You have a few factors. Medicaid eligibility and more strongly location is a rough approximation for likelihood of complications or infant mortality.
I consulted for an opioid surveillance program several years ago at a large scale data analytics level.
It was very sobering - basically the scientists could score each pregnancy in the state on a scale of 0-100, where 0% was <5% probability of being born addicted and 100 was >95%.
Most addicted mothers got hooked on painkillers as a result of injury or sickness. The biggest factors were access to healthcare, access to transportation, and health insurance. “Bad” zip codes were like 10x more likely to have an opioid addiction and 25x more complications.
Being a new dad at the time, it really affected my view of politics in this space. I realized that as cynical as I am, it’s not enough.
800 meters from where I live a comfortable life kids are born into a statistically shorter, poorer life with more disease, more mental unhealth and fewer healthy years.
People there live, on average, 4.5 years shorter than people where I live with something like 10 fewer healthy years.
The statistics comparing people born here and there are pretty awful. Despite having the same hospital, infant mortality rate is something like 2x (although the statistical uncertainty is high).
How is this not a failure of society? In addition: I live in a country with pretty high social mobility, especially compared to the US...
I've seen it reported recently that both infant and maternal mortality has gone up in states that have outlawed abortion. Even if the doctor forecasts that the pregnancy will be a risk to the mother or infant, they don't want to be second-guessed after the fact and prosecuted for performing the abortion.
The variations don’t seem to follow any rhyme or reason. Adjacent red and blue states do similarly (Idaho/Oregon, Minnesota/Iowa, Utah/Colorado). But North Dakota and South Dakota are quite different. Many states do worse (MD/VA) do worse than poorer ones (Oregon).
French doctors and the government have expressed concern over the rising mortality rate in recent years. Some worry that a contributing factor may be the relatively low rate of C-sections. Difficult labor is a major risk factor for neonatal death. It is also associated with intracranial hemorrhages at birth, which are common, but even more frequent and severe after complicated deliveries. These hemorrhages are often linked to wrongful diagnoses of shaken baby syndrome, which is commonly assumed whenever an infant presents with intracranial bleeding. France is likely among the countries with the highest rates of such misdiagnoses. Obvious to say the consequences for the child and their parents are severe.
Weird that some countries with high C-section rates are not particularly good in mortality (Turkey, Bulgaria, Romania, Poland), while the best countries by mortality has low C-section rates, Norway being the country with the lowest rate in the world.
Curiously, France has almost the exact same rate than Finland (200 per 1000 births), which tops the mortality list.
Meanwhile, in Finland, midwives here are worried abou the increase in C-sections in the last few years. It is most likely a cause of the decrease in funding towards healthcare, causing understaffed maternety wards and a push for c-sections on slow labor.
The opening graph is from the Office for National Statistics. The ONS is a UK govt department. So there is some UK data in the article but this is first-year mortality, not first month.
The author does say:
> The precise birth data needed to produce these comparable rates is not available for all countries, so only a selection of OECD countries is shown.
Perhaps this is the reason.
I do agree though, it's mighty strange to have not included an interpretation of you're going to mention them in the text. Perhaps they were excluded after the text was written?
China naturally cares a lot about the classification of the successor government of the other side of a previous civil war. The reason people in other countries care strongly is less clear to me.
The article's item 3 would have to be backed up for me to take anything away from this. There have traditionally been very different methods for tracking this information, where countries drew lines, etc. When i did medical software (granted long ago) you could not compare data between the USA, France, or Japan in any useful way.
Isn't the more appropriate question is where in the world are babies AND mothers at the lowest risks during all 9 months of gestation and then childbirth?
It's certainly another question, but the question of appropriateness seems to depend on context. I don't see what would make either more appropriate than the other in terms of "things a data analysis website might assess"?
This is looking entirely at the survival of children after birth. It even normalizes for situations where children are born before full-term.
What happens to postpartum mothers is certainly another question worth looking at, but the two questions are separate enough after birth that one can still be perfectly "relevant" without the other.
It would be interesting to see infant mortality rate as a function of GDP per capita -- that would tell you which countries make best use of their wealth to ensure the health of their citizens' children.
(It's not hard to guess which country would come out looking even worse than it already does in this table.)
