How Do We Prepare for an Epidemic without Bankrupting Ourselves?

In keeping with the spirit of the times, I will share some of my thoughts on the coronavirus outbreak, and what kind of public policy response would enable us to prepare for such an outbreak while still being economically feasible. I've found much of the popular writing on what the right policies are for epidemic preparedness to be disappointing. It is often too vague (Bill Gates's article in the New England Journal of Medicine struck me as short on details, which is bizarre given his foundation's involvement in many specific responses to the ongoing epidemic). Sometimes it's shamelessly partisan. Sometimes it's implausible (e.g., demanding that society constantly have enough resources on hand to handle an epidemic at any moment). But I think the problem must be talked about in greater detail for viable policies to be conceived and implemented.

In this post, I won't discuss such key elements of epidemic policy as how to enforce quarantines, travel restrictions, etc. to quell the spread, or how best to promote vaccine and drug development. Perhaps in a later post I'll give my opinions on those, but today, I'd like to focus on what seems like the biggest problem right now: healthcare system capacity. It is likely that America's failure in this respect will be the greatest danger, even beyond it's inability to stop the spread or the likely unnecessarily long time it takes to develop treatments and vaccines. As the healthcare system is overrun by people sick with covid-19 as well as people sick with the cold or flu who think they're sick with covid-19, the quality and availability of care will deteriorate, not merely for those with the virus, but everyone else as well. We will need far more beds, more staff, more respirators, and more masks than we have.

So, what is to be done? Well, a simple answer is: increase the capacity of the healthcare system. Frankly, this is economically impossible. Currently, healthcare is about 18% of our economy. Let's suppose an epidemic like the one we're facing effectively doubles demand for healthcare; we need twice as many beds, twice as many employees, and twice as much equipment (respirators, etc.). Well, to permanently maintain the capacity necessary to be ready for an epidemic like the one ongoing, we would have increase that to 36%, and during the vast majority of the time, when we're not facing a massive epidemic, 18% of our resources will be idle, just waiting. Now, they may not be strictly idle, but there would be a massive oversupply of healthcare workers, who would either work less or be less productive while they worked. Some of the excess capacity would be soaked up by an increase in healthcare consumption that would otherwise be considered frivolous and unnecessary. But regardless of how these resources spend their time during ordinary times, a big chunk of our collective resources will be diverted from otherwise productive activity to, essentially, just waiting for epidemic to happen. The decline in our standard of living during non-epidemic periods would be enormous, I suspect more than high enough to render such a situation politically untenable.

And my doubling scenario, I should note, is not especially unrealistic in the case of coronavirus. In reference to the situation in New York, the New York post reports: "New York City’s entire hospital system has about 26,000 beds. If 10 percent of the city’s population becomes infected, and only 5 percent of those 850,000 people ­become seriously ill, that would still require 42,500 hospital beds — and nursing staff to tend to the patients, along with ventilators and other equipment." (https://nypost.com/2020/03/12/nyc-government-doesnt-look-as-ready-for-coronavirus-as-de-blasio-claims/) That's more than a 60% increase in beds needed if 10% of the city becomes infected. Peak infection levels could conceivably be higher, as could the percent needing hospitalization. And as the Post notes, not just more beds but more staff and equipment are needed too.

So, if we can't simply permanently increase the healthcare system's capacity to meet the level of demand seen in extreme scenarios, what can we do? Well, we can increase the flexibility of our healthcare system's capacity. We need a healthcare system which, in a time of crisis, can rapidly double the size of its workforce, it's available space, and it's available equipment. Then, when the crisis is over, it must also be able to reduce its capacity back down to normal levels; the excess staff must go back to work in other sectors, the excess space returned to other use.

