Hacker News new | past | comments | ask | show | jobs | submit login
Link between health spending and life expectancy: US is an outlier (2017) (ourworldindata.org)
72 points by ColinWright 20 days ago | hide | past | web | favorite | 60 comments

A few things off the cuff as a physician-- the US has dangerously unequal access to care. I suspect most of the "drag" on life expectancy is from patients without access (this is conjecture but based on personal experience). Even patients eligible for medicaid don't necessarily go through the steps to get it. Often we see patients show up to the hospital who haven't seen docs in a many years until they've developed terrible sequelae like gangrene requiring an amputation. At that point, during admission, the hospital generally helps get them on medicaid. Also, obesity is becoming the new normal in US with all of its associated co-morbidities (a common morbidity/mortality sequence is obesity -> diabetes, hypertension -> coronary artery disease, end stage renal disease, general vascular disease -> death. More healthcare spending is required to treat patients, but the root cause for a growing part of the population's illnesses can't be cured by medicine yet. Also, economically speaking-- the US spends a lot more per capita on mostly everything-- its GDP per capita is higher than these other countries. But, mostly, I suspect the US's health mortality problem is access related. We're spending a lot of money on things that don't help people with poor access to care until their original diseases have done irreversible damage.

I agree with most of what you said, though I want to highlight something as a matter of looking at it from a different angle.

That the US has a higher GDP (and therefore spends more on most things that can't be imported) isn't divorced from the health policy of the US. It is the prioritization of GDP in the public policy sphere in the US that has, partially, led to the GDP being higher. For example, drug companies in the US make more from patients partially because there is no national collective bargaining program there. This gives them enough profit to meet the testing requirements to get into other countries around the world. There are many similar examples of this.

I'd also argue that more effective people in America are discouraged from working for the state due to cultural prestige differences with other western countries.

The French are healthier. Not as wealthy, but healthier. Healthcare spending is just one part of it. The other part is a culture that emphasizes other parts of life over raw income or cost. Nobody is tallying up the healthcare benefit of eating fresh veggies instead of a pizza pocket, but it is there, even if it is invisible in the numbers.

Do you have any insight into the reason for the unequal access? Is it that poorer or uninsured patients cannot afford non-emergency healthcare? Is it that they are too far away? Is it education/knowledge/social - belief that they shouldn't go to a doctor for "nothing". Are they "too busy".

Also, are there any studies to show that populations in other countries use their medical systems more frequently across the board, or that they use more frequent preventative visits?

You mention the problem with obesity and its co-morbidities. Are there any studies or plots that show health outcomes (lifespan, infant mortality, maternal mortality) as a function of % of population who are obese? I wonder if poor health is linearly correlated with obesity, without regard to healthcare spending...

It's common for baby boomers (at least the males) to refuse to the go to the doctore.

Refusing to go for checkups, screenings, even after mild heart attacks, refusal to take prescribed medications for blood pressure or diabetes...

> Do you have any insight into the reason for the unequal access? Is it that poorer or uninsured patients cannot afford non-emergency healthcare? Is it that they are too far away? Is it education/knowledge/social - belief that they shouldn't go to a doctor for "nothing". Are they "too busy".

I am not a Doctor observing patient populations, but I have lived and worked in East Palo Alto, CA for the last few years, which from my perspective, is a stark case study in all sorts of inequalities (For non-SV residents, East Palo Alto is a small historically socio-economically depressed city wedged between Palo Alto, Facebook, and Google. Zuck's house is a stone throw away from crack houses)

Anecdotally, what I have seen regarding poorer/uninsured people not utilizing health services involves all the things you have mentioned.

Non-emergency healthcare without insurance is pretty much out of the question. Even with insurance, co-pays can be prohibitively expensive. There are people I know living paycheck to paycheck with zero savings and like 50%+ of income going to housing expenses. Even a $50 unplanned-for copay is a painful blow.

Distance is also an issue for a population that may not have consistent access to a car--and the bus system takes too much time. If you are working a few jobs, it's nigh-impossible to take a random 1hr+ bus ride to a medical facility.

Education/knowledge is, unfortunately, also a huge issue. I have observed some astonishingly unhealthy habits that stem primarily from ignorance. Additionally, even navigating our health system requires some level of education, literacy, and time that often seems prohibitive for many in the population I have interacted with. And, with regards to distrust of the medical system, I have observed that as well. I have also seen this coupled with exploitative homeopathic "care" providers...which is doubly sad. These are more complex issues that I don't feel well-equipped to begin to speculate the underlying reasons for. Perhaps some of it is cultural. I am not sure if that is the primary story though.