"Japan, Sweden, and Finland are at the very bottom. They’re consistently among the best, even when we adjust for reporting inconsistencies. The other Nordic countries — Denmark and Norway — also have very low mortality rates."
I have no idea how this tracks to births, but I did some studies before of Nordic homicide rate, and if you drop an actual Nordic citizen in similarly white homogenous states like new hampshire their murder rate isn't terribly worse than it is in a Nordic nation.
My hypothesis is there may be something like this effect happening, where once you control for nordic people in similarly white states I bet they'd have much closer to nordic birth risk.
Ham-fisting statistics to support some model of eugenics and racially perfect societies is an old tale.
We need to be very careful with statistics, because rarely does one or even a dozen cover the whole picture. For example, in the US we can see that black individuals are more likely to commit crime. We could easily run with that and draw some unfavorable conclusions.
But, they're also much more likely to be impoverished, more likely to live in communities with low infrastructure funding, more likely to live in communities with drugs, more likely to have much poorer access to education, and more likely to face barriers to employment.
OK but blacks in Seychelles (adjusting for purchasing parity, about the richest black nation in existence to the point they're not far from the black american GDP per capita) have a gini coefficient far better than the US (in fact almost same as Sweden) but still have a pretty similar infant mortality to blacks in the US. So I don't know about the thesis they just need to be made more equal to the rest of USA or something.
Definitely not interested in accepting eugenics either, just maybe we should acknowledge and be OK with the fact some demographics are just different and we shouldn't be forcing them to be like the nordics.
I don't think it's a matter of demographics being different, I think it's a matter of us wanting to believe some demographics are inferior because that makes us feel better. I mean, we're not describing positive, but different, traits, now are we?
In regards to Seychelles - are we looking at levels of education? Are we looking at income inequality? Are we looking at corruption? Are we even considered how developed the nation is? How long have they been developing? Have we taken a gander at any population pyramids?
Again, it feels like we're hamfisting these things to draw conclusions that we want to draw. I don't think just having white people fixes things, we need to look at what those white people are doing that actually works. Even money doesn't necessarily fix things, if you're already underdeveloped or you have high corruption.
I have a thesis that could be wrong. I have some very weak data. What you have is that would be an inconvenient hypothesis.
It is quite an inconvenient hypothesis. Unfortunately I have some weak data that leads the question to persist and no strong evidence against it. I see no reason why it can be dismissed.
It's not just that it's an inconvenient hypothesis, it's that it's an hypothesis we've already been trying for several hundreds of years. It's how we justified slavery, the holocaust, colonialization. These people weren't dumb, they genuinely believe black people were inferior, so they would therefore have the best quality of life by being slaves. You see, we're really helping them by taking them from their backwater countries of tribal war and giving them opportunities to do structured labor that their feeble minds could handle.
Time and time again, we prove those hypothesis wrong. As countries develop and equality is prioritized, we see more and more differences disappear.
Ultimately, I have no reason to believe we have achieved the apex of development or equality. Systemic racism is real, because racism doesn't just disappear. We implemented integration just a few decades ago in a lot of communities, and that disconnect and resentment that built doesn't just - poof! - disappear. It continues on and manifests in less opportunities for education, more drugs in communities, and a lot of second order effects that transcend generations.
We even still see black populations today distributed as we saw them during slavery. And that was 150 years ago. These things don't get solved, they just get better, a little bit at a time over very long periods of time.
I dismiss the hypothesis because we have already dismissed the hypothesis more times than we can count. And, I have no reason to believe this time it's different.
I feel like we're on the same page then -- my hypothesis isn't intended to drive policy which is what you appear to object to. I don't want to 'enslave' black people into a systemically racist society that tries to force them into different infant metrics.
If even sovereign, nord-level equal in income, relatively well off black nations have US-black level infant mortality then it's not really my place to tell them to be more like the white man and do something that drops it but has god knows what other unintended consequences. Maybe they benefit in some other way. I have no idea.
Indeed, unfortunately when it comes to children/babies everyone is busybodies and seems to think they can act on behalf of the child in their interest and trump the actual parents. The biggest actor in the US in this capacity is CPS themselves, who investigate and take black children at far higher rates than the others races, believing they can stomp on black parents to implement their idea of bettering the outcome of babies to closer match that of white ones.
