tl;dr. A document saying it opposes the conventional wisdom on U.S. health care spending actually, in my view, supports the conventional wisdom on U.S. health care spending.
The conventional wisdom (which I agree with)
The conventional wisdom is that Americans spend too much on health care and get too little. This conventional wisdom resonates with just about anyone who’s ever had to deal with the U.S. healthcare system, also there are graphs like this one, based on data from 2007:
The story
An anonymous correspondent points us to this analysis of U.S. healthcare costs that argues against the conventional wisdom of the reason for these healthcare costs, instead making the point that richer countries spend more on health care than poorer countries, and the U.S. is one of the richest countries in the world, so we spend a lot, so the U.S. doesn’t stand out at all from the crowd:
I don’t quite buy the above graph, as the line going through USA seems reliant on the iffy quadratic term in the regression—but even if you draw a straight line and drop the mysterious “ARE” point, the USA would not be much higher than the fitted line, only 10%-15% higher, it appears.
What does the U.S. get out of all that spending? The author of that post seems to agree that we don’t get much, writing:
America’s mediocre health outcomes can be explained by rapidly diminishing returns to spending and behavioral (lifestyle) risk factors, especially obesity, car accidents, homicide, and (most recently) drug overdose deaths. . . . The diminishing returns are evident in cross-sectional analysis:
In the earlier graph shown above, Luxembourg and Norway are also on the high end of spending relative to life expectancy, but nothing compared to the United States.
My take
My main reaction is that the main take-home point of the post is the above-cited bit about diminishing returns, which seems consistent with the conventional wisdom that the U.S. overpays for health care. Maybe Luxembourg and Norway do so too, to a lesser extent, but that doesn’t make me feel any better!
This connects to a general statistical issue that came up a few years ago, which we called the “all else equal” fallacy. The comparison of the U.S. to other countries convincingly shows that richer countries tend to spend a greater proportion of their consumption on health care, with the U.S. not standing out much except for being richer. But, as discussed above, I think this is all consistent with the conventional wisdom that we’re overpaying. Indeed I think this conventional wisdom is supported by the argument in the linked post about diminishing returns to health-care spending.
To put it another way: we in the U.S. are overpaying for health care. It might be that as other countries get as rich as we are now, they’ll overpay too, just like we do! Or maybe not. Maybe they’ll learn from our experience and decide not to spend an increasing share of their consumption on aspects of health care that aren’t improving health.
I like the general approach of the above-linked post, which is data-focused, and all about drawing the most direct inferences from available data rather than being flashily counterintuitive. It just seems to me that the title of the post, “Why conventional wisdom on health care is wrong,” does not fit its content.
The political angle
There’s also a political angle here that I don’t fully understand.
It seems to be a position on the left or center-left who argue that U.S. health care spending is out of control, and a position on the right or center-right that our system is just fine.
I kind of get this correlation. Until the Obama-era health-care law, the alternative to the U.S. system was considered to be some sort of socialized system, a national insurance or Medicare for all, and often these proposals were motivated by comparisons to Canada, France, or other systems. So if you’re a U.S. liberal, it makes sense to view the American system as worse than that in these other countries, whereas if you’re a U.S. conservative, it makes sense to argue that, despite appearances, our system is just fine, perhaps even better than elsewhere.
On the other hand, now that we have some sort of universal health insurance, maybe there’s motivation for liberals to say good things about our system and for conservatives to complain.
But, beyond all this, there’s a lot of government involvement in the U.S. health care system. Consider all the weird results of Medicare spending rules, which somehow percolate through the entire system. It would seem very natural for a liberal or a conservative to attribute some of the problems with our current system to a tangled bureaucracy. To support our crappy system just cos it’s not officially socialist . . . that just seems like a mistake. Go back to the above-linked post, see the bit about diminishing returns and the bit about richer countries being willing to spend more, put that together and you can see what a mess we’ve gotten into here! From the other direction, there’s no reason for people who see problems with our health-care system to think they need to argue with many of the points in that post.
Just commenting about the mystery, ARE stands for United Arab Emirates.
Thanks!
A couple things. Is the x-axis on the first graph government spending while the x-axis on the second graph is total spending per capita, including personal out-of-pocket spending? Otherwise the values don’t make sense to me and seem to be measuring slightly different things. The values on the x-axis on the third graph seem consistent with the first, though. But then, I also don’t understand why anyone would expect there to not be diminishing returns for life expectancy as a function of health care spending. We don’t live forever, after all, and no amount of money can (yet) seem to increase the maximum length of human life all that much.
Cuba is the best example.
Chronic shortages of medications, decades old equipment, and higher life expectancy than the US. That should make people wonder what they are paying for.
Eg, the 80 million people on blood pressure meds with NNT of ~100 and chance you’ll accidently under/over-dose yourself then get classified as a drug poisoning death.
“and higher life expectancy than the US”
If younger people are gradually increasing doughnut consumption and reducing exercise (for example) and becoming over all less healthy, how long will it take to show up in the life expectancy data? IMO it should take decades at least, shouldn’t it? So isn’t there some lag in life expectancy data, at least for certain types of health factors?
Which country are you referring to? I’m not sure about Cuba, but in the US you are encouraged to eat healthy and exercise*. Unfortunately “eat healthy” got defined as s low-fat/high-carb diet.
Anyone can prove to themselves with a few days of eating under 100 g carbs (~20% of daily calories) that you simply crave less food.
But that is the point. If cheap and safe interventions like diet/exercise are more effective than these other ones, why is $4 trillion a year being spent?
Life expectancy in the US peaked in 2014 and has plateaued, at best, since then. A few years ago I looked at the CDC wonder data and saw this was primarily due to an increase in middle-age deaths attributed to over/under-dosing painkillers and blood pressure meds. That is when it appears the US healthcare system became a net negative for society.
*except during covid when I saw the signs encouraging children to go outside and play get taken down
I am pretty sure that ethnicity plays a role in Cuba’s performance. The World Bank estimates Cuba’s 2019 (ie, pre-COVID) life expectancy at 78. https://data.worldbank.org/indicator/SP.DYN.LE00.IN?end=2019&locations=CU
In contrast, US life expectancy for Hispanics in 2019 was 82. https://www.researchgate.net/publication/359948254_United_States_Life_Tables_2019
And more generally, the substantially higher life expectancy for Hispanics versus whites (79) casts some doubt on the value of life expectancy as a proxy for access to effective health care.
Then the US spends 10x more per capita but it isn’t enough to overcome whatever genetic and cultural differences.
The conclusion is still that there is far less benefit than typically expected. If you read the medical literature carefully this is no surprise though. Even taken at face value, it contains many standard interventions with very minor net benefit on average. It once again goes back to “statistical significance”.
Are our health care policies insurance or Insurance plus maintenance? Imagine what your car insurance would cost if it included all maintenance costs. High income seniors pay for Medicare IRMAA plus Gap policies plus concierge medicine before any actual health problems.
Quite right and you should also add Long Term Care Insurance. Next year we hit the highest Medicare premium category only because we sold our home and had a large capital gain that bumped our AGI way up. When you add up all the insurance payments (we both have concierge practice as you can hardly find internists in the DC area that are not in a plan), we will pay about $25K before even seeing a doctor.
When comparing health care costs in America vs. the rest of the developed world, I suggest that all that (numerical) graphing is unnecessary. The essential point is summed up in the phrase,
“with benefits,”
an entirely mysterious concept to the rest of the developed world where health care insurance is untethered to employment. Or is dependent upon age, with the worst age in the U.S. being not quite old enough for Medicare.
To get a quick glimpse of the weirdness of the medical situation in the U.S., go to
https://www.npr.org/series/651784144/bill-of-the-month#:~:text=NPR-Kaiser%20Health%20News%20Bill%20Of%20The%20Month%20%3A,can%20tell%20us%20about%20the%20health%20care%20system.
Insurance companies don’t actually pay those huge bills the patient sees:
https://en.wikipedia.org/wiki/Chargemaster
As it says there, you need to recognize it is a “starting point for negotiations” and submit your own low-ball bid. The actual price will be 1-50% of the initial offer.
