Taking this into account, wouldn't cluster nature itself be an argument for the more prolonged kind of exposure over an accidental, one-time casual contact? In other words, if casual transmission was indeed likely, wouldn't we see more smooth and uniform progression of the disease in the population? With over 3 months of data at this point the relation between the granularity of epidemiological spread and the transmission parameters of the disease can be investigated at this time.

This is important because it may help in adjusting the policy to more effective control of the disease focusing on prevention of longer-time exposure and safety practices in "dense" environments (such as workplaces, nursing homes etc) rather than complete elimination of casual contact. For example, it's an established practice in many jurisdictions that after a mandatory quarantine of usually 14 days a person is deemed completely recovered and can return to their workplace or residence etc while there are indications that production of viral agents in non-symptomatic carriers can last considerably longer than that.