I think that is when you'd start referring to other statistics such as confidence intervals. There are extreme examples of different aspects of viral infections all the time, this does not mean they are the norm or should be the basis for policy.
For example, in my direct area of expertise, Zika virus can be found in the semen of infected men for 6 months after infection. I think the same or similar is true of Ebola. But, these are not typical cases and can sometimes have complicating factors driving the prolonged shedding. To suggest a 6 month quarantine for such men would be extremely expensive and not all that supported by evidence of how common that is. This is why it is important to have more than just medians and ranges. You are correct in saying that we shouldn't be deciding these things based on medians either, I will grant that
.
As far as quarantines go: with regards to cruise ships as you pointed an HK-bound cruise ship issue, the cruise ship currently quarantined near Japan (Diamond Princess), has been handled in such a way that it is now becoming a fascinating, but sad, transmission study in an enclosed space. The quarantine itself onboard the ship might be what is driving the infection to spread.
Another example would be some of the images and tales that are coming out of the people in quarantined areas within China via Weibo, etc. of people running out of food, supplies, not getting adequate care, etc. I suppose we shall see how much of it is true and how much of it is alarmist. But I will say this: heavy handed quarantines are not always the answer, and many times, do not end up being particularly effective at stopping epidemics if one looks at the literature. They will delay, but not stop it.
(Realizing now that I need to make this clear to head off some misunderstandings I've had with others, not on this forum, over the past week. Isolation is the term for if someone is already sick. Quarantine is a pre-emptive measure for people for whom you suspect had exposure. I 100% stand behind the efficacy of isolation procedures.)