And I thought I was an expert on the peculiarities of American medical billing! Thanks to Anoneuoid, I now am aware of the term “Chargemaster.” From the website he (she? they?) mentioned
“Furthermore, most of the items on the charge master file are written in code so that only the hospital administrators and a few experts in the field can interpret their meanings.”
One of the more obvious inefficiencies in our medical system is the large number of people who spend their time dealing with coding issues, both on the medical care side and on the insurance side. Even the doctors have to think about how to phrase their reports to fit the codes.
https://www.econtalk.org/vivian-lee-on-the-long-fix/
I don’t think medicare is a good benchmark. Likely other prices go up in response to artificially low medicare prices.
But, in many cases even those “artificially low” medicare prices are probably still to high for the actual (rather than percieved) benefits people are getting.
Anoneuoid said:
” Likely other prices go up in response to artificially low medicare prices.”
Exactly. The low reimbursement by medicare functions as a tax on everyone else.
By the same token, prices for equipment and drugs in the US are higher because other countries have price controls. So, in effect, Americans (mostly) or any other country without price controls is subsidizing the cost of drugs and equipment for countries that have price controls, just as private payers are subsidizing medicare recipients.
Among US medical equipment manufacturers, the discussion is always “remind me: why do we sell for virtually no profit in Europe? (or Canada or Australia or Japan)”
These issues need to be addressed through trade agreements.
One of the (many) tragedies of the American health care system (or systems, if you prefer) is the lack of data for sensible analysis. There are now requirements for transparent hospital pricing – but those are list prices which almost nobody pays. The Medicare prices are publicly available – but, as pointed out, these are absurdly low prices that would not be sustainable in the market without higher payments from insurers and individuals. The commercially negotiated prices are only available to researchers with an inside track or to the insurers themselves. To conduct a meaningful analysis of health care prices is almost impossible given the data readily available, and the fragmented pieces of data reflects the fragmented nature of American health care.
I personally view the lack of data as an American tragedy, indicative of a number of more serious tragedies in our system. While some of the care is the best you can find anywhere, much is not, and a huge drain of resources goes to the administrative overhead of the complex and incompatible billing systems (I’ve seen estimates as high as 20%, but I personally think it may be higher, and in any case, I haven’t seen any data that would permit a careful analysis of the administrative costs). I could view these administrative costs as worthwhile, if the competition between insurers really was about the quality of care they enabled – but the system is so opaque that it is difficult to see how insurers are competing in any ways that improve quality and/or choice.
If anyone wants to think about these things, just take a look at Medicare Part D coverage (drugs). Every year, you get to play a Vegas-type gamble: which particular drugs to you think you will need, and then shop for the plan that provides the best prices for those particular drugs (Medicare even provides a good tool for doing this shopping). Of course, if you suddenly need a different drug, you are stuck – but only until next year’s lottery begins again. The only thing you can’t afford to do is forgo part D until you need it – unless you sign up for part D when you become Medicare eligible, you pay a huge penalty to sign up later. Who ever thought it made sense to design a system asking patients to try to guess which drugs they will need each year?
You pretty much make the point of the paper Andrew is criticizing: Handing out luxury health care benefits is a way for companies to attract talent. The “luxury” aspect of these benefits *intentionally* increases the spending without necessarily correlating it with outcome.
People should actually read the analysis. The author notes:
” If you want to see further substantiation, please click through as I’m mostly focused on covering the highlights in relatively broad strokes.”
I like this graph, which shows the cost of hospital services increasing 200% over a 20yr period, while the price of everything else declines. Sounds like big inflation – except that people spend much less time in the hospital than in the past because of developments like laparoscopic surgery. Even when I had a “normal” surgery, I spent less than six hours in the hospital.
Overall I agree that if you’re going to have a nationalized health care system, connecting it with employers is a poor choice. But then again I don’t support a nationalized system.
Connecting your health care system to employers is incredibly bad full stop.
Even if you’re going to have a market based system, the only connection to employers should be through money exchange like any other good. Just make employer provided health insurance illegal. And require insurance companies to accept anyone on the basis of a price for individuals based on a short list of criteria. Age, sex, height, weight, smoking status, results of some small set (0-4) of medical tests for example.
Daniel,
Everyone talks about the US health care system, but there is no such thing. There is a private health care market and a variety of government programs at every level of government. Of course these systems interact with one another indirectly (as Anoneuoid points out regarding the cost of medicare), but they don’t comprise a system.
Your idea regarding making employer sponsored health coverage illegal has crossed my mind but in the end why should employers not be able to offer health coverage as part of a compensation package? If anything wouldn’t the profitability of high-end employer sponsored coverage make it more likely that companies would offer lower cost low-end coverage?
There’s a whole bunch of parts and they interact… That’s what’s a system is.
Daniel, by that definition there is nothing that is not a system.
That’s right. An atom is a system of interacting subatomic particles. An economy is a system of people interacting through trade and companies and markets. A galaxy is a system of interacting gravitationally bound stars and planets… It’s systems all the way down.
It looks that it is generally agreed that the USA pays the most for healthcare without corresponding leadership in broad based outcomes. Very few dispute this. Is there an acceptable justification for this? I understand how the rich spend more on a given good than lesser advantaged people or nations. However, do the rich spend a greater percentage on some good/service? I spend more on wine from St. Emillion now than I allocate on wines from Modesto, but I am not spending a larger percent of my income. I have read that an MRI in Japan is a lot cheaper and without the pre-approval hassles of the American system. We spend so much on healthcare that it is simply imprudent not to look closely at the situation.
Americans pay for care in three ways. There are governmental subsidies that come out of our taxes. This is less for us than other nations. There are employer provided benefits that obviously come out of your paycheck. We also pay out of pocket where we have no economic clout; when you break your leg you don’t have a chance to be a smart shopper. I believe that the relative power of the individual buyer versus the big corporate provider puts a big advantage on the big, unitary provider versus the atomized consumer. It’s only a market when you have the option to walk away. When you’re in a dark alley facing a guy with a gun, that’s not a market. When the Xray shows a mass in your mediastinum, that’s not a market either.
As a patient, the Japanese system is seriously wonderful. There’s no BS. Whatsoever.
Everyone is insured, the premiums are affordable, you get to see a doctor whenever you want. Referals to specialists are provided without problem. Actual out-of-pocket expenses are tiny (by US standards). Pretty much everything is covered. (In my experience, a treatment being not covered means it’s something that’s actually a bad idea, e.g. cosmetic surgery, things that are still experimental.)
What’s the trick? Draconian price controls. How do the doctors make money? They largely don’t. Except for the non-covered stuff. I’m perfectly happy with metal molars, but if you are an actor or insecure, you’ll have to pay for ceramic. Ceramic is covered for front teeth. (Note that dental is covered. For everyone.) A Go-playing dentist friend once confided that providing patient services to folks rich enough to afford the high-priced/high-profit cosmetic dental stuff was exhausting, since rich folks are obnoxious jerks requiring constant praise and arse-kissing.
Note that the Japanese government had a sleazy industrial policy to artificially support the Japanese MRI/CT scanner industry, so Japan is waaaaaay oversupplied with scanners, so prices are gloriously low.
One result is that the doctors form political groups to lobby right-wing politicians to increase payments. A brother in-law is a dentist, and I once played in the band at a political rally in support of an LDP candidate. Some blokes two or three levels up in the dentists association hiererchy got charged with bribery and other nefarious things shortly thereafter.
I agree that it’s hard to draw conclusions from statistics, since the populations are so different. But the US system really is completely and totally insane by current standards in other industrialized countries.
I’m a health services researcher, so I think about this problem a lot. My (increasingly discouraged) opinion is that 1) this is indeed conventional wisdom (the diminishing returns thing) but 2) this phenomenon is driven by such massive economic, political, and historical forces that there is no practical solution. A lot of solutions people discuss start from an unstated premise of having god-like power to restructure major components of society or the economy, and then explain how this would solve the problems. Like yeah… obviously.
