We see the flooding issue right now--an evacuation would have killed far more people.
A dirty bomb? Once the bomb goes off you will have plenty of time to evacuate where you need to. Keeping radioactivity contained would be more problematic than removing people.
A biological agent? You want to quarantine rather than evacuate.
A fire? San Diego showed how to deal with that--phased evacuations ahead of the actual fire path.
Earthquake? Well, once it's done you have plenty of time to evacuate.
I can't really think of anything that would affect something the size of Houston simultaneously.
Because the cost to not be in it is enormous.
[1] https://www.cms.gov/research-statistics-data-and-systems/sta...
Enormous compared to the value provided.
> We spend 3.2 trillion a year on Medicare
No, we don't. As your own source says, it is only $646 billion on Medicare, and $3.2 trillion a year in total combined public and private healthcare expenditures.
> Imagine if just 1K per person per year (300B) was spent on infrastructure every year to make mass evacuations over a few days possible.
Then it still probably wouldn't be enough, and we'd get a lot more value for nearly doubling infrastructure spending [0] if we spent it on something else.
[0] $416 billion/year in 2014, https://www.cbo.gov/publication/52463