I hope someday society can see your plan of less interventionism is the more correct one.
Sure, take a look at Bahamas or Seychelles, "rich" black countries almost as high GDP per capita as some of the nordic nations (PPP adjusted in Seychelles case). Still with infant mortality in the relative shitter.
Or compare hispanic to blacks at similar income in USA, they knock them out the park on infant mortality.
> Sure, take a look at Bahamas or Seychelles, "rich" black countries almost as high GDP per capita as some of the nordic nations. Still with infant mortality in the relative shitter.
The Seychelles Gini coefficient is higher than Sweden. There doesn’t seem to be data on the Bahamas.
Gini coeff is bullshit anyway. Just look at who's topmost at income equality (Ukraine) and also at the bottom of the list for wealth inequality. This makes no sense as a practical indicator. Especially for the racist take the other poster is making.
Having mostly flat income inequality can mean either of those -- everybody is rich and paying taxes and people with money are tax evading really good, so poor don't have access to healthcare.
Then again, just because country has allocated money for healthcare, the outcome depends on how many sick/old/poor people need it and how well it's used up, because corruption exists and isn't reflected in Gini coeff.
The Nordics is where American conservatives and liberals find unity on harmonious white homogenity.
> My hypothesis is there may be something like this effect happening, where once you control for nordic people in similarly white states I bet they'd have much closer to nordic birth risk.
Historically (at least up until the late 90s) Japan, France and maybe others had a much different point of death included as death at birth that skewed infant mortality numbers. I'd like to see more than a bullet point 3 to understand how/when that was changed.
The U.S. has beginning to end health care for pregnancy and child birth. Its infant mortality rate (5.4) isn’t really much higher than Canada (4.7), which has a socialized healthcare system.
Which seems kind of okay, except when you consider that we pay significantly more per-capita for healthcare as opposed to Canada. Yes, that includes if you take into account taxes, and it's not even close.
Sure, we're not the absolute worst, but we are the most expensive. And, for that, we get close to the worst results. Clearly, our healthcare system is broken in a variety of complex ways.
As far as I can tell, we pay more for healthcare for the same reason we pay more for schools and more for subways. We’re a low competence, low trust society, and have to compensate for it by making everything subject to litigation to the point where the country is effectively run by lawyers.
My six year old boy ran into a table and got a black eye. Took him to the doctor (because my wife made me), who physically examined him and saw he was fine. But ordered a CT scan anyway (which we got the same morning because this is America). No sane healthcare system would order a CT scan for this! But in our litigation-driven system, the doctor has to do it, because in the extremely unlikely situation that there was an undetected internal bleed, he’d get sued. And some expert would get on the stand and say the standard of care is to order a CT scan every time a six year old boy does a six year old boy thing.
I don't think this is the sole reason why, I think private sector inefficiency is a big reason why, too.
We have an extremely fragmented system that breeds inefficiency. Thousands of insurers, so hospitals have hundreds of billing specialists. Thousands of plans, so the complexity of what is and isn't covered explodes beyond belief. There's no streamlining, no centralization, no authority. Just bickering and "erm, ackshually" from every party. Every interaction has extremely high friction that comes with a massive, fragmented system.
It's like a microservice architecture with thousands and thousands of microservices. Except their contracts aren't always published, sometimes you need to call them on the telephone. And sometimes you just have to try requests and see if they get denied.
Also, I think a CT for head injury is fairly standard practice. I think they do that in Europe. Anyway I had some pain somewhere inconspicuous once and it was cancer, so. I don't think the issue is we image too much.
We very definitely image too much. There are apparently as many MRI machines in Massachusetts as there are in all of Canada. "We image too much" is a pretty common complaint in health policy discussions; Rayiner is not just making that up himself.
I don't know, maybe, but at the same time our cancer screening recommendations are pretty conservative and we're actively looking to lower the screening ages for some of them because enough are slipping through the cracks.
We do image a lot for injuries. Maybe it's to give radiologists something to do, I don't know.
First, when we're talking about imaging overuse we're generally talking about injuries and back/joint paint.