Capital Stock

First, let's discuss physical capital, such as medical equipment, which is probably the easiest aspect of preparedness (or at least, the one with the most obvious solution). I suspect the cheapest way to be prepared on this front is simply to stock up on excess equipment. Assuming we want to be prepared for six months of crisis capacity at all times, each hospital would stock up on a six months worth of surgical masks, rubber gloves, respirators, etc. The shelf life of unopened surgical masks and rubber gloves are 5 years and 3 years, respectively, so assuming storage costs are pretty negligible, that means we can expect an increase in the cost of 'healthcare incidentals' of about 10-20%. That's not too bad, since such 'incidentals' are a small fraction of healthcare costs. More expensive technology makes up a greater share of healthcare costs, but I imagine also lasts longer, so overall, that figure may be a reasonable one estimate for the increase in 'capital costs' incurred by the necessary preparedness. There is, of course, an alternative to stockpiling: hospitals could pay the manufacturers of equipment - or the manufacturers of other goods - an annual fee to be ready to increase production as necessary at a moment's notice. The factory could store extra machines, rent extra space (and lease it out short-term during non-epidemic times as warehouse space), and make sure enough staff are willing to work overtime in the event of an emergency. Alternatively, maybe (and I'm making this up) a shower cap factory can be fairly easily converted into a rubber glove factory. A hospital could pay a shower cap factory to maintain the necessary equipment (and give workers the necessary training) to be able to immediately convert from the production of one good to the other. Which strategy - stockpiling or paying suppliers to prepare to shift/increase production - is cheaper probably depends on the good, of course.

Space

Space, or real estate, may be a kind of physical capital, but it is different from other capital in key ways and thus deserves a separate section. I argued above that it's probably untenable to permanently maintain twice the usually necessary supply of healthcare. This is almost certainly true of space (hospital beds, or really, rooms). We can't really afford to have each major hospital maintain a duplicate version of itself, left vacant almost all the time, to prepare for a healthcare emergency. Stockpiling, I'm afraid, is not really viable with space. There are some established ways to temporarily expand capacity that are (thank God) already in use: in New York, for example, hospitals are building tent hospitals in their parking lots. This is a good idea, but I think parking lot hospitals will prove to be insufficient by themselves, and there are probably some procedures that can't be performed in a tent. One thing hospitals can do is make arrangements with landlords of neighboring buildings - essentially, procure 'emergency leases' for these buildings. I'm unaware if such a thing exists, but what my idea of an emergency lease would look like: the hospital would pay nearby office buildings 'rent' (which would be a small fraction of the rent for actually occupying the building), and in return, and under certain conditions (e.g., if the hospital hits 90% capacity usage, or if a state of emergency is declared, or if the CDC says so), the current occupant of the office building has to evacuate immediately, and the hospital can move in and occupy the building for as long as the emergency condition lasts. Then, once the beds and necessary equipment is moved in, you have some more patient rooms, waiting rooms, and staff space. As with tents, of course, there may be some things it's very difficult to do in auxiliary space. However, in all likelihood, auxiliary space would be mainly used for recovering patients and those with less serious cases, allowing the main hospital to devote itself more to emergencies and serious cases.

Labor

How could hospitals maintain capacity flexibility in labor? A couple ideas come to mind from Italy: recruiting retired healthcare personnel and medical/nursing students is a good idea. However, rather than waiting until the crisis is happening, this should be part of a pre-existing contingency plan. Medical and nursing schools could recruit - or even require - students to sign up to serve as medical personnel in the event of a health crisis. Non-emergency physicians or nurses (e.g., cosmetic surgeons) could be repurposed during a crisis. Of course, this is all best not left to be arranged once the crisis has broken out. Somewhat similar to the arrangement I speculate hospitals could make with landlords to expand spatial capacity, hospitals could 'hire' people in the aforementioned groups as auxiliary personnel.  These auxiliary employees would agree - under a specified condition, such as the hospital approaching capacity or a declaration of emergency - to leave their current employment and work for the hospital for a pre-arranged wage. The hospital could pay them an additional regular fee as well if need be to recruit a sufficiently large auxiliary staff.