Some things would surely help at the margin. Like, it is very broken to have this 4 party economic system between employers, insurers, providers, and patients where basically no one is paying directly for the services they receive so all the normal capitalistic signals break down. Restructuring that to somehow be more transparent would make a big difference; all you need to do is completely revamp what is currently 20% of the economy that has been entrenching itself over the last 70 years or so.
Regarding the relationship between healthcare spending and GPD: a lot of the things we spend lots of money on (e.g., cancer treatment, end of life care) there is no cost effective alternative. The alternative is to die. There’s a certain “pascal’s wager” type logic where even if this treatment is not likely to save you, it’s still worth spending all the money you have left for the tiny chance (even going into debt, if you can). ~25% of our healthcare spending occurs in the last 3 years of life. The more money you have, especially towards the end of life, the more you’ll waste. And I don’t expect this phenomenon to change under a different economic or healthcare system setup (though it might change across cultures, where less individualistic cultures might see the value of putting that money towards the benefit of others instead).
How are you defining alternative? As in expensive enough that someone paid for FDA approval? In that case you will only get expensive interventions.
Eg, chemo/radiation therapy probably partially (if not mostly) works via caloric restriction (nausea, vomitting, appetite loss). So if you want something with a chance of extending your life by 6 months you can try that rather than expensive poison. There are plenty of alternatives out there with at least small chance of minor benefit.
“Eg, chemo/radiation therapy probably partially (if not mostly) works via caloric restriction (nausea, vomitting, appetite loss). So if you want something with a chance of extending your life by 6 months you can try that rather than expensive poison.”
Somehow, against impossible odds, we have again reached a new low.
The warburg effect is 100 years old science. Theres thousands (maybe even millions) of papers on caloric restriction slowing tumor growth. And we know chemo/radiation damages the GI-tract since it is filled with frequently dividing epithelial and immune cells, which reduces absorption of nutrients.
What issue do you have with it?
Anoneuoid –
> What issue do you have with it?
What’s “it?”
Consider that maybe he doesn’t have an “issue” with the Warburg effect and if so, there might be another a spect of your comment that he was responding to.
Maybe he denies the typical treatment for advanced cancer only appears to extend life by around 6 months on average? Or that chemo/radiation is expensive, or that it is poison with the idea it kills the cancer before it kills the organism?
I don’t see anything that an informed person would disagree with, but would really like to know what objections exist.
What you don’t see is interesting to me.
I think it’s a form of arguing from incredulity mixed with a lack of perspective taking/cognitive empathy.
Having worked on a paper where I modeled survival after diagnosis with different cancers, I can provide some context for the cancer claim. It is indeed true that survival post diagnosis is pretty grim for many cancers. For example pancreatic or glioblastoma or a couple others. In those cases it’s entirely likely that the treatment does very little. But in other cases the treatment may actually do a lot. For example there are treatments involving antibodies that cause your immune system to kill the cancer cells. That has extremely good effect relative to the older treatments that simply target dividing cells.
https://www.biorxiv.org/content/10.1101/2020.08.26.268987v1.full.pdf
In the section on survival see figure 8, this is posterior estimates for aging multiplier. Glioblastoma (GBM) has a multiplier around 20, meaning your risk of death increases like a person who ages about 20 years worth for every actual year you live. And that’s **with treatment**.
Daniel –
> For example pancreatic or glioblastoma or a couple others. In those cases it’s entirely likely that the treatment does very little.
Of course. But it’s hard to dictate to people not to take a long shot. That the average benefit might be minimal can be a hard sell. As a spouse of a long time hospice nurse, I’ve often seen many situations where in my judgement people would have been better off focusing on quality of life over long shot treatments while feeling miserable, but I don’t don’t get to make the call for other people. Perhaps the worst scenario is when doctors seem to overemphasize treatment and not give palliative care sufficient emphasis. Judging the reasons why docs do that is complicated.
At any rate – Anoneuoid’s comment led me to a Google search and I found the first hit intersting:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8749320/
Daniel, heres some of the common side effects of monoclonal antibody treatment:
https://www.mayoclinic.org/diseases-conditions/cancer/in-depth/monoclonal-antibody/art-20047808
Since this always happens for apparently effective cancer intervention, perhaps instead of a “side effect” this is at least partially how these interventions work. I suspect it is a large part, but I’ve never seen a clinical trial even attempt to account for this obvious factor.
To Joshua, I am empathetic to the patients (which could be any of us), but not to those they trust who have failed to investigate such an obvious explanation.
My priority is actually understanding then curing cancer. If that means researchers/doctors need to abandon some sacred cows, ok. Those feelings are not my priority.
Anoneuoid –
> Since this always happens for apparently effective cancer intervention,…
It happens for most chemo interventions whether they’re effective or not. And many cancers are associated with a loss of appetite.
Seems to me what you’re doing here is analogous to p-hacking.
Insofar as those symptoms correspond to caloric restriction, we should expect the intervention *was* effective at slowing the growth/spread of the tumor. It was just more “effective” at killing the patient.
And yes, a natural response to cancer can be reduced appetite. It is probably evolutionarily advantageous in some way.
So, if you want a moonshot cancer treatment without going bankrupt that seems like a good alternative. If someone ever gets the funding to learn more detail or even show otherwise, then great.
NB: It is sad to have to mention it, but obviously too much restriction leads to malnourishment and will not be a net benefit.
Anoneuoid –
Even assuming caloric restriction always has a net beneficial effect on cancer progression (which seems a huge assumption, particularly since I’d imagine there are potential costs both physiological and psychological), you have yet to articulate (at least that I can see) a discernable way that you’ve differentiated any potential benefit from the benefits of the chemo itself (where countless studies have explicated the mechanistic action), let alone quantified that beneficial effect to support your contention that it outweighs the benefits of the chemo (or actually explains that those putative benefits are spurious).
And to top it off, you Just-So storify to reverse engineer evolution to confirm your theory.
It’s almost a wonder that anyone ever died of cancers that reduce appetite before chemo was developed.
“Always has a net benefit” doesn’t make sense. The idea of “net benefit” is that that more people are helped than harmed.
And there is a huge existing preclinical literature on this topic, the only assumption is that what we see in animals also applies to humans. That assumption should be checked, and the details worked out.
It is baffling that this hasn’t been done, and from this discussion it seems people actively resist it. Why would you not want this research done?
Anoneuoid –
> “Always has a net benefit” doesn’t make sense. The idea of “net benefit” is that that more people are helped than harmed.
Net benefit per individual. Presumably there are some individuals where there is no net benefit or even a net harm. And for others it is marginal and less than the benefit from chemo. Again, I can see no way you can differentiate let alone quantity those different scenarios.
> and from this discussion it seems people actively resist it.
And…
> Why would you not want this research done?
Again, you might benefit from working harder at perspective taking. Why would you (wrongly) think I would?
It seems we agree it should be investigated. In that case, perhaps researchers should figure out how to distinguish between the different mechanisms (in general, killing cancer cells directly vs starving them via caloric restriction).
That is great, then we can agree it is possible to figure out what is going on using science. Do you have any ideas as to how we can distinguish between the mechanisms?
Anoneuoid –
Yes, we agree it seems very much worth investigating. The article I linked above indicates it is happening.
> Do you have any ideas as to how we can distinguish between the mechanisms?
Not really. Seems very complex to me to tease put the differential effects, particularly given the ethical complications of withholding chemo treatments. As that article I linked above indicates, continuing research into the molecular mechanisms seems an obvious path forward.
Distinguishing between the possible explanations is an essential part of science. Without it, you have no science.
Keep in mind there is no proving or disproving in science, rather you see which explanation fits the observations best. So we aren’t looking for perfection here.
You can do it, come up with some ways to check whether caloric restriction is part (or even all) of how chemo/radiation slows tumor growth. It really isn’t difficult, just allow your mind to think scientifically rather than using authority, consensus, or arbitrary rituals.
Sorry if that sounds patronizing but I feel you may be close to a breakthrough.
Anoneuoid –
> You can do it, come up with some ways to check whether caloric restriction is part (or even all) of how chemo/radiation slows tumor growth.
