Post #945: Masks, part III: The Germans have the good sense to issue N95s. Medicare should copy.

Posted on January 12, 2021

I have heard it said that whatever bright idea you have, somebody’s already had it, and posted it on the internet.  Twice.

And so it goes with the idea of having the government issue N95 masks to the general public (Post #942).   Based on the recent JAMA-published mask test, use of an N95 respirator results in a roughly 14-fold reduction in virus exposure relative to a typical procedure (surgical) mask or cloth mask.   My guess is, use of N95 respirators in place of cloth or disposable surgical masks is the single most effective step that could be taken to reduce the population’s exposure to cornavirus.

Seems like a no-brainer to me.

Turns out, this idea has not gone unnoticed.  Outside of the U.S.

A friend of my wife’s reposted my mask post, and got a response from yet a different friend who lives in Germany.  And guess what?  The German government is providing N95 masks to the entire over-60 population of Germany, as a way to get through this winter wave of coronavirus.

And they’re doing it in a way that’s smarter than what I suggested, which was to mail masks to everyone.

You can read a clear description of what they’re doing in this article.

Their mask distribution started on December 15.  At that time, anyone age 60 and over or with a medical condition that puts them at high risk could walk into a pharmacy and ask for three FFP2 (essentially, N94) masks.  They either show an ID to prove age, or attest that they have a relevant condition (e.g., diabetes).  After January 1, masks will be issued two more times, but at that point individuals need a certificate from their insurance company.  (Virtually everyone in Germany has health insurance through one of their privately-run sickness funds.)

And so the Germans achieve the same goal I was looking for — every targeted person gets a few N95 masks — in a far more efficient fashion.  It’s efficient because only those individuals who want to wear an N95 will bother to pick them up.  And it’s efficient because they targeted high-risk groups.  By contrast, I would have wasted many quality masks by mailing them to individuals who wouldn’t be bothered to wear them, and by mailing them to the entire population regardless of risk.

The sole downside that I see to the German approach is that, by report, they actually have to go to the pharmacy to get them.  But surely there are ways around that.  I also note that they paid the pharmacies 6 euros (about US $7.20) per mask for the mask, and for dispensing the mask, which seems a little generous.  On the other hand, if you want wide participation by pharmacies, it doesn’t hurt to make it worth their while.

And so, let me copy what the Germans did, but modify it for the American context. 

First, forget the idea of mailing out the physical masks to all US residents (Post #942).

Second, start by distributing masks via the Medicare program.  This targets the elderly (More than 95% of those over age 65 are enrolled in either Part A or Part B of Medicare) and the disabled, as well as those with end-stage kidney disease.

Medicare already has plenty of infrastructure for mass distribution of health care goods via the U.S. mail.  So distribute them initially using existing mail-order suppliers, both Part B (Durable Medical Equipment (DME) suppliers) and Part D (drug suppliers).

Individuals could sign up to receive a few N95 masks via the existing MyMedicare.gov portal.  In addition, in the traditional (fee-for-service) portion of Medicare, you could simply send a few masks automatically to anyone with DME- or drug-based indicators of having a relevant condition.  Anyone receiving diabetes supplies, or taking drugs indicating heart failure or cancer treatment, and so on.  (Or for that matter, anyone with a physician visit in a group facility such as a nursing home or assisted living facility.)

If this were well-received, the program could be expanded along the German lines.  Just show up to a pharmacy, show an ID or sign a form, and collect a few N95 masks.  There would probably be some abuse of that system, but, seriously, I don’t think that’s really our greatest worry at the moment.

Interestingly, the Medicare program has “demonstration authority” to implement this on a city-by-city basis.  That is, for example, how they initially tested the current bid-based program for supply of DME.  So the fact that you don’t have the ability to implement this nationwide is not a barrier to some form of implementation.

I suggest that Los Angeles would be a great place to start. And if you were really of a mind to test this, then do as was done for DME, and pick an additional random sample of cities for implementation.