However, there are only so many students, retired, or nonessential clinical personnel to be reallocated. If one needs to double capacity, these groups likely won't be enough. The healthcare system would have to, in addition, recruit auxiliary personnel from outside the medical field. How might this work? Well, one place to look for inspiration is the military. Historically, militaries have been perhaps the most prominent institution which  have faced this dilemma: maintaining a large standing army is expensive, but maintaining a small one or none at all is risky as it leaves one open aggression by opportunistic neighbors. One way militaries have dealt with this is via conscription. I am not suggesting we employ this method to deal with crisis shortages of healthcare personnel though. Leaving aside the ethical concerns of a 'healthcare draft,' randomly selected conscripts aren't going to be very well-trained, prepared, or motivated, and those shortcomings are serious enough issues for conscripted infantrymen; they're probably even worse for healthcare providers.

A better inspiration for how to deal with the issue of capacity flexibility is the army reserves. Members of the reserves enlist voluntarily, can be assigned roles before they're needed so they can hit the ground running, once called up, and they are required to attend regular training to 'stay sharp' (one weekend a month and two weeks per year). What I propose is something like a Medical Personnel Reserve (MPR for short, so I don't have to keep retyping that). An MPR could be a large, centralized organization like the army reserves, which would assign members to hospitals near where they live (it could also be organized at the state or city level), or each hospital system could maintain its own MPR. These organizations could recruit people for different jobs based on education and/or competence level, then provide them with basic medical training; they would be required to attend periodical retraining sessions to keep them up to date on relevant knowledge. They could be assigned roles based on their actual occupations as well. For example, construction workers might be allotted the task of constructing emergency facilities. MPR employees would get acquainted with the facilities they work in, learn where the fire exits are, etc. so when called to work in an emergency, they would be reasonably prepared. Until an emergency takes place, of course, these auxiliary staff would work at their day jobs. They would agree in advance to the wages they'll paid should they be needed, and may be paid simply for joining the MPR. One might wonder how this will sit with the erstwhile employers of these workers, and I imagine it would managed rather similarly to how benefits like parental leave or prolonged sickness or disability leave. But since auxiliary healthcare workers would be paid in their emergency jobs, there would be less pressure to procure paid leave from their regular jobs in the event of an emergency.

Obviously, MPR employees would almost certainly not be as good as professional doctors and nurses. However, I think it's possible to train many people not in the medical field to perform some of the tasks usually performed by nurses, paramedics, and technicians. This would allow the professional staff to spend more time on the kinds of tasks we would only want to trust medical professionals with. Moreover, nurses and technicians could probably take over some of the tasks usually reserved for doctors, so even if we assume we can't train 'temporary doctors,' maintaining an MPR could still greatly increase the supply of physician services by freeing physicians up from their less 'physician-specific' tasks.

I suspect the best way for MPRs to be organized is by individual hospital systems and local healthcare institutions (but perhaps I have undue deference for the principle of subsidiarity). Hospitals would be paid (I'll discuss by whom later) to maintain a certain degree of auxiliary capacity, and they would decide how many auxiliary employees they need for various job categories, and would recruit as they see fit.

Regulation

In order for all of this to happen, especially with respect to labor and space, regulators need to be accommodating. To be blunt, that means they need to get out of the way. I won't try here to convert those with generally favorable views toward regulations to my broadly deregulationist point of view. I think even if one believes regulations, as they exist, are perfectly optimal and enforced with perfect efficiency, one should still recognize that, in times of crisis, our standards must fall, and that means regulations that enforce our 'societal' standards must relax in order to remain optimal. Ordinarily, we might require extensive training and credentialing for someone to work as a doctor or a nurse. Presumably, this is because there is a risk associated with being treated by a subpar clinician. However, during a health emergency, such as an epidemic of an unusually fatal infectious disease, not only does the demand for medical care go way up, often beyond the number of people who can provide optimal care, the cost of doing nothing goes up dramatically as well. Medical care that would be substandard during ordinary times is better than no care at all during such situations. And almost by definition, expanding the supply of labor, in the short run, means lowering our quality threshold. The best people for the task are already medical personnel; once they're overwhelmed, it is a fact of life that we have to settle for the next tier of available labor. What this means is that credentialing and other occupational restrictions on medical personnel must be temporarily relaxed or suspended during a healthcare emergency for the system's capacity to expand accordingly. We've already seen some of this in the US, with restrictions on physicians' ability to practice medicine who are licensed in other states being lifted in Massachusetts to help deal with physician shortages.

