Comment by bzbarsky | Hacker News Reader

Comment by bzbarsky | Hacker News Reader

Parent Full thread

The problem with infant mortality figures is that different countries measure them differently. There are cases (e.g. very premature births) that the US considers a live birth (with a high risk of mortality) while other countries consider them miscarriages and hence don't count them in infant mortality.

There have been attempts made to correct for these discrepancies. They end up accounting for a large fraction (but not all) of the difference in infant mortality rates, if I recall correctly.

Most of the rest of the effect is that premature births that everyone considers "births" have a higher prevalence in the US (for various reasons, not all of which are clear). Premature births have higher mortality, obviously.

https://uk.reuters.com/article/us-health-infants-mortality/u... is actually trying to look at an apples-to-apples comparison, and a few things jump out at me:

1) Birth defects. I wonder how much of this has to do with differences in abortion availability and again differences in definition of "birth". Many other countries count a child that dies within some number of days after birth from birth defects as a stillbirth, not an infant mortality event.

2) SUID (aka SIDS) is a huge contributor in the US. In the US this is highly concentrated in ehthnic groups that are largely missing in the comparison countries. Why that is is a good question, but makes this comparison less apples-to-apples.

To know where you'd want to be you want to know the outcomes for your specific demographic in the different countries, which is not something that gets reported very much...

"Infant mortality rates for full-term babies vary across the U.S., but all states are worse than many European countries, a new study suggests."

From your own source, literally doesn't help your case at all.

I am aware of what the article says.

The bit you quoted does not negate my two points about the data in that article.

But just to be very clear, Connecticut is cited as having an infant mortality rate that is slightly above the comparison country rates. Note that those countries were picked for a comparison because they have the 6 lowest mortality rates in Europe. [1]

Anyway, Connecticut is maybe somewhat similar to the comparison countries in terms of ethnic group demographics (mostly negating my point (2)) and has a more expansive definition of "infant mortality" than those countries last I checked, which is part of my point (1).

Mississippi, at the other end of the spectrum, has a vastly different ethnic makeup, has much worse abortion availability (increasing the number of babies with birth defects carried to term). And still has the more expansive definition of "infant mortality".

Now I'm not saying that it's a good thing that Mississippi has higher rates of SUID/SIDS. And I'm not saying limited abortion availability for cases when the child wouldn't survive is good. But I am saying that drawing conclusions about US healthcare here, as opposed to other societal factors, requires apples-to-apples comparison. And drawing conclusions about outcomes for a specific case (specific family) requires understanding whether those societal factors apply to it.

Or to put it another way, families that are willing to do an abortion when told their child, if carried to term, would not live more than a few months will have a lower infant mortality rate than families that are not willing to do said abortion.

One other thing which _is_ mentioned in the study itself but not in the article: "survival rates among preterm infants in the US were found to be very similar to those of the same European countries". Please model that with the "the US just sucks" model.

[1] The numbers in the actual study: Connecticut has an estimated FTIMR of 1.29 with a 95% confidence interval of 1.08 - 1.53; note that they don't have an actual hard number for reasons I haven't figured out yet. The six European countries involved are Austria, Denmark, Finland, Norway, Sweden, and Switzerland. Their numbers range from 0.97 to 1.24 according to the study.

Reply on news.ycombinator.com