But it's also probably not true that we screen less aggressively than Europe. For instance, I think we start breast cancer screening earlier than Europe. There are European countries with better rates of colorectal screening, but that's a patient compliance issue as much as anything else (and 10 years from now standard of care is unlikely to be imaging-based for that screening for most pts).
I agree our private sector is inefficient. But our public sector is too (hence my transit and educational examples). We suffer from cost disease across our entire economy, both in public and private spheres.
>Also, I think a CT for head injury is fairly standard practice. I think they do that in Europe.
Chiming in here from a rich EU country. CT for head bumps at the ER is not standard unless the doctor deems it absolutely necessary in grave injuries since the public system is already clogged up. Only X-ray on the spot is standard.
When I had my bicycle accident they did no CT scan, only head Xray. They said they'll do a head CT only if concussion symptoms don't go away or worsen after a few days.
In poor EU countries, you don't even get an Xray if your skull isn't cracked wide open because there's already 100 people in the ER waiting with even bigger issues than you. My dad slipped and fell on a concrete floor and the ER sent him home after looking at him for 3 seconds telling him "it looks fine". If he went to a private hospital he'd get all the imaging he wants since he'd be paying out of pocket and they'll never say NO to money.
I feel like Americans live in a parallel universe where healthcare has infinite money so they throw expensive checks procedures at the wall since insurance pays anyway, but that's not the case in public systems where money is tight than the government demands frugality from hospitals and doctors.
What's the point of using OECD data anymore when countries like India, China, MENAPT and Indonesia are excluded? Any study that uses OECD data should be disqualified as a shit study outright.
Seems to me that data quality and consistency is much more important than the number of countries. If China or Indonesia collect data of similar quality and consistency - by all means, OECD + these countries will make for a better dataset. If not, mixing shit data with good data is a better indicator of a "shit study".
Then the title of the article ought to be "Where in the developed world are babies at the lowest risk of death?". Straight up ignoring more than 50% of the world's population in one go, lmao.
For the record, I'm pretty sure most of the world has caught up on methods to gather mortality data, especially all of the countries I had mentioned.
The rest of the world has caught up on methods, but each country uses its own definitions.
International comparisons often use OECD data, because it's the best data source that exists. The organization has spent decades collecting data based on consistent definitions across its member states. Even the same data collected directly from the national authorities of OECD members is often worse. The data released for domestic uses often uses national definitions, which vary subtly from country to country.
>"Where in the developed world are babies at the lowest risk of death?
Where there are fewer black people or native americans.
In the US it almost follows to a T a heat map of where black people and reservations are. Except the places where there are enough hispanics to counter balance them, because for whatever reason hispanic identifying populations do comparatively quite well even with lower incomes.
It said world. The US isn't the world. And even in the developed world, without including oil rich Middle Eastern countries, the study is flawed. Heck, for all we know, the lowest risk might be in the theocratic monarchic state of Brunei.
Because it's more useful to compare wealthier countries with other wealthier countries (not by total size of the economy), for these types of metrics. But yes, I agree it would have been interesting to see India and China for comparison (and in fact, they are listed in the OECD report that was linked to in the article).
Update: I noticed that in the linked to OECD report, only the OECD countries, plus Romania, report the same 22-week threshold. That may be why those countries were picked (for a proper comparison with the same IVs.)
I consulted for an opioid surveillance program several years ago at a large scale data analytics level.
It was very sobering - basically the scientists could score each pregnancy in the state on a scale of 0-100, where 0% was <5% probability of being born addicted and 100 was >95%.
Most addicted mothers got hooked on painkillers as a result of injury or sickness. The biggest factors were access to healthcare, access to transportation, and health insurance. “Bad” zip codes were like 10x more likely to have an opioid addiction and 25x more complications.
Being a new dad at the time, it really affected my view of politics in this space. I realized that as cynical as I am, it’s not enough.
People there live, on average, 4.5 years shorter than people where I live with something like 10 fewer healthy years.
The statistics comparing people born here and there are pretty awful. Despite having the same hospital, infant mortality rate is something like 2x (although the statistical uncertainty is high).
How is this not a failure of society? In addition: I live in a country with pretty high social mobility, especially compared to the US...
