4 ms·

I'd wager odds you would have survived without are above 95%, so its beneficial, but when averaged out, a very slight improvement, quite possibly too low to be

by midjji 5y ago

I'd wager odds you would have survived without are above 95%, so its beneficial, but when averaged out, a very slight improvement, quite possibly too low to be captured with significance in 10k people over 10 years with confidence p=0.99. A few years latter the effect of a month+ sick at home is almost certainly absent from blood pressure, etc. Insulin might show up, though its prevalence is what, <2% ? And most diabetics do survive without, they just have to be very careful about what and when they eat, and a bit lucky. Psychiatric meds probably dont count, as its often not a part of healthcare, for similarly sane reasons as dentistry. Even if it does, it pretty much has to prevent suicide, or it will count as zero, as neither preventing forced hospitalization, nor homelessness will count as improved health.

pas 5y ago
I've reread your comments and while I generally agree, my hypothesis is that simply having good healthcare improves the economy which then in turn leads to improved life expectancy. Of course it's next to impossible to measure this, because of the endogeneity problem. That said, I'm not sure I understand what you wrote about psych meds ("not a part of healthcare for [...] sane reasons [...]"). To me it makes the most sense to have psych meds socially paid, to ... you know keep people sane, keep them in the workforce instead of waiting for their problem to grow so big that now they need to be put into the psych ward (or court mandated rehab).
- midjji 5y ago
  I am pretty sure that convenient access to good healthcare significantly improves health and is good for society in general, just not the way measured by these studies. I dont think it was intentional, but when carelessly read they needlessly discredit public healthcare. More importantly, I think lifespan is overrated in terms of health, it isn't what people care about the most, nor what is valuable to society. Proxies for economic productivity, and quality of life should be used as well. Difficult to be sure, but not impossible. The problem lies more in the ability of the analyst and the limited, highly abstracted(compressed), often anonymized(lossy) data used. I think the kind of AI we need is already here, and we are slowly but surely transitioning into a true information society. A futuristic idea is that people will all wear sensors which provide a bunch of medical statistics which are then used by semi automated healthcare, but I dont think thats the way it will come to pass, or at least not the entire story. Why wear a device measuring your pulse, when a webcam can do the same while being used for other things, or just not turned off. Your pupillary response speed is another useful metric that a high res webcam can measure. To my knowledge the two aren't used to measure ad responsiveness or engagement to unaware users yet, but they will. This kind of non-abstracted, non-anonymized data already allows some companies to predict health, especially mental health, very well, and that's by accident. Oh I mean the same reason that dentists are considered blacksmiths rather than doctors, people are just dumb sometimes. I may have sounded a bit pessimistic about healthcare, but I am more disappointed than pessimistic. I think medicine can and should be improving much faster. I think part of the reason it isn't, is because the public is missinformed about the medical outcomes. And if they knew, far more funding would become available. But faith in doctors ability, has been and, may still be more important than treatment. By this I dont mean mean placebo, I mean it would scare people in general, and that good empathic doctors would burn out fast without an optimistic view of their outcomes. This was even a stronger factor in the past, and as a result this idea is strong, and prevents more ambitious programs.

by midjji 5y ago

pas 5y ago

I've reread your comments and while I generally agree, my hypothesis is that simply having good healthcare improves the economy which then in turn leads to improved life expectancy. Of course it's next to impossible to measure this, because of the endogeneity problem. That said, I'm not sure I understand what you wrote about psych meds ("not a part of healthcare for [...] sane reasons [...]"). To me it makes the most sense to have psych meds socially paid, to ... you know keep people sane, keep them in the workforce instead of waiting for their problem to grow so big that now they need to be put into the psych ward (or court mandated rehab).

midjji 5y ago
I am pretty sure that convenient access to good healthcare significantly improves health and is good for society in general, just not the way measured by these studies. I dont think it was intentional, but when carelessly read they needlessly discredit public healthcare. More importantly, I think lifespan is overrated in terms of health, it isn't what people care about the most, nor what is valuable to society. Proxies for economic productivity, and quality of life should be used as well. Difficult to be sure, but not impossible. The problem lies more in the ability of the analyst and the limited, highly abstracted(compressed), often anonymized(lossy) data used. I think the kind of AI we need is already here, and we are slowly but surely transitioning into a true information society. A futuristic idea is that people will all wear sensors which provide a bunch of medical statistics which are then used by semi automated healthcare, but I dont think thats the way it will come to pass, or at least not the entire story. Why wear a device measuring your pulse, when a webcam can do the same while being used for other things, or just not turned off. Your pupillary response speed is another useful metric that a high res webcam can measure. To my knowledge the two aren't used to measure ad responsiveness or engagement to unaware users yet, but they will. This kind of non-abstracted, non-anonymized data already allows some companies to predict health, especially mental health, very well, and that's by accident. Oh I mean the same reason that dentists are considered blacksmiths rather than doctors, people are just dumb sometimes. I may have sounded a bit pessimistic about healthcare, but I am more disappointed than pessimistic. I think medicine can and should be improving much faster. I think part of the reason it isn't, is because the public is missinformed about the medical outcomes. And if they knew, far more funding would become available. But faith in doctors ability, has been and, may still be more important than treatment. By this I dont mean mean placebo, I mean it would scare people in general, and that good empathic doctors would burn out fast without an optimistic view of their outcomes. This was even a stronger factor in the past, and as a result this idea is strong, and prevents more ambitious programs.