Despite reaching conclusions without supporting evidence, you now tell me you can’t describe how to do it without telling me you can’t describe how to do it.
Despite your condescension, arm-waving at complex processes doesn’t suffice.
I note that I repeatedly observed you have no way to make the kinds of differentiations you argue you have made. And still you don’t address that. So instead you arm-wave.
Given you have nothing substantive to offer I’ll go with what the researchers who are actively involved in the molecular-level research have to say.
> rather than using authority, consensus, or arbitrary rituals.
And
> Sorry if that sounds patronizing but I feel you may be close to a breakthrough.
You had the opportunity to learn after making obviously wrong conclusions. Seems not to have slowed you down even a bit. While I’m certainly capable of making mistakes, after that I gotta say I found those comments amusing. Your lack of insight (and perspective taking) is notable and from what you’ve displayed rather intractable/intransigent.
Obviously time to move on. Thanks for introducing an interesting subject.
Start with checking the relationship between weight loss and survival. You can also look at the severity of GI distress symptoms.
That is data already being collected so is cheap and easy. If designing a study from scratch you would want better data. Like actually monitor the size of the tumors and blood/urine/stool markers of nutritional state. Eg, ketone bodies.
As I said, it is baffling no one ever checks this. To my knowledge it has never been published for any cancer clinical trial.
Anoneuoid –
> Start with checking the relationship between weight loss and survival. You can also look at the severity of GI distress symptoms.
That would be interesting information as one part of an analysis. It would nonetheless have quite limited utility for, as I said, analyzing the effect of caloric restriction on cancer progression – given the many confounds, not the least of which is the chemo. Or of differentiating the effect of caloric restriction or nauseation from the chemo intervention from the other effects of the chemo intervention (as you claim to have insight into to make your assertions).
You’re still just arm-waving. Clearly, you aren’t going to address the actual issues. But you’re entitled.
> ~25% of our healthcare spending occurs in the last 3 years of life.
When we do cross-national comparisons of the relationship of dollars spent on healthcare and outcomes like life expectancy, to me this seems like a critical variable to consider.
For example, a lot of spending across many people where it brings little quality of life (perhaps even a negative return there given prolonged life during chemotherapy) but a lot of people living a little bit longer could skew this comparison. So even though life expectancy in the US is relatively low, a more nuanced analysis might show even lower returns in that so much money is spent for little gain in quality of life during an increased life expectancy.
I could be wrong but sense is that in a number of other countries people are more wiling to allow very ill people to die. If that’s so, the crappy US system might actually be even crappier than it looks just by comparing dollars spent against life expectancy.
I suppose another factor would be the amount spent on a relatively small number of people who are chronically very ill – largely because ironically they can’t afford to get good quality preventative care.
I wonder also how this would look in a cross-national comparison. Is the large number of dollars spent on a relatively few number of high need patients a more common feature in the US? And if so, is that attributable to our pay to play system or to larger societal factors that result in poor preventative care/unhealthy lifestyle tendencies. (Of course the causality there is multi-directional and multi-factorial and very complicated.)
“in a number of other countries people are more wiling to allow very ill people to die. ”
If I’m not mistaken, “willing” has nothing to do with it. I’m pretty sure that countries with national systems like Canada and the UK are capacity limited – that is, there is *always* a waiting list with people in varying levels of need. So for a given patient, when a treatment has a very low chance of success, it’s simply not offered to the patient, because there are other patients waiting with higher probability of success. Outside of walk-in clinics visits, everything takes four or five times as long as in the US. This becomes a restriction on treatment because the system can’t keep filling up with longer and longer wait times, so physicians have to raise the bar for what constitutes “necessity” for any given treatment.
Canada has horrible wait times. Half a year. In some cases it can be much longer because sometimes in a given region there is only one specialist for a given procedure, and they may do others as well.
That’s from the Fraser Institute. They may be grinding an axe there. May,
https://en.wikipedia.org/wiki/Fraser_Institute
Here’s what Wikipedia says about health care in Canada.
“Canada has performed close to, or above the average on the majority of OECD health indicators since the early 2000s,[14] and consistently ranks above the average on OECD indicators for wait-times and access to care, while receiving average scores for quality of care and use of resources.”
https://en.wikipedia.org/wiki/Healthcare_in_Canada
David is quite right. The Canada wait time issue is a favorite for libertarians who detest public provision of health care services. It is true that wait times in Canada are very long for some services, but not for others. Good data is hard to come by and politicized think tanks are not the best source: I’d suggest using a source such as https://www.oecd-ilibrary.org/sites/242e3c8c-en/1/3/2/index.html?itemId=/content/publication/242e3c8c-en&_csp_=e90031be7ce6b03025f09a0c506286b0&itemIGO=oecd&itemContentType=book. The picture varies a lot depending on what kind of services you are talking about. Discretionary services tend to have long wait times in Canada – emergency services not so much. I think it is fair to say that Canada exhibits a somewhat sensible sort of prioritization of care where essential services are provided quickly, at the expense of longer wait times for nonessential services. Of course, the definitions of “essential” and “nonessential” are subject to debate. I think it is also important to not focus on averages – wait times vary considerably across groups. I suspect that the US has greater variability in wait times across demographic groups than does Canada – but I haven’t seen any data on that, so I am prepared to be educated.
Gents:
I’m aware that the wait time issue is a partisan issue. The reason it is a partisan issue is because it’s real and lots of people – lots of Canadians – identify with it. It’s in the news in Canada frequently because it’s a real issue.
It’s not appropriate to compare average wait times in Canada to the US. In making that comparison, you’re comparing mandatory wait times in Canada to optional wait times in the US. In the Canadian system, you don’t have the option to travel from your home in Halifax to a hospital in Quebec City to have a surgery because there is spare capacity in QC. In the US you can go wherever you want if you’re willing to pay, and in many cases with insurance as well. Most people don’t, but you can.
Neither of you challenged the point I made that care is withdrawn or not offered in Canada when the patient has lower odds of success than the next patient in line. This is a necessity in a capacity / dollar limited system. I agree it’s a sensible thing to do from just about every perspective. However, when viewing the US system from a financial perspective, you have to take into account that Americans have the option to spend the money for treatment under these conditions and often do spend it, where Canadians simply don’t have the option. How much would Canadians spend if they could?
Also note that (sans publicly funded plans) Americans voluntarily enter into system restrictions when they choose a health care plan. Canadians simply have no option but to use their local services, however good or shitty they may be.
Chipmunk –
As other commenters seem to suggest, I think your response amounts to your fairly typical banging on a political agenda drum. Obviously, comparative costs and benefits off different healthcare systems can’t be reduced to a single dimension analysis, and neither can decisions and attitudes related to end of life care.
But…. your comment did send me into some Googling which turned up quite a bit of interesting material.
My original comment about cross-national comparisons of end of life care was mostly based on ignorance but also on (1) reading articles about people traveling to other countries to get legalized physician assistant suicide and (2) reading about how in Sweden early on in the pandemic they largely just didn’t treat COVID infections in very ill, very old people in congregate living facilities because the cost (e.g., spreading infections) wouldn’t be balanced by the benefit. I thought that interesting because such a policy would never fly among the people in the US who were fawning over Sweden’s COVID policies.
But after doing some reading (I could provide some links of you’re interested) I see that the picture is more complicated than I thought. One comparison between Japan, the US, Italy, and Brazil showed surprisingly few differences (with Brazil a bit of an outlier as being less prone to palliative care). Another suggested that Western Europe is more receptive to palliative care than either Southern Europe or Scandinavia (with levels of religiosity playing a role).
But the whole issue seems really complicated. Attitudes seem to be affected by access to care, and of course also, the embrace or lack thereof of palliative care in the medical community. One thing that’s interesting is that it seems that almost universally many people are open to palliative care but don’t hear much about it from doctors.
It seems that maybe the UK and the Netherlands stand out a bit but it’s not clear to me now that overall the US is distinguished by a comparatively more conservative gestalt regarding palliative care or what they call “passive euthanasia.”