Likewise, regulations on the necessary conditions of hospitals would need to be relaxed as well. Parking lot emergency rooms or office building hospital suites are not going to be up to code. However, as the risk of the disease causing the emergency goes up, the relative dangers posed by unregulated medical facilities declines. If the issue is between people dying in waiting rooms (and likely infecting others in the likely very crowded waiting room) or dying at home because they no there's no point in going to the hospital, and being treated in a parking lot tent, then the choice is clear. Again, when ordinarily substandard care becomes better than no care at all, the regulatory standards must fall to accommodate the change in relative risks. More over, the temporary relaxation of the relevant regulations during an emergency should be written into the law and go into effect automatically, so hospitals aren't waiting left in limbo for a discretionary decision that no one knows when - or even if - it will come. Such uncertainty could hobble the ability to mobilize auxiliary resources.


Paying for it...

I've sort of glossed over a huge issue so far. I've written things like, "the hospital would buy such and such.' But how? And why? Paying partial rent for an office building so you can move in in the case of an emergency is going to be expensive. So is paying thousands of people to ... Hospital systems themselves clearly don't have the incentive - or even the means - to pay for such preparations. On both counts, this is because no one is paying - or will pay them - to maintain such capacity flexibility. They're not likely going to turn a profit on many covid-19 patients.

Realistically, in the system we have, the entities best-positioned to pay hospitals to maintain auxiliary capacity are city, state, and local governments. Governments could simply pay annual subsidies to hospitals determined by the amount of auxiliary capacity they maintain. I think the most cost efficient way to do this would be set an auxiliary capacity goal (either relative to standard capacity, e.g., 2X, or in terms of the number of patients the system can tolerate) for a city (or metro area or district, depending on the circumstances), and 'buy' auxiliary capacity from hospitals (and perhaps other medical providers, such as urgent care clinics) in the area, taking the lowest bid, up to the point where the state has 'purchased' as much auxiliary capacity for its constituents as it wants. This way, hospitals would only commit to auxiliary capacity at a price that compensates them for the costs of doing so, unlike if the state just mandated a degree of auxiliary capacity; conversely, since hospitals would have to bid against each other for funding for auxiliary capacity, taxpayers would be less likely to end up overpaying for it, and the hospitals that can most afford to maintain auxiliary capacity will be the ones that end up doing so. This system would thus be more efficient and cost-effective than simply doling out grants of arbitrary size in a politicized process. The state could then occasionally audit providers to make sure they are maintaining the auxiliary capacity they've been paid to maintain, and it could be made clear that a hospital that demonstrably fails to live up to its capacity commitments during a crisis will be held appropriately liable afterwards.