Having worked with or around the Medicare program for many years, it’s not clear that they are agile enough to do this in anything less than a six-month timeframe.  At least, not within existing “rule making” process.  For everything they do, they have to give public notice of the proposal, take public comment, and then issue the rule.  Assuming they could expedite it, that by itself would take months.

But that’s OK, because the U.S. Congress would have to authorize this anyway.  They’d have to make those N95 masks a covered benefit under Medicare.  And as part of that Act of Congress, they could instruct the Medicare program to bypass the normal rulemaking process in this emergency situation, and just get it done under their inherent reasonableness authority.  (I.e., that Medicare can just pay for stuff, as long as the price seems inherently reasonable.)

And this all depends on their being no critical shortage of N95s any more.  Which is what I argued in my original post on this topic (Post #942).

If you think this is a good idea, write your Member of Congress.  Suggest that Medicare begin supplying N95 masks to the U.S. elderly population.


A little cost-benefit calculation

The question here is, how effective would N95s have to be in order to pay for themselves in this context?  Just roughly speaking, not with any great detail.

Cost:  Last month, I bought a bag of Kimberly-Clark N95s for about $1/mask, on Amazon (Post #918).  (They are now also out of stock on Amazon.)  Those appear genuine and work fine, but are hard to put on properly.   For a program aimed at the elderly, you’d be better off with a hard cup-type N95, like a 3M N95.

So price those out at $3 each.  Each elderly person gets three, plus postage and handling, so call that a generous $20/person/month.  (Or roughly what Germany is paying.)

As noted earlier, based on Medicare statistics, Medicare is currently paying roughly $25,000 per COVID-19 hospital admission, per this reference on the Medicare website.  (And for all the loonies out there, please note that Medicare only pays additional for the cost of COVID if there’s a documented positive PCR test for COVID in the medical record (Post #802), so these aren’t made-up hospital admissions.)

And so, strictly speaking, using those assumptions on costs, and only including hospital inpatient facility costs, provision of N95 masks would be cost-effective if it would avoid one COVID-19 hospital admission per month for every ($25,000/$20 = ) 1250 elderly US residents.

What is the current rate of hospital admissions for COVID-19, per month, per elderly US residents?  Assuming I can read this CDC chart correctly, in total, at this point, more than 1 percent of the entire U.S. elderly population has been hospitalized for COVID-19. 

Boomer remover, indeed.

Most recently, that cumulative rate went from 908 / 100,000 to 1040 / 100,000 over the past four weeks, or roughly 140 / 100,000 / month.  Inverting that, the current rate is a current rate of COVID-19 hospitalization of about 1-in-700 elderly Americans per month.

The bottom line is that if use of N95 masks cut the current COVID-19 hospitalization rate for the elderly roughly in half, then the reduction in inpatient facility costs, by itself, would pay for the masks.  (Under the assumption that it costs Medicare $20/person/month to supply the masks.)

As far as I can tell, there’s no data to support that large of an actual, in-the-field effect of wearing N95s versus other masks.  It simply has not been measured, I think, because that outcome (hospitalization) is rare among the populations for whom mask use has been formally tested (typically, health care workers.)

And so, narrowly measured, this policy would not be guaranteed to be free (in the sense of paying for itself via likely reduction in inpatient facility costs for mask wearers).   But it’s not that far off, either.  Whether or not it would be free if a broader measure of costs were used is not clear.  (For example, including physician-related and post-acute costs, and most importantly, including a measure of cost savings from reduced transmission of disease to others.)

But when in doubt, look to the smartest person in the room for guidance.  And as far as I’m concerned, among the Western nations, that has been Germany.  Starting with their efficient use of pooled COVID testing (written up as a set of math problems in Post #605), and their early ban on church singing (noted in Post #679), they seem to have been one step ahead of the curve all along.  And now, they’re issuing N95s to their entire elderly population.

We probably should take notice of that, and, if possible, follow suit.