Generalized ambient anxiety and depression is also, I believe, the more ultimate, albeit distal, cause here, rather these more proximate poverty mechanics. The emotional toll of living in poverty, primarily extreme _relative_ poverty (remember Zuck across the freeway and the swarms of Tesla commuting through your city on their way to high-paying tech jobs that are inaccessible to you) results in all sorts of extremely unhealthy compensatory behaviors. Heavy alcohol and drug use--even if just Marijuana, binge eating fast-food, violent communication styles that contribute to always-on fight or flight reactions, dangerous activities like racing cars or weaving bikes through traffic--perhaps done as some way to "feel something" or perhaps find identity in the only way available--acting out anti-social power displays as a way to reify self-worth contra a system that clearly treats them like shit on nearly all levels. And more.

Frankly, unequal access is both a moral failure and also, I believe, soon to be an instrumental failure for the strength of the United States. Having a large population of enfeebled, diseased people does not bode well for nice things we care about like innovation, improved quality of life, stability of our governance systems, or even things like the strength of our fighting force.

However you want to cut it, the issue of unequal health access is a shame.

I'd be curious to get your take on the Oregon Medicaid study, where a randomized subset of Medicaid-eligible patients were given access, and no significant effect on health outcomes between those who enrolled and those who did not was observed.

If the US has higher GDP per capita and pays more for everything per capita, then maybe the reason is not that the US is so exceptional, but that the US dollar is artificially stronger than it should be. And a reason for that might be the willingness of the US to wage war (economic war if not outright bombing and invasion) against anybody who tries to trade oil in anything but US dollars.

> but also in child mortality

Different countries have different definitions of infant mortality. When is it categorized as a miscarriage and when an infant death? The US tries very hard to save premature babies, and counts the failures as infant death rather than miscarriages.

I think you'll find that the US is not alone in trying that and that other countries simply don't limit that kind of efforts to the well insured elite. When visiting a hospital is a potentially bankrupting event, a lot of young uninsured mothers might end up not going to a hospital or going too late. Also, prenatal checks are not optional. The attitude towards abortion combined with religious conservatism also pushes a lot of young women to look for solutions outside the traditional healthcare system; which has all sorts of risks for the mother.

I find it ironic that the net result of pro-life policies is increased mortality rates for babies and mothers.

But even if they all tried equally hard, it's possible that there could be recording differences which throw the stats off. When exactly is an event counted as a miscarriage vs. an infant death, that's a paperwork question not a medical one.

At least I think that's what GP was saying. I don't actually know what rules are applied. It would be interesting to see a graph of miscarriages month-by-month as well as deaths month-by-month.

> I think you'll find that the US is not alone in trying that

It's all over the map, there are even societies where the baby isn't considered a person until some time after birth.

It's certainly enough to question the relevance of the statistics.

"it's all over the map, there are even societies where the baby isn't considered a person until some time after birth."

What societies? Some of the countries in the graph?

In Europe? Who?

Why are you restricting this to Europe?

If you look at doi 10.1257/pol.20140224 you can see that the main difference in mortality is not at day one but late(er) in the first year. So likely the main difference is not due to the distinction between miscarriage and infant death. This is also matched by the fact that children born to middle class parents do statistically as well as kids born to European parents. It is children of poor parents that have the inflated mortality.

I’ve always wanted to see this chart with the US broken down by state.

I’d like to see the E.U. as a whole vs US. It would also be interesting to see the States mapped against the countries in the article.

I'd also be interested in how Life Expectancy at Birth is calculated in advance.

It’s really just death rate at the time the newborn is born: if x% of the population died in 2016 and that never changed going forward, how long would the mean person born that year live (given that people born in earlier years would be dying at their rate, contributing to the formula.

So technically not actually predicting an expected lifespan (thermonuclear war could happen after all) but allows longitudinal and secular analysis.

ok, so infant mortality from that year doesn't effect the number?

(any more specifically than those deaths adding to the overall death rate)

Yes it does


I described the approach in my first comment: for 2018 (now it's over) look at the death rate and then apply estimate it uniformly going forward for all recorded births during the year until the surviving population goes to nil. The death rate is all the people who died, whether 1 jan or 31 dec regardless of the age of the deceased.

If you want greater resolution on a process like this you'll do better to search yourself than to reply on HN commenters like me.

