybe you'll get lucky and won't have to pay out more than you take in. All the best with that. > Your example also has the problem of measuring the "lifestyle questions" being presented. How would you prove a person isn't exercising enough? If I know it will get me denied I'm not going to self report. And then the insurance company can decide what level of fraud it will tolerate (something HN has discussed previously[0], and the linked article[1] is enlightening) and thus adjust its costs, and perhaps premiums. > Your dismissal of my specific example is silly - I don't want to sue a doctor for misdiagnosing a relatively common issue. If it's not a problem then don't sue. If it is, that is what the court system is for (or whatever system doctors and medical companies would put in place to avoid going to court). [0] https://news.ycombinator.com/item?id=38905889 [1] https://www.bitsaboutmoney.com/archive/optimal-amount-of-fra... |