Anyway, just a brief summary of a cursory Googlkng from someone now slightly less than completely uninformed.
Thanks for providing the impetus
Dale:
your point about “essential” vs “non-essential” is the real rub. You say:
“Canada exhibits a somewhat sensible sort of prioritization of care where essential services are provided quickly, at the expense of longer wait times for nonessential services. ”
If you had began your statement with this qualifier: “For a resource-limited system…” I would have agreed with your statement. Limited resources have to be doled out in a way that gets the most bang for the buck. However, the US *does not* have a resource limited system, and that’s why typically “non-essential” (e.g, you won’t die this year if you don’t get the service immediately) services are readily available in the US but not in Canada.
One thing I find interesting: it’s very difficult to find data on how many Canadians travel abroad for health care services. I don’t think it’s a huge amount, but it does seem to be significant. For all public systems, private systems are important safety valves for “non essential” services that help keep the demand down.
And FWIW, a quick Google suggests that while on the whole Canadians rank their healthcare considerably lower than Americans do on an “access” scale, overall their approval for their healthcare system is much higher than among Americans.
Make of that what you will.
> However, the US *does not* have a resource limited system,…
What?
The US has a highly rationed healthcare system with the rationing criteria being, to varying degrees and in varying ways, ability to pay.
I suppose you might mean something else by “resource limited system” but I can’t see any reaonable way to portray US healthcare as being non-resource limited.
The US does have a resource constrained system – all countries do. Also, while it would be interesting to know how many Canadians seek health care in other countries, I’d also like to know how many Americans do so. It is not unheard of for Americans to be medical tourists.
> When we do cross-national comparisons of the relationship of dollars spent on healthcare and outcomes like life expectancy, to me this seems like a critical variable to consider
The available evidence indicates health spending is similarly concentrated throughout the developed world and that this dimension is not obviously cross-sectionally related to the level of national health spending. About 5 to 10% of total health spending gets spent on the last (calendar) year of life amongst high-income countries, and the United States doesn’t particularly stand out.
https://randomcriticalanalysis.com/2017/04/15/some-useful-data-on-the-dispersion-characteristics-of-us-health-expenditures/
Another issue with this type of analysis for identifying “waste” is that we only have a very foggy notion of when someone is apt to die ex ante.
https://www.science.org/doi/10.1126/science.aar5045
More generally, the ex post end-of-life populations and high-cost populations only partially intersect. Most end-of-life are high cost, but most high cost are not end-of-life.
When I was a grad student in the US, I needed some minor dental surgery that was not an emergency, but it was not something I could delay indefinitely either. My insurance did not cover it.
I called a specialist and asked for the price (of a specific treatment — I consulted with my dentist back home and he explained what he would do). They acted as if I wanted to know something that is 1. inherently unknowable, because it depends on my insurance, and 2. is of no concern to me, because it is covered by insurance. I pursued the issue and got a “list price”, which was crazy. Then a day later I got a call from them saying that they can “adjust” this price for me. The kind lady on the phone asked about my financial circumstances, and when I explained I was a grad student, they ended up with about 15% of the original price they first quoted after a couple of phone calls.
It was still more though than the price of a roundtrip ticket to Europe purchased at a short notice, so I decided to do it there (where I was insured still as a citizen). But I found the whole experience very weird.
The health share vs. income is dimensionally odd, since the x correlates closely with the denominator for the y. Fitting the graph to a quadratic is equivalent to fitting health costs to a cubic.
Sort of. A cubic that has no constant term.
Assuming the incomes are calculated in terms of Purchasing power parity, it’s really a regression of two dimensionless numbers. The income is (total income in local currency / cost to purchase a given basket of goods in local currency) * cost to purchase the basket of goods in US
That final US dollar scale factor is a constant across all data points so you can think of it as ignorable.
If you design the basket of goods to be a basket of monthly fixed costs, then you’d gain a lot of insight by the size of the x axis. Suppose the basket is rent, food, utilities, taxes, and transportation to work. Then when income is about 1 you are living in borderline poverty. Much below one and you must be skimping on shelter, food, or utilities etc. People in that situation probably have little to spend on healthcare. On the other hand if healthcare is govt provided, then we should count that as both income and expenses. It’s not “free” it’s equivalent to a govt check that is spent on healthcare only. As people’s income goes much above 1 they can afford “luxuries” and that would include a bigger amount of healthcare. But in the US where you have lots of broken incentives and essentially no market for healthcare (pricing is hardly transparent, there isn’t any ability to know the price of what you’re buying) you would expect healthcare to extract large rents. And they do. And that is a net negative on US longevity. There’s nothing much mysterious here. It’s the same broken economics throughout the US economy. Monopolies, oligopolies, rent-creating regulatory structures, manipulation of money supply, subsidies, patents, copyrights, lawsuit trolls, etc. Everywhere you turn in the US individuals are being fleeced by powerful organizations that have winnowed their way into a deeply entrenched rent extraction regime.
I’ve lived in Spain and the US each for 15+ years as an adult. The main difference between their health care setups is that one is primarily public and one is primarily private. This matter in part because of the different incentives the different setups generate.
The general observation is that in a public system medical practitioners are incentivized to avoid prescribing potentially unnecessary treatments, particularly those of high cost, while in a private system they are incentivized to prescribe some treatment, particularly if it is high cost. Money is made by prescribing expensive drugs and treatments. Money is saved by avoiding them.
So when a medical practitioner faces the same patient in the two settings the decision is different. Basically public systems incentivize rationed use of expensive treatments while private systems incentivize liberal use of expensive treatments. Since the benefits of many of these expensive treatments are not great relative to their costs, this could be part of the mechanism underlying the diminishing returns.
Thank you. I appreciate the link and your kind words.
As I see it, the conventional wisdom is that U.S. healthcare is *uniquely* broken. That is, we spend much more than the typical country *because* of some distinct features of our healthcare system, and we also get much less in return, presumably primarily because of prices. The implication is that the rest of the world has healthcare well figured out and that if only we modeled our healthcare system like any other, we would cut spending in half and get much better outcomes.
—
This NY Times article is representative.
https://www.nytimes.com/2018/01/02/upshot/us-health-care-expensive-country-comparison.html
“The United States spends almost twice as much on health care, as a percentage of its economy, as other advanced industrialized countries….a few decades ago American health care spending was much closer to that of peer nations. A large part of [what happened] can be found in the title of a 2003 paper … ‘It’s the prices, stupid.’ The study… found that people in the United States typically use about the same amount of health care as people in other wealthy countries do, but pay a lot more for it. Ashish Jha…said. ‘The U.S. just isn’t that different from other developed countries in how much health care we use. It is very different in how much we pay for it.'”
—
My perspective is that the U.S. healthcare system is not uniquely broken. Our higher spending is almost entirely a product of our higher income levels. There’s always latent demand for more and better care. What differentiates rich countries is that they are more willing and able to pay for it. Our higher marginal spending might, arguendo, be so wasteful as to be not worth the added cost. However, changing our system is unlikely to change patient expectations and, more importantly, our willingness to pay as a society through various third-party payment schemes. Regardless of the system, we will likely spend approximately the same amount in the long run, conditional on income levels.
I would go further. It’s not just the total level of spending. Many (not all) of the things people believe to be unique byproducts of our system, whether its (well-measured) prices, utilization, or intensity, are pretty well explained by our high average income levels or, at least, income levels explain most impactful things, and the U.S. is rarely the outlier many people assume it to be. The vast majority of the alleged waste and diminishing returns to spending exist on a continuum. The U.S. is further along this frontier due to its income, and other countries will likely continue to move in this direction as their income allows.
Most of the action is in the intensity of care. If reformers genuinely want to reduce spending by much, they must sell the public the idea that so much high-intensity medicine isn’t worth the cost. Though such discussions occasionally make their way into scholarly literature, I see little discussion of this sort in the mainstream media, social media, or popular blogs to explain spending. Instead, we tend to get repetitions of “it’s the prices, stupid” or allusions to nebulous “waste,” which is bad by definition, but good luck getting much agreement on quantitively significant specifics.
