As for the original question, in Canada, the PM simply said the harbours were closed to cruises. Could any of the politicians in the U.S. do the same? Maybe it's more a governor level than a federal level, but is that power available to an elected leader or only to the CDC?
While the PM made the public announcement, the restriction is actually done through Transport Canada regulations, as they are the relevant regulatory body. See this backgrounder for when the restrictions through Oct 31 were announced:
https://tc.canada.ca/en/initiatives...measures-cruise-ships-other-passenger-vessels It also notes that smaller vessels were subject to provincial/territorial/local requirements.
For the US, this article by the National Conference of State Legislatures gives a good overview of State quarantine and isolation statutes:
https://www.ncsl.org/research/health/state-quarantine-and-isolation-statutes.aspx Even if for reasons they couldn't actually outright close the port [I don't know if they have that authority; eg there may be issues re: interfering in interstate trade and such], they could do things like impose quarantine requirements on anyone entering a state/locality, which could have the effect of nobody would be allowed to get off a ship, or you unless you were a state resident you would have to be eg qurantining for 2 weeks before you could board the ship, which is not feasible for most people.
As for the OP's question, I think there are a lot of factors that are going to influence whether the CDC needs to change its current order wrt cruise ships or stay the course. The one thing the current order has is a lot of *time* built into it, which even if all goes well and everything goes exactly according to plan, means sailing with "actual" passengers (not test sailings) doesn't start until into 2021.
Consider too that even if they were allowed to sail, some of the cruise lines may choose not to if for example, land case # were skyrocketing, as they may not wish to assume the risk. They will have to do their own risk calculations on top of the regulations, which will affect things like how many ships they actually have sailing, itineraries, etc.
As for spreading within a household, the studies I have read where they have looked at covid spread within households have shown that usually the majority of household members do NOT catch it when one member gets infected. While covid is contagious, it is not
as contagious as something like measles. There are many factors that will influence the situation of course, from mask-wearing of other household members to once the infected person is known to be infected are they isolated from the others, etc. There is also much that is not known -- the Diamond Princess is a good example for this: they had a range of situations where stateroom mates got it, others where they didn't, others where some did and others didn't... so there is much unknown about why some people catch it and others don't even when they are sharing the confined space of a cruise stateroom with someone who is infectious and presumably similarily exposed. Here is a Public Health Ontario summary of one study looking at household spread of covid:
https://www.publichealthontario.ca/...ics-household-transmission-covid-19.pdf?la=en
The info about spreading will have an influence on the rules that get set for how cruise ships handle cases. I assume they will be erring on the side of caution, though. But it informs things like "what do you do when one person in a stateroom develops symptoms/tests positive" in terms of isolating them in the same stateroom, in a separate stateroom, etc. Since not everyone in the stateroom is "doomed" to come down with it, even though they would have been exposed, I assume they would take the symptomatic/positive person and quarantine them alone elsewhere, and then isolate and monitor the rest of the stateroom people in their stateroom [is some cleaning done in there too ?]. This fits with the finding of the study PHO summarized above: "
When index patients were isolated while at home immediately after symptom onset until hospitalization, the secondary attack rate was zero, compared to a secondary attack rate of 18.3% when index patients were not immediately isolated while at home. "
SW