I wonder how the consistent gestational period adjustments would affect the statistics for the various states.
https://www.nbcnews.com/health/womens-health/pregnant-women-...
and
https://jamanetwork.com/journals/jama/article-abstract/28302...
Curiously, France has almost the exact same rate than Finland (200 per 1000 births), which tops the mortality list.
>A baby in the UK, France, or the US can be two to three times more likely to die than one in Japan or Finland
why do they reference the UK while none of the graphs feature the UK?
The opening graph is from the Office for National Statistics. The ONS is a UK govt department. So there is some UK data in the article but this is first-year mortality, not first month.
The author does say:
> The precise birth data needed to produce these comparable rates is not available for all countries, so only a selection of OECD countries is shown.
Perhaps this is the reason.
I do agree though, it's mighty strange to have not included an interpretation of you're going to mention them in the text. Perhaps they were excluded after the text was written?
Be kind. Don't be snarky. Converse curiously; don't cross-examine. Edit out swipes.
Comments should get more thoughtful and substantive, not less, as a topic gets more divisive.
Please don't fulminate. Please don't sneer...
Eschew flamebait. Avoid generic tangents. Omit internet tropes.
Please don't use Hacker News for political or ideological battle. It tramples curiosity.
https://news.ycombinator.com/newsguidelines.html
Not saying that other question isn't worth looking at too. It's just a separate one from the one in this article.
What happens to postpartum mothers is certainly another question worth looking at, but the two questions are separate enough after birth that one can still be perfectly "relevant" without the other.
(It's not hard to guess which country would come out looking even worse than it already does in this table.)
"Japan, Sweden, and Finland are at the very bottom. They’re consistently among the best, even when we adjust for reporting inconsistencies. The other Nordic countries — Denmark and Norway — also have very low mortality rates."
Interesting in the sense that the Nordics have a vastly different healthcare system than Japan.
My hypothesis is there may be something like this effect happening, where once you control for nordic people in similarly white states I bet they'd have much closer to nordic birth risk.
We need to be very careful with statistics, because rarely does one or even a dozen cover the whole picture. For example, in the US we can see that black individuals are more likely to commit crime. We could easily run with that and draw some unfavorable conclusions.
But, they're also much more likely to be impoverished, more likely to live in communities with low infrastructure funding, more likely to live in communities with drugs, more likely to have much poorer access to education, and more likely to face barriers to employment.
Definitely not interested in accepting eugenics either, just maybe we should acknowledge and be OK with the fact some demographics are just different and we shouldn't be forcing them to be like the nordics.
In regards to Seychelles - are we looking at levels of education? Are we looking at income inequality? Are we looking at corruption? Are we even considered how developed the nation is? How long have they been developing? Have we taken a gander at any population pyramids?
Again, it feels like we're hamfisting these things to draw conclusions that we want to draw. I don't think just having white people fixes things, we need to look at what those white people are doing that actually works. Even money doesn't necessarily fix things, if you're already underdeveloped or you have high corruption.
It is quite an inconvenient hypothesis. Unfortunately I have some weak data that leads the question to persist and no strong evidence against it. I see no reason why it can be dismissed.
Time and time again, we prove those hypothesis wrong. As countries develop and equality is prioritized, we see more and more differences disappear.
Ultimately, I have no reason to believe we have achieved the apex of development or equality. Systemic racism is real, because racism doesn't just disappear. We implemented integration just a few decades ago in a lot of communities, and that disconnect and resentment that built doesn't just - poof! - disappear. It continues on and manifests in less opportunities for education, more drugs in communities, and a lot of second order effects that transcend generations.
We even still see black populations today distributed as we saw them during slavery. And that was 150 years ago. These things don't get solved, they just get better, a little bit at a time over very long periods of time.
I dismiss the hypothesis because we have already dismissed the hypothesis more times than we can count. And, I have no reason to believe this time it's different.
If even sovereign, nord-level equal in income, relatively well off black nations have US-black level infant mortality then it's not really my place to tell them to be more like the white man and do something that drops it but has god knows what other unintended consequences. Maybe they benefit in some other way. I have no idea.
I hope someday society can see your plan of less interventionism is the more correct one.
It’s baked into the law and has resulted in the essential elimination of people with downs for example.
https://www.theatlantic.com/magazine/archive/2020/12/the-las...