***
While I still maintain that healthcare is subject to rapidly diminishing returns vis-a-vis life expectancy and that this does much to explain America’s declining status in rankings (less affluent countries have much more opportunity for big gains and thus convergence), it doesn’t follow that spending more than the median high-income country is necessarily bad. For one thing, patients are not only interested in avoiding mortality. Attributes like mental health, faster recovery times, lower rates of complications, aesthetics, quality of life (chronic diseases), and more are likely highly valued by patients, even if these things don’t move the needle on life expectancy.
Likewise, we are likely to care about the healthcare experience of patients and their families above and beyond mortality effects. Waiting weeks or months to get an MRI, see a specialist, or what have you may not affect national mortality rates much, but it’s still very unpleasant when you need or want medical attention. Is it unreasonable to suggest that paying a few more percentage points is worth it to avoid the indignities and frustrations associated with the rationing typified by the United Kingdom’s NHS? I doubt I am alone in being willing to pay a premium to minimize such issues substantially.
We don’t typically approach other areas of life in such a unidimensional Green eyeshades fashion. I doubt we are even willing to do this for healthcare in practice. I do not mean to suggest that we should treat healthcare decisions as entirely unquantifiable. Instead, a more realistic model would recognize that people care about other dimensions of healthcare besides just its effects on mortality. Poor countries may have little choice but to target the lowest cost and most efficacious treatments (e.g., antibiotics). Still, as wealth increases, they will increasingly relax fiscal constraints, accept more modest life-extending effects, and ultimately pursue remedies that improve life in other subtler ways.
In addition, the life expectancy ~ health expenditure slope amongst more developed countries is probably partially confounded by some of the adverse effects that are *also* associated with higher income levels (e.g., obesity/diabetes, drugs, and automobiles). Though the life expectancy returns to spending undoubtedly diminish, the average causal effect of marginal expenditures may be greater than zero, even if the observed slope is zero or even negative. Higher-income countries still do many things apt to extend life at much higher relative rates, such as transplant surgeries or pacemaker & ICD implantation. However, such interventions are costly at present and affect a smaller fraction of the population than some earlier interventions. That countries with lower baseline rates of cardio-metabolic disease (substantially “Western” diseases of affluence) may enjoy lower rates of deaths associated with the disorders treated by such technological innovations despite employing much fewer technological innovations does not imply that such technology does not work. (Indeed, it’s probable that the gradual improvement and diffusion of these innovations at and from the frontier drive a substantial fraction of improvement in health outcomes globally — large spillover effects).
Above all, I take the effects of differences in national income very seriously. Not only is national income likely to be the dominant (likely causal) determinant of national health spending (mainly via a willingness to pay), but the relative price of goods and services changes sharply as a function of income levels. The same income share buys considerably more of most goods and services (particularly goods) in the United States. The high-income elasticity of health spending does not imply that health care will displace everything else in *real* terms. Such high levels of health spending might be suboptimal, but it’s not the disaster many assume it to be.
Moreover, as income rises, the marginal utility of most other expenditures is also likely to diminish, and we are likely to place a higher monetary value on extending each quality-adjusted life year. We face a different set of tradeoffs from those that we faced decades earlier and different from the average OECD country. The forgone marginal utility of other forms of consumption may not necessarily exceed the utility we derive from our marginal health expenditures (when viewed more comprehensively).
Admittedly, I am playing devil’s advocate here to some extent. The dominant role of third-party payment in healthcare doubtless complicates the assessment of tradeoffs. Spending patterns would be quite different if people had to pay out of pocket, even if we were to issue patients comparable lump sums to dispose of as they pleased (I’d wager many lower-income households, in particular, would pocket the money even when it’s to the detriment of their health).
https://marginalrevolution.com/marginalrevolution/2017/07/medicaid-isnt-worth-cost.html
These spending patterns may, in some sense, be irrational. Nonetheless, they still have ultimate material (economic) causes. We don’t have to like it or agree with them for views on other dimensions to contribute proximally. What people believe and value matters. To think we can simply technocrat our way out of such issues as a society is delusional. The decisions of technocrats are unlikely to remain insulated from the beliefs and values of voters, politicians, and other interested stakeholders in the long run, particularly over what are perceived to be high-stakes, high-visibility decisions, particularly within democracies like ours. Medicare, for example, is heavily socialized with direct federal control, yet it offers extremely generous coverage by international standards, and it indeed spends much more per capita on an age-adjusted basis than the rest of the OECD. People who want to explain our spending habits by appealing to its marginally higher degree of privatization must grapple with this.
***
The pursuit of radical, ultimately unrealistic reform has caused a lot of unnecessary grief and strife in our society. Cutting spending in half through top-down socialized care and dramatically cutting the average price of healthcare is a lost cause. The same is true for some on the right who believe we could solve high spending by radically shifting to an out-of-pocket HSA model. Neither approach takes people’s baseline attitudes and income effects seriously enough. Perhaps more importantly, it’s gotten in the way of more politically feasible and ultimately more realistic solutions to solvable problems.
As I see it, the main issues we have today relate to people falling through the cracks in the various systems, particularly among those not covered by the three major systems, i.e., employer-sponsored plans, Medicare, and Medicaid. The uncertainties and vagaries around reimbursement are primarily issues for the uninsured or poorly insured, who are at least nominally exposed to the billed/list prices, which reflect neither costs nor the average price paid. Tackling these narrower problems doesn’t require top-down remodeling of the entire healthcare system aimed at solving much larger perceived problems. I do not believe health outcomes or total spending hinge very much on this, as such people tend to be relatively young and relatively healthy, and most of their care gets provided one way or another. Still, we could reduce the stress and uncertainties associated with large unexpected bills through more modest reforms with much more limited goals.
***
I take your point about the “all else equal.” Some extrapolation and inferences are undoubtedly required, but when isn’t this true when making judgments at this level of analysis? It cuts both ways here. We cannot know for absolute certain that the U.S. would sustainably spend much less without actually doing it, likewise, for changing our system and health outcomes. The quality of such inferences isn’t likely better when discussing issues of specific countries rather than cross-cutting factors like this. Though there is good reason to believe the U.S. is substantially different from other Western countries in other dimensions besides our healthcare system(s), many, nonetheless, have no problem asserting we are one technocratic fix away from “better” results.
Insofar as what other countries will spend when they have incomes comparable to ours today, it substantially depends on whether I am correct about the overriding role of national income. Little is absolutely certain in life, but my argument is considerably stronger. In horse races between objective indicators of need (e.g., age structure, chronic disease, etc.) and ability to pay (well-measured income), the ability to pay wins hands down every time. Rich countries spend a much larger share of income, which is largely determined by the quantity of care provided, primarily via higher intensity. It’s difficult to explain the vast preponderance of the evidence without positing income is the major causal factor and that health spending is highly elastic at the national level, particularly parsimoniously. Though many assume otherwise, it’s also difficult to draw a quantitatively significant bright line between the behaviors of the U.S. health system and other countries that don’t ultimately track income.
People argued that the United States was terribly out of control decades ago, yet many OECD countries are spending more than we did then. Many behaviors people decried decades earlier, such as high MRI usage, have also been surpassed by many other countries in recent years. Yes, prediction is very difficult, especially if it’s about the future. I’d still gladly wager on this. I predict (1) health spending will continue tracking with income levels for other wealthy countries — much like it has for the U.S. — income growth may decline, and thus health spending growth rates may continue to slow, but the elasticity isn’t likely to decline, and (2) the life expectancy ~ health spending slope will continue to decline within the OECD. The notion that this “disconnect” is a particularly American pattern will become increasingly untenable for those paying attention.
I also predict countries that have historically had conditionally good outcomes, like South Korea, Japan, and Israel, will increasingly enjoy longer life expectancies than high-income Western Europe. Idiosyncratic differences that already contribute to outcomes will increasingly rise to the fore as marginal differences in healthcare contribute proportionally less. Little of this will be convincingly explained by some healthcare systems spending their resources more effectively than others, conditional on spending.