I would imagine that there is maybe a few small states that does well, but it's unlikely to be one outlier that draws the entire chart down.

there could be a few large states where the expenditures are astronomical, pulling the whole US to the right... Who knows, since we always see the US vs Denmark vs Luxembourg vs Monaco vs Vatican City

Given statistics that show that blue states tend to do much better on life expectancy I have the feeling that nobody wants to publish that break down as it will only result in a partisan shouting match and drown out the larger message that the US is really failing to provide health care (as opposed to a for-pay health system) to its citizens.

The Bible Belt is running out of notches


An attempted pun on a belt running out of notches as you loosen it to accommodate increased girth, and the Bible Belt leading the obesity statistics. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3866220/figure/...

It would be interesting to see a chart normalized to Americans vs. Others with similar BMI ranges. For example, Americans with 18-25 BMI (I know you're out there!) vs. French with the same.

By taking out those with a very general correlation for 'does not take care of their own health', we'd get a truer measure of the relative effectiveness of the respective health care systems.

"Taking care of your health" is in significant part an outcome of your health care system.

BMI is a poor indicator of health and and even poorer indicator of whether one takes care of themselves. In the US, unhealthy food is abundant and strongly advertised. Much of the country is sprawl where the only way to really get around is by car. People are more stressed and work longer hours & don't have the time to cook for themselves or exercise. The US lacks universal healthcare, some people live in food deserts where healthy food is not accessible, etc. The United States has such terrible infrastructure and social programs that it's unfair to pin people's poor health entirely on their own decisions

> BMI is a poor indicator of health

But it's an available statistic.

If we had doctors report on body fat % (even done with shotty capacitance) or hip/waist ratio we'd be better off.

Asking to learn. What is this kind of graphing, the one that plots the trend over animated time, called?

Gapminder's moving bubble charts seem to be called "moving bubbles"


I thought that was Rosling's company, awesome!

Try looking at Hans Rosling's work, he made similar charts, has a company called gapminder and there's a nice ted talk with animated charts on world data.

What is going on with Greece in the first life expectancy chart? Large reversal in cost increase coupled with unperturbed steady increase in life expectancy.

Also has impressive numbers in other charts.

2008 financial crisis forced drastic cost-cutting.

The question is what the lag on health effects of that was. We know from things like smoking prevention and childhood diseases that the lag can be huge, anything up to half a century.

Likely to be a delayed time effect, people aren't born and live to 80 in the space of ~15 years.

So in the last 25 years, the medical costs more than doubled in the US and maternal mortality actually went UP?

Has anyone ever disputed this? Isn’t what we should be trying to prove whether or not that increased spending is actually providing innovation in drugs and procedures and thereby enabling other countries to put varying degrees of price controls in place without severely limiting the quality of care?

Framed in the opposite direction: could the US put in any form of price control to equalize this metric without significant impact to the continued investment in the field?

If the US managed to save money on healthcare, that extra money should go to universities so that the results of the research are not proprietary to pharma companies, but instead can be used by anyone.

That's not how it works. Commercialization of drugs from University research is what already feeds a multi billion dollar industry.

That's how it could work.

Then why restrict it to public universities? Give private universities and pharmaceutical companies funding as well, with the stipulation that research resulting from that funding be public domain. Public universities are not special entities in this (which is the GP's point).

How exactly do you see a system working that way? We establish a single payer plan and just start taxing everyone based on the current estimated healthcare expenditures and blindly shove money at universities?

You shove money at universities and don't patent the drugs they come up with so that companies can compete on the price of manufacturing them.

While financing innovation could explain the higher costs, it does not explain why the US performs bad in almost any health-related metric.

Of course not. That’s not related to money, that’s because we all eat anything we want and haven’t walked further than the car since we were 15... if anything it’s amazing that we’re as healthy as we are.

Wonder what this chart would like if you plot quality of care/life. It would also be interesting to see how health spending in the US (particularly drug spend) influence health related metrics in RoW.

Unfortunately, there are no free market health care countries to compare with.

There is free market health care. It's called "veterinarians".

Granted this won't do for comparing all metrics, but it's still an interesting reference point. Here's one article about this:


Fortunately or not, the US is far from having free market healthcare, especially considering its public health expenditure per capita alone beats most developed countries' public and private expenditures combined.

That's right. And before the government got massively involved in it, health care price increases pretty much matched inflation. The big ramp up in prices, that have never slowed down, started in 1968 with Medicare.

Is that even true? Is there a definition of "free market healthcare" that only the US meets?

Pricing is pretty much free market. It seems they can charge whatever they feel like.

Turns out that life or death has very small demand elasticity.

The vast majority of health care is not life or death.

Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact