The known cost isn't really insurable. A premium based on expected costs would include the full amount of the known future costs and also some amount for less predictable costs.
Okay, now we're getting somewhere. Let's stay with the example of asthma. I pay a premium based on that condition, and the expectation that maybe one day I'll get cancer. Sure, I can understand that.
So then I get cancer, and the bill is more than I can afford. Where does the money come from? Not my premiums, because I've only been a subscriber for 3 months, and my premiums won't even cover the cost of treatment for a week of cancer treatment. If someone else isn't paying for me, and I'm not paying for me, who is paying for my treatment?
Other people do pay for the cancer treatment. Make up a pool of, say, 1000 people. Make up a cancer rate of say, 5/1000. Make up an asthma rate of say, 50/1000.
The people that are going to get cancer are unknown, but the whole group is willing to pay 0.5% of the cost of cancer treatment for a contract that covers 100% of the cost.
The 950 people that know they don't need expensive asthma treatment don't really want to pay for contracts that will cover expensive asthma treatment, so (in a pure insurance market) either the cost has to be included in the contracts for the 50 that do have it or the treatment can be excluded from the contracts.
So then I get cancer, and the bill is more than I can afford. Where does the money come from? Not my premiums, because I've only been a subscriber for 3 months, and my premiums won't even cover the cost of treatment for a week of cancer treatment. If someone else isn't paying for me, and I'm not paying for me, who is paying for my treatment?