It is hard to respond to such a concise diatribe. I don’t disagree with much of what you say, but one component I think needs to at least be challenged. You claim that Medicare “indeed spends much more per capita on an age-adjusted basis than the rest of the OECD.” It is (very) hard to come by good data, but my impression is that Medicare prices are far lower than comparable international prices – indeed, it is the below-market Medicare reimbursement rates that make it a poor model to use for estimating what a single payer system would cost in the US. I chose MRI prices as they are somewhat less complicated than other services, though they are still not straightforward.
In any case, I found 2017 data (https://healthcostinstitute.org/hcci-research/international-comparisons-of-health-care-prices-2017-ifhp-survey) showing an average outpatient MRI price of $1430 in the US with most international prices of $500 or less. I found a 2018 Medicare price for an MRI of the lumbar spine of $229 or $386 for the brain stem (https://www.brookings.edu/wp-content/uploads/2021/03/Imaging_Paper_Final.pdf). While there are additional complications for making these comparisons and this is just one selected service, from my experience this is typical for the Medicare fee schedule – it is typically 10%-20% of the commercial prices, often lower than average international prices (being cross subsidized by private insurance prices in the US). So, I question your claim regarding Medicare spending relative to the OECD. If my price comparison is anywhere close to accurate, then it is possible that the US just does a lot more MRIs than other countries – but while the US appears higher than average, it doesn’t seem to be unduly so (https://data.oecd.org/healthcare/magnetic-resonance-imaging-mri-exams.htm).
As is common with some of your past comments, the sheer volume is somewhat overwhelming – but I’m not convinced of the accuracy.
I see no tension between the position that Medicare prices are relatively low and that the spending on the Medicare age group is high by international standards. Indeed, I have long argued that prices explain little and that rich countries spend more mostly because they treat much more intensely. This is particularly true for Medicare, which combines relatively low prices with high intensity. Though there’s probably something to cross-subsidization, Medicare (FFS) can, in part, get away with substantially lower reimbursement rates and its low administrative overhead because it is exceptionally liberal and mechanistic in what it pays for and under what conditions.
If one accepts the premise that national health spending is determined by what we are willing to spend (ultimately a function of mean income levels), there’s little to no reason to expect modest variation in price levels to independently predict the overall expenditure level in the long run. We can expect prices to correlate with spending because rich countries combine higher real healthcare consumption with higher prices (the productivity gains typically enjoyed by high-productivity countries in labor-intensive services like healthcare aren’t quite enough to eliminate the effects of the higher average wage). However, healthcare prices predict little conditional on income levels.
> You claim that Medicare “indeed spends much more per capita on
> an age-adjusted basis than the rest of the OECD.”
> It is (very) hard to come by good data
Though some cognitive dissonance may exist here, this assertion is not controversial. It’s self-evident if you follow the data closely, but I know several published estimates indicate as much.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7411536/
“the Medicare-eligible population in the US still spent 100% more per capita on health care than older adults spent in the 7 comparator countries…Moreover, in absolute dollar amounts, not ratios, this spending gap actually increased to be the largest among individuals 65 years and older, with the typical person in the US spending nearly $18 600 more at approximately age 80 years than the typical person in these other high-income countries.”
You might also consider that Medicare and Medicaid spent considerably more per head than private health insurance plans, about 2.3 and 1.5 times more in 2019, respectively. Yes, the beneficiaries are older and sicker, but collectively these major public plans account for about 25% more spending nationally than private plans. Private health insurance can only explain so much of the excess spending.
> my impression is that Medicare prices are far lower
> than comparable international prices
In absolute terms? I rather doubt that. They may be somewhat lower than the average price paid in some wealthy countries, conditional on income, but public prices tend to be lower than private prices.
> Medicare fee schedule – it is typically 10%-20%
> of the commercial prices
That seems like a large overestimate. Conventional estimates indicate that Medicare FFS prices for hospital services are roughly 50% less on average and roughly 30% less for professional services. The weighted average price reduction appears to hover around 30%.
> it is possible that the US just does a lot more MRIs
> than other countries
My point was *not* that MRI usage arithmetically explains (much) US spending today, but behaviors that people just assumed to be highly idiosyncratic features of our system turned out not to be. Many countries are using MRIs more than we once did, and some are even using them more than we do now.
My larger perspective is that technological innovation in healthcare is the primary proximate cause of excess cost growth. The usage of such innovations tends to be highly income elastic at the national level, particularly at the outset. Still, the latent demand for specific technologies gets sated at some point as prices decline and national income levels increase throughout most of the world to make them more broadly affordable. Whereas MRIs were once highly predictive of spending within the OECD, it’s less and less predictive, which is to be expected as MRI technology is nearly 40 years old now. Newer, more expensive technological innovations are where most of the action is to be found today.
https://randomcriticalanalysis.com/why-conventional-wisdom-on-health-care-is-wrong-a-primer/#rcatoc-better-technological-indicators-are-required-to-fully-capture-rising-volumes-through-proxy-measures
I will modify my previous comments. My experience with Medicare is that the Medicare fee schedule is indeed 10%-20% of the list prices for the services. Since I don’t have access to the private insurance prices, I have no way to comment on that comparison. I did find one study that looked at that: https://www.kff.org/medicare/issue-brief/how-much-more-than-medicare-do-private-insurers-pay-a-review-of-the-literature/. It appears that the Medicare prices are around double those of private insurance (according to that study and recognizing that there was a large variability in the comparison).
The underlying question of importance is the degree to which the higher US per capita spending is driven by higher usage and/or higher prices. I don’t think the higher prices can be ignored, nor do I think they are caused by higher incomes in the US. The higher usage levels I would agree are largely due to income differences – although a closer examination would need to look at the relative usage of different treatments and services in the US relative to other countries. I suspect the picture varies a lot, depending on whether we are talking about drugs, hospital services, long-term care, diagnostic services, outpatient care, etc. Other countries usage of these categories is largely determined by government policy – in the US, it varies considerably across demographic and socioeconomic groups (even within Medicare).
I’m still not convinced of your central argument – that the US is not really distinctive, other than what would be expected due to income differences between the US and other countries. Consider the counterfactual: if incomes in France (or substitute any other country) were equal to those in the US, would spending per person be comparable to that in the US? It seems that you are arguing that the answer is largely “yes.” I’m not so convinced, since the large differences in how the systems work (particularly related to market pricing, services approved, and private options) might still account for large differences. Let’s suppose that the US does many more expensive hip replacements (I’m not sure this is a good example, but let’s suppose it is) than they do in the UK. If the UK’s income was higher, then the question is whether they would then approve a similar frequency of treatment to that in the US. I accept that this would possibly be true, but it is also possible that the centralized NHS would ration expensive treatments according to some model of QALYs while in the US it would still depend largely on what sort of insurance an individual has, where they live, their educational and economic background, etc. In other words, I’m still unsure how much of the difference is due to income levels and how much can be attributed to market structure and policy.
Dale said:
“would spending per person be comparable to that in the US? ”
The French – and pretty much all of western Europe – won’t even spend to defend Ukraine and Europe from Russia! They just let the US do it, and we’re stupid enough to keep doing it.
It’s amazing this discussion is still going. It shows how useless data is in many situations – the data doesn’t matter, it’s whether people are willing to accept it’s meaning that matters, and you’re obviously not willing to accept the pretty obvious reality that Americans spend more on health care because they have more to spend.
But, again, the last thing a data analyst wants is to solve a problem and thereby end the stream of cash upon which he depends to survive. You know the old saw that Phil loves so much:
“It Is Difficult to Get a Man to Understand Something When His Salary Depends Upon His Not Understanding It”
chipmunk
What about this “obvious reality?” The US has higher income than all other countries, yet the cost of mobile communications services (https://datahub.itu.int/data/?e=USA&i=34619 shows data for the high usage basket – data for low usage basket is also available and shows similar patterns, with the US lower than most countries) is lower than most countries, although higher than most countries in Western Europe. Clearly income matters, but so does market structure and regulation. Similarly, my point is that these things matter for medical care as well – income is part of the explanation, but not all of it. Random Critical Analysis seems to suggest it is virtually all due to income. I accept that income is part of it, I’m just not convinced how much.
