The advent and spread of the coronavirus to the United States poses a serious challenge to our insurance system. Insurance is supposed to help with unexpected adverse events, and the virus certainly fits that category: it is unpredictable, contagious, interferes with normal activities, and can lead to a need for costly care. What makes this unforeseen epidemic different from other threats to health and triggers of medical spending is precisely that — it was not predicted, and it potentially strikes a large number of people at the same time.

Compare it with the seasonal flu, which happens every year, to a greater or lesser degree. Influenza is also contagious, so testing for it and limiting contact with those who test positive makes public health sense. While the medical costs associated with it are usually modest, some fraction of cases are severe and expensive to treat. And it can keep people away from work, so their employers either pay wages for no productive work time or the workers lose wages along with feeling bad. The difference is that from society’s, insurer’s, or employer’s perspective, when it comes to the flu, we cannot predict who will get it, but we can predict fairly well how many will get it from a given population. The coronavirus, on the other hand, was an unexpected add-on source of misery that no one was prepared for.

What changes in public and private spending might make sense in response? The most obvious need is for full public coverage of the cost of testing as a public health measure. Just as we would not expect people to pay to have their water or air tested if there was an unexpected new threat, so, too, should they be able to have a test for the virus to help limit its spread and alleviate anxiety.

Things get less clear when we move on to other consequences. What about the potential for medical bills to treat the illness? That is what 90% of Americans have health insurance for. The biggest issue here is for middle-class people who chose or whose employer chose for them insurance with fairly high deductibles — so that premiums would be lower or wage increases larger. Seasonal flu would present the same kind of modest risk of a high medical bill as does the coronavirus — so people with less generous insurance should have been prepared to handle the possibility of an out-of-pocket payment. Those who paid more in premiums or sacrificed more in wages for lower cost-sharing might reasonably feel it unfair if the government were to help others who made the wrong bet. There is a role for government as “reinsurer of last resort” to help insurers who failed to anticipate the additional cost of the epidemic — but insurers make profits in good years from low risks, so the case for bailing them out in bad years is weak.

However, the biggest challenge posed by the epidemic is the potential need for people to stay home from work or school if they test positive for the virus. If this happens to you, both the monetary cost and the frustration cost is likely to be high — and anticipation of such cost is widely expected to deter people with symptoms from seeking to be tested. The administration has floated the idea of a payroll tax holiday, but it might be better to use the equivalent amount in a much more targeted way. Specifically, people who seek the (free) test and who test positive could be compensated with a lump-sum payment. Not only would such “insurance” (because that is what it is) fill the gap in protection against unexpected loss of wages, but its availability would strongly motivate people to seek testing — since the impact of bad news would be well cushioned. Those employers who now pay some of the cost of missing work might share in this payment, especially if high sick leave benefits pose a threat to the viability of already stressed businesses.

How much might the payment for a positive test be? This is ultimately a political judgment, but a payment of $2,400 (10 lost eight-hour working days times $30 an hour) for a two-week quarantine should be in the ballpark. If the number of confirmed cases, once people present for testing, remains low, the cost of such a program would be modest — but if the epidemic sickens tens of millions, the cost should be high because the help is needed.

Compared to providing tax breaks to all workers and employers, such a more targeted approach ought to be considered. It is too late to say that we should have planned for this already, but it is not too late to devise thoughtful and flexible plans for what lies ahead.

Mark V. Pauly is a Bendheim Professor in the department of health care management, a professor of health care management, and a professor of business economics and public policy at the Wharton School, and a professor of economics in the School of Arts and Sciences at the University of Pennsylvania. He is also a member of The Inquirer’s Health Advisory Panel.