This is exactly the problem. Despite spending what is in my many cases billions of dollars a year on IT and related underwriting systems, most of the major health insurers in the US still have some manual touch involved in their pricing, which means that they can't price in anything near real-time. Even the best can do perhaps half in a fully automatic way, yet generally choose not to do it in realtime (for reasons I'll never fully grasp).
One major insurer that I'm familiar with receives applications from the biggest of these sites by fax, on paper. They're then shipped to a low cost country, where they're scanned, and then passed along to another low cost country to have the OCR corrected. Then the work comes back on-shore to be manually adjudicated. Figure 3 days minimum, a week on average to get all of this done.
A crowd-sourced solution would be astronomically useful. For it to be accurate, you need to capture a decent amount of medical information (most insurers ask a pretty standard set of ~20 questions), but even knowing that the average person pays 1.4x the quoted rate for someone of their age/gender/zip would be a great tool.
One major insurer that I'm familiar with receives applications from the biggest of these sites by fax, on paper. They're then shipped to a low cost country, where they're scanned, and then passed along to another low cost country to have the OCR corrected. Then the work comes back on-shore to be manually adjudicated. Figure 3 days minimum, a week on average to get all of this done.
A crowd-sourced solution would be astronomically useful. For it to be accurate, you need to capture a decent amount of medical information (most insurers ask a pretty standard set of ~20 questions), but even knowing that the average person pays 1.4x the quoted rate for someone of their age/gender/zip would be a great tool.