@Dale Lehman
> Medicare prices are around [half]
For hospital services. The private premium for professional services is considerably smaller, and private insurance is less heavily weighted toward hospital services. Plausible estimates place the weighted average price savings for Medicare FFS at about 30% less (comparing prices paid for the privately insured just before eligibility–it’s probably even less for younger people). Of course, as you indicated, Medicare prices are unlikely to scale nationally due to cross-subsidization, so this is likely a significant overestimate of potential national price reductions. These lower prices also come with tradeoffs, which make them even less generalizable to net cost reductions.
Some back-of-the-napkin math might be instructive. Commercial insurance is less than half of health spending nationally (45% in 2019 and decreasing). .3 * .45 implies roughly 13% of national health spending might be saved if private health insurance paid the same rates as Medicare, assuming those prices can scale and that intensity isn’t adjusted upward. That’s not much, and the actual effects are likely smaller still. Remember that it’s already well established that Medicare spends roughly twice as much as other high-income countries for the same age group and that the ratio for the predominantly privately insured (non-elderly) age group isn’t much different.
> nor do I think they are caused by higher incomes in the US
Health prices are unmistakably positively correlated with mean income at multiple levels of analysis. We should expect the overall price of healthcare to be higher in the United States because it is labor intensive, and real wages are substantially higher in the United States (not just in healthcare!). The best estimates of international healthcare prices, i.e., those produced by OECD-affiliated researchers working on health PPPs with the cooperation of national health authorities, indicate the United States is about where we would expect it to be.
https://randomcriticalanalysis.com/2020/02/13/its-still-not-the-health-care-prices-2017-edition/
However, well-measured healthcare prices, both domestic time series and international price indexes, rise more slowly than income levels, likely because of modest productivity gains in healthcare amongst higher-income (more productive) economies. If the quantity of healthcare per capita remained roughly constant, the price statistics imply that the health share of income should have declined with income growth.
Healthcare prices may contribute to excess spending growth insofar as productivity gains cannot keep up with manufactured goods or even agriculture without some fundamental technological leap (e.g., AI + robotics). In an alternate universe where the overall productivity growth in healthcare resembled Moore’s Law, it’s entirely plausible that the expenditure trajectory would be much flatter. However, the productivity gains and changes in price levels observed in healthcare, both inside and outside of the United States, are nowhere near that level (other services also look pretty similar). Thus healthcare remains a luxury good at the national level, wherein we spend what we think we can afford to spend as there is always latent demand for more and better healthcare. Variation in healthcare prices is unlikely to factor appreciably into the level of expenditure in the long run without some massively disruptive technology (We are still a long way away from AI having this sort of impact on healthcare IMO).
> Other country’s usage of these categories is largely determined by government policy
All countries, including the United States, have policies restricting healthcare utilization/intensity (it’s mostly intensity). However, said “policy” isn’t independent of income levels in the long run. Even countries particularly intent on aggressively rationing care ultimately respond very similarly to changes in their economic circumstances. A technocratic regime might optimistically hope to have some long-lasting effect on the intercept. Still, the slopes are very similar, particularly allowing for mean reversion (towards the expected level of spending at a given income level).
> If the UK’s income was higher
The UK’s NHS is one of the few systems that has credibly sustained significant cost reductions *conditional* on well-measured income levels for a few decades. The UK is much more plausibly positioned as an outlier amongst high-income countries, albeit on the other end, than the United States.
https://randomcriticalanalysis.com/misc/the_uk_is_the_outlier/
However, the establishment of the NHS in the United Kingdom was historically contingent. It was developed in the aftermath of WWII in the context of broader economic rationing that started during the war and persisted long after. Most other developed countries didn’t move toward “universal” coverage until well after WWII. This scarcity mindset never had the same purchase outside the United Kindom, and it certainly hasn’t stuck.
The NHS cost savings are very much dependent on aggressive rationing. Its days as an aggressive rationer are probably numbered. The NHS is still notionally popular with many in the UK, but its extensive rationing is not. Spending has been increasing and has tended to converge closer to expected values over the past few decades, yet there were growing cries of “austerity” even before covid hit. Although the UK may still have its “NHS” in the coming decades, if it’s still around, it will likely loosen up the purse strings considerably, increasing overall spending and willingness to pay for more intensive technological innovations. The combination of population aging (much less spare capacity) and demographic change (fewer people grew up during or in households that lived through the war) is particularly likely to do it sooner rather than later.
Its long-term political viability aside, I am not convinced the savings the NHS has tended to eke out with aggressive rationing have been worth the squeeze. The act of rationing itself has costs (e.g., delays, travel time, etc), as does the actual aims of rationing, i.e., a non-trivial reduction in access to technological innovation and likely less innovation in the long run. The ~20% savings the NHS has plausibly rung out ought to be balanced against everything that comes with that cost containment regime.
I’m sorry but if you honestly believe this you are hopelessly uninformed. You would think that it would be obvious to anyone with a passing knowledge of Economics that our lack of single payer is why we pay so much more than every other country. To literally quote Investopedia:
“A monopsony is a market condition in which there is only one buyer. Because there is only one buyer for a good or service, the buyer sets the demand, and therefore, controls the price. Monopsonies, like monopolies, are inefficient to a free market, where supply and demand regulate prices to be fair for consumers.”
Common sense would dictate that a country concerned with the wellbeing of its citizens would seek to lower the costs of non-discretionary spending like healthcare, instead of forcing them to walk for miles and /or wait in lines for days for the annual free clinic to roll through town. https://www.theguardian.com/news/2016/nov/23/enormous-pop-up-clinic-trying-bridge-americas-health-divide
Of course, it would take a uniquely evil country with a complexly bought off corrupt political class to preemptively tie its own hands and refuse to negotiate lower drug prices against the wishes of 92% of the public. https://www.brookings.edu/articles/government-regulated-or-negotiated-drug-prices-key-design-considerations/#:~:text=Unlike%20other%20nations%2C%20the%20U.S.,prices%20significantly%20requires%20government%20intervention.
So, do we just love free markets so much that we are willing to pay more for healthcare? Nope. We have all too happily let, and even encouraged, merger after merge in the health sector.
https://www.thebignewsletter.com/p/how-obamacare-created-big-medicine?utm_source=post-email-title&publication_id=11524&post_id=113116212&isFreemail=false&utm_medium=email
“Richard G. Frank and Conrad Milhaupt from Brookings noted this trend last month. Payers could buy providers, and then send revenue to related businesses. The two firms leading the charge here were UnitedHealth Group (UGH) and CVS. UHG, one of the big four insurers, formed a subsidiary in 2011 right after the passage of Obamacare called Optum. Optum began rolling up physician’s practices, software and analytics firms, medical clinics, and pharmaceutical middlemen. In 2018, CVS, which owned large pharmacy chains and the PBM Caremark, bought the health insurer Aetna. Earlier this year, CVS completed its acquisition of Signify Health and Oak Street. In 2018, Cigna bought Express Scripts, the largest PBM in the country. Humana bought Kindred, a health care delivery firm. Elevance, formerly Anthem, became a PBM, and cut deals with a large number of medical providers.
In 2019, UHG sent 18% of its payer revenue to itself, while CVS’s Aetna sent 13% to its own clinics and pharmacies. That number has no doubt increased dramatically over the last three years.”
Results (same article):
“And yet, something was off. Obama had promised on the campaign trail that he would sign a universal health care bill into law, and one that would “cut the cost of a typical family’s premium by up to $2,500 a year.” In 2004, the average insured family of four paid $11,192 in health care costs, by 2022 that amount was $30,260. That increase in cost for a family of four is the price of a small car, every single year. And that’s because prices have gone up, and not because there are more doctors, beds, or care.”
And you know you are not getting a good deal when you are making that deal with Private Equity. Which has been going hard into buying up small practices and teaching hospitals. That is the rot in our system. Well, to be fair, if rot didn’t exist neither would this country.