U.S. life expectancy drops sharply, the second consecutive decline

I think this map makes very clear that there are certain regions where life expectancy is much worse than others: https://en.wikipedia.org/wiki/List_of_U.S._states_and_territ...

There's a strong correlation between state life expectancy and whether that state has embraced the ACA expansion of Medicaid as well:

https://www.kff.org/medicaid/issue-brief/status-of-state-med...

Every significant disease can have longstanding impacts. Damage is often cumulative. Actual causation can come via many routes. One person might have diminished physical fitness due to weeks/months laid up with the disease. Another might have impact from various drugs/treatments administered during the disease. Others might come away with diminished organ functionality (heart/kidneys etc) that in turn stresses other systems. The "what doesn't kill you" mantra is just that. It isn't based in science. It is almost always better to avoid getting sick in the first place.

When things like tax revenue-per-acre starts adding up, it’ll also be more affordable to live in denser areas.

By removing zoning restrictions, some communities will start becoming more walkable & mixed-use than others. Since it’s more cost efficient, it’ll be cheaper to own a townhome in a dense area than an equivalent townhome in a sprawling area (after the market equalizes & we can keep up with demand, because it’ll be sky-high for those walkable townhomes at first). But if zoning is reduced, supply will start to catch up.

People in sprawling communities will start needing to pay more for taxes for things like road maintenance etc since they have much higher costs per person than a dense area, further starting to push people to favor dense areas. Are people willing to pay the actual cost of living in remote areas like suburbs?

It’ll happen slowly at first, but just removing the current laws that prohibit that sort of development may actually enable those changes to happen automatically

This study is one of a number of different ones from the same lab (Al Aly), taking a dataset of older, sicker people (US veterans who seek hospital care at the VA), and looking at groups who had two hospital confirmed infections, vs. those with one hospital confirmed infection, vs those who went to the hospital with no confirmed infections.

Per the paper's abstract:

> We show that COMPARED TO PEOPLE WITH FIRST INFECTION, reinfection contributes additional risks of all-cause mortality, hospitalization, and adverse health outcomes

(emphasis mine)

All this paper shows is that if you follow a cohort of older, sicker people -- people who are seeking hospital care! people who were infected and possibly hospitalized in the first wave! -- long enough for them to get multiple reinfections (in this case, about 2 years, from the original Covid wave through the Omicron wave), the overall rates of a (selected) set of different health conditions goes up over time. They have not shown that the increased risk is because of Covid. The paper has no way of following the younger, healthier people (particularly in the later waves, where home testing was common) who never set foot in a VA hospital and had a Covid test, despite having Covid. There is a huge selection bias [1] problem here. There may be others, such as immortal time bias. [2]

What you would like to see in a study like this is that it is prospective (the participants were selected randomly in advance and followed over time); but this study is retrospective. You would also like to see a control against a similar illness, such as flu or non-Covid pneumonia; this study does not have such a comparison (despite being trivial to make), and resorts to using "atopic dermatitis and neoplasms" as a "negative control" instead.

As an aside, Al Aly has a history of using a weird grab-bag of negative control illnesses against this same dataset across his various papers. One wonders why he doesn't always use the same illness as a negative control -- presumably atopic dermatitis didn't suddenly start being unassociated with Covid between his studies, right? Would the results not be the same if he was consistent? I'm guessing not.

In short, this study is confounded, and Eric Topol really needs to tone down his fear-mongering. Even if you believe the top-line results, unless you are an elderly person sick enough to be seeking regular hospital care, this has very little relevance to your risk profile.

[1] https://catalogofbias.org/biases/selection-bias/

[2] https://catalogofbias.org/biases/immortal-time-bias/

Then obviously something else would have to be the main cause of the bump in excess deaths that's been observed.