Or compare hispanic to blacks at similar income in USA, they knock them out the park on infant mortality.
The Seychelles Gini coefficient is higher than Sweden. There doesn’t seem to be data on the Bahamas.
https://ourworldindata.org/grapher/economic-inequality-gini-...
Having mostly flat income inequality can mean either of those -- everybody is rich and paying taxes and people with money are tax evading really good, so poor don't have access to healthcare.
Then again, just because country has allocated money for healthcare, the outcome depends on how many sick/old/poor people need it and how well it's used up, because corruption exists and isn't reflected in Gini coeff.
> My hypothesis is there may be something like this effect happening, where once you control for nordic people in similarly white states I bet they'd have much closer to nordic birth risk.
That follows from murder rates? I don’t follow.
That's why the US is always at the very bottom of rankings like this.
Sure, we're not the absolute worst, but we are the most expensive. And, for that, we get close to the worst results. Clearly, our healthcare system is broken in a variety of complex ways.
My six year old boy ran into a table and got a black eye. Took him to the doctor (because my wife made me), who physically examined him and saw he was fine. But ordered a CT scan anyway (which we got the same morning because this is America). No sane healthcare system would order a CT scan for this! But in our litigation-driven system, the doctor has to do it, because in the extremely unlikely situation that there was an undetected internal bleed, he’d get sued. And some expert would get on the stand and say the standard of care is to order a CT scan every time a six year old boy does a six year old boy thing.
We have an extremely fragmented system that breeds inefficiency. Thousands of insurers, so hospitals have hundreds of billing specialists. Thousands of plans, so the complexity of what is and isn't covered explodes beyond belief. There's no streamlining, no centralization, no authority. Just bickering and "erm, ackshually" from every party. Every interaction has extremely high friction that comes with a massive, fragmented system.
It's like a microservice architecture with thousands and thousands of microservices. Except their contracts aren't always published, sometimes you need to call them on the telephone. And sometimes you just have to try requests and see if they get denied.
Also, I think a CT for head injury is fairly standard practice. I think they do that in Europe. Anyway I had some pain somewhere inconspicuous once and it was cancer, so. I don't think the issue is we image too much.
We do image a lot for injuries. Maybe it's to give radiologists something to do, I don't know.
But it's also probably not true that we screen less aggressively than Europe. For instance, I think we start breast cancer screening earlier than Europe. There are European countries with better rates of colorectal screening, but that's a patient compliance issue as much as anything else (and 10 years from now standard of care is unlikely to be imaging-based for that screening for most pts).
Chiming in here from a rich EU country. CT for head bumps at the ER is not standard unless the doctor deems it absolutely necessary in grave injuries since the public system is already clogged up. Only X-ray on the spot is standard.
When I had my bicycle accident they did no CT scan, only head Xray. They said they'll do a head CT only if concussion symptoms don't go away or worsen after a few days.
In poor EU countries, you don't even get an Xray if your skull isn't cracked wide open because there's already 100 people in the ER waiting with even bigger issues than you. My dad slipped and fell on a concrete floor and the ER sent him home after looking at him for 3 seconds telling him "it looks fine". If he went to a private hospital he'd get all the imaging he wants since he'd be paying out of pocket and they'll never say NO to money.
I feel like Americans live in a parallel universe where healthcare has infinite money so they throw expensive checks procedures at the wall since insurance pays anyway, but that's not the case in public systems where money is tight than the government demands frugality from hospitals and doctors.
For the record, I'm pretty sure most of the world has caught up on methods to gather mortality data, especially all of the countries I had mentioned.
International comparisons often use OECD data, because it's the best data source that exists. The organization has spent decades collecting data based on consistent definitions across its member states. Even the same data collected directly from the national authorities of OECD members is often worse. The data released for domestic uses often uses national definitions, which vary subtly from country to country.
Where there are fewer black people or native americans.
In the US it almost follows to a T a heat map of where black people and reservations are. Except the places where there are enough hispanics to counter balance them, because for whatever reason hispanic identifying populations do comparatively quite well even with lower incomes.
Update: I noticed that in the linked to OECD report, only the OECD countries, plus Romania, report the same 22-week threshold. That may be why those countries were picked (for a proper comparison with the same IVs.)