But we should also ask, how would epidemic preparedness be paid for if we transitioned to a freer health care market (I can dream)? Well, in the 'extreme' scenario where ER's charge patients the market rate per visit, and hospitals are profit-maximizing entities, then the price of an ER visit would skyrocket during a crisis, so even if a crisis is unlikely, the windfalls hospitals would stand to see from them would lead them to maintain more capacity than they would use in during ordinary times. As it stands, hospitals don't really profit from seeing more patients in times of crises; their reimbursement rates are determined by political machinations rather than market prices, and as a result, the appropriate level of preparedness is not necessarily worth achieving. The suppression of the price mechanism in healthcare also obscures how much excess capacity it is worthwhile to maintain. Of course, even in free market utopia, people probably won't pay for ER visits per visit; they would probably by insurance. If, in order to obtain treatment during an epidemic, you need 'epidemic insurance,' then the number of beds or staff needed in the event of an epidemic would be conveyed by the epidemic insurance market: insurers would collect the premiums, then pay hospitals to maintain the capacity to take care of their clients in the event of an emergency. Insurers would have the incentive try to predict the probability and severity of epidemics and coordinate their predictions with the number of clients. They could also vary prices and the amount of excess capacity purchased according to client risk factors, such as age or geographic location. Maintaining the right number of beds is of course ultimately an actuarial problem. Obviously, the capacity achieved wouldn't be perfect, as even the strong efficient markets hypothesis doesn't predict market omniscience, but insurers would have stronger incentives to be right - and prepared - than either governments or bureaucrats would. The latter two are either unlikely to suffer the consequences of ill-preparedness because voters fail to hold them accountable, blaming someone else (or conditioning their blame on whether the politician belongs to their opposing party) or will blame them for the crisis no matter what, no matter how well they handle it, as voters often hold elected officials responsible for things beyond their control that happen during their tenure.

One needn't, I'd add, simply deny ER access to anyone without epidemic insurance during an epidemic. Simply requiring people without it to pay a significantly higher deductible should they need the ER (depending on how much of a bleeding heart one is, one could restrict this to people who can afford it). Some surely find this to be cruel, but medical care is a scarce resource the supply of which is (or at least should be) responsive to demand, and people who buy insurance signal the possibility that they may need it. One may think of this as the 'pro-social' thing to do if one prefers: it allows medical personnel to adequately prepare for their potential arrival. It also allows them (or, insurers, functioning as intermediates) to collect information on prospective patients and assess both the likelihood and nature of care they would need in the event of an epidemic. Those without insurance, on the other hand, render themselves a surprise to the healthcare system; and the shock to the system of unexpected patients not only renders the system less capable of accommodating the uninsured, but everyone else, incurring both a private and a social expense. In my opinion, it is thus not only economically but also morally appropriate that they pay a price for their lack of foresight.

Lastly, even if one believes a free market would leave us underinsured against epidemics, and thus leave the healthcare system underprepared, that isn't an argument for restricting the market or centralizing decision-making. Rather, the optimal solution would be to treat it like an externality and subsidize the purchase of epidemic insurance, or penalize not purchasing enough epidemic insurance. To behavioral economic enthusiasts, such subsidies or penalties would constitute 'nudges' in the direction of procuring the right amount of auxiliary capacity inasmuch as people would tend to underinsure themselves against healthcare crises.

Conclusion

If we hope for our healthcare system to be prepared for the next epidemic, we must make the system more flexible. In this post, I've gone explained what I believe are plausible ways to do this in some detail. Beyond merely coming up with ideas for how to do so, however, we must also ensure that  hospitals and other healthcare providers have incentives to maintain the necessary degree of preparedness. A makeshift solution, given our current healthcare system, would be for the federal government to pay hospitals to maintain a prescribed degree of flexibility on the one hand, and agree to automatically suspect or ease restrictions on the provision of healthcare in the case of a crisis on the other. A still better solution, however, would be to utilize markets to most effectively convey information necessary to plan the optimal amount of auxiliary capacity, and also to give hospitals the incentive to take measures like the ones I've outlined in order to maintain the degree of auxiliary capacity necessary to meet the needs of prospective patients in a crisis.

Finding ways to enhance the capacity flexibility of our healthcare system may be challenging but does not appear impossible. However, it's accomplishment would require more than just some clever ideas of ways to temporarily repurpose labor and resources during a crisis. It also requires a regulatory regime that is flexible to the changes in the changes in demand for healthcare as well as relative risks faced by patients during a crisis. Finally, it requires a healthcare system that has the incentive to be responsive to the demands and expectations of its prospective patients, and to make use of the available information to optimally prepare for a healthcare emergency like the one we are currently facing.