I hope vaccines are more of the reason for the fall than I'm giving them credit for, that would be what is best for everybody. I'm just cautious to jump on that train because like I originally posted on Israel, their hospitalizations are falling at the exact same rate they were last year, just a couple weeks sooner. I hope we see them fall and never come back up, and we'll get a good test on that in a couple months. I bang the seasonality drum as hard as I do because it has almost been completely ignored by the media and people that are pushing the NPI narratives, when it's been the biggest factor across the US over the last year.
I agree that Lebanon and Israel should be a good comparison, though their cases pre-vaccination don't have a high enough correlation for me to agree that the main difference in the last month or two is vaccines. Unfortunately Lebanon doesn't have good hospitalization data that I could find to compare the two.
I think a better comparison for USA vs. Canada would be bordering states vs. Canada.....
Actually, before I finish that last paragraph, I think one thing about my argument that I haven't made clear enough is the following. I'm saying seasonality is by far the biggest driver for when cases rise vs. when they fall. Vaccines absolutely can/should play a role in limiting the magnitude of surges, and can/should play a role in how low cases/spread gets during periods of less than favorable conditions for spread. In the case of Israel, I don't think vaccines had much/if any to do with them peaking in mid-January, but they certainly should be contributing to the fall and how far they fall. It would be wonderful if the reason that they started falling again after a temporary plateau in late February was because of vaccines. My biggest caution there is that the slope of the fall is about the same as last year's in Israel at this time. Similarly in the US, I don't think vaccines had anything to do with when regions rose or fell (especially those regions that peaked and fell before vaccine rollout even ramped up or started), but it should be playing a role in where we are headed over the next couple months. Michigan and Ontario surged at almost the exact same time this spring, but Michigan's peak was lower than Ontario's. I hope that was vaccine related.
If I'm right, we should see cases/hospitalizations continue to decline across most of the US through May. In the northern US, this trend should continue through June/July at least. Around the start of June we should see cases increase again across the sunbelt region, with the worst areas being Arizona, Florida, S and SE Texas, and to a lesser extent Alabama/Georgia/Mississippi/Louisiana. I really think this time period will be a very good early barometer for the level of success the vaccines will have in the USA.
As for places like Colombia, it's possible that my seasonality theory is mostly worthless in tropical regions. There's some papers on flu seasonality in those regions, but I haven't read through them (just googled them a few minutes ago) to see what conclusions they came up with.
Well, I think we've discussed this some in the past, and I don't remember exactly where we left it, but my take is that although seasonality plays a role, some of the data correlation you see is really more about how people cycle their behavior in response to weather, rising cases, media attention, lockdown fatigue, etc.
Sure, UV exposure, RH, and temperature impact the survivability and spread of COVID. And I know you would absolutely agree that weather conditions definitely influence people's behavior in a given area as it pertains to activies that are more/less likely to transmit/spread COVID. For example, after a period of several weeks with cold temps and heavy snow, if there's a few days or a week with a nice thaw and sunshine, people's behaviors are going to change massively as far as what they do, where they do it, and how they do it.
Similarly, I think that event calendars throughout the year vary substantially on what activities are performed and where (inside vs outside and activities conducive the warmer or colder weather). Clearly COVID disrupted this, but most states kept fairly similar profiles in what they allowed and when/how with a few outliers here and there.
My point is that the climate variable, IMO, is more about how different climates and different periods of weather drive changes in human behavior and activities versus any impact on the virus itself and the virus' own behavior. The latter still is influenced by climate and weather but we shouldn't see such large variations between states who've had extremely similar weather for the past several months and share as very similar geography and climate.
As an example, why in the world has Ohio and Indiana not looked anything like Michigan for the period of March - yesterday? Looking at its surrounding states, there's not really a state anywhere around Michigan that has had similar case increases/decreases over the same time period. You could say Minnesota and Pennsylvania line as far as the shape of your case count changes on a graph but the slope and trajectory of cases in those states are a fraction of what happened in Michigan. A little odd, no?
To expound a bit further, we can also look at states that border Canada. I'm sure you're very familiar with how the population is distributed in Canada. Border crossings and population movement between Canada and the United States (excluding Vancouver) are going to be 95%+ Ontario/Quebec into Michigan/New York.
I found a COVID graph for the 5 counties of Western NY surrounding the Buffalo area (2nd busy USA/Canada crossing site after Michigan).
5 Western NY counties:
https://buffalonews.com/news/local/...cle_261f0820-dcb5-11ea-9ade-d3e35b30860b.html
Michigan:
https://www.worldometers.info/coronavirus/usa/michigan/
Canada:
https://www.worldometers.info/coronavirus/country/canada/
It is stunning how similar Canada and Michigan have been this year until vaccine rollout increased in the USA. The western NY map is not quite as heavily correlated but still looks more similar to Michigan and Canada than all the states in Michigan's region. No state in the Midwest has that kind of aggressive slope that Michigan has. But the western NY counties get close and Canada is just dead-on similar. Thus, my point is that the reason Michigan and the Western NY counties are seeing such a quick and sharp drop is because of better vaccination rates. Canada's numbers are slowly decreasing at a much lower rate of change.
It's just a theory. I would expect Michigan and the 5 western NY counties to be somewhat similar due to climate/weather and Canada border crossings. I think the border crossings and Canada's lower vaccination rates is a plausible explanation for why Michigan and the 5 western NY counties looked more like Canada than any place in the United States until recently. Just a hypothesis. I may be totally wrong.