https://www.twcenter.net/forums/showthread.php?802465-Coronavirus-outbreak-From-China-to-the-World&p=16011601#post16011601
 Originally Posted by PoVG
Countries that followed strict lock down procedures properly like New Zealand certainly benefited the most. Those that did a half-assed job merely mitigated the outcome while also lengthening the process. On the other hand, those that had less lock down procedures enacted ended up creating a lock down environment as well as the extra deaths.
This assertion has been reduced to an unfalsifiable appeal to purity built on an anecdote. It’s worth repeating that the pandemic was declared before NZ went into lockdown, and at that stage, WHO recommendations indicate mass quarantines will no longer be effective and should therefore not be used as the costs of implementation likely exceed potential benefits. Parsing demographic and geographic factors from the lockdown impact in a single country will require the full perspective of hindsight, but the odds of a potential counter factual being significant enough to countervail the trend we’re seeing in the current as well as previous pandemics regarding the (in)efficacy of lockdowns are inherently low.
This would make NZ an outlier, best case, given its geography and low population density isolated it from international spread since the beginning of the outbreak by default. It would have been in a completely different place relative to more interconnected regions that were sitting on hundreds or thousands of undetected cases by the time major NPIs were adopted.
It’s also untrue overall that less restrictive interventions led to increased case growth relative to more restrictive measures. As cited earlier:
Implementing any NPIs was associated with significant reductions in case growth in 9 out of 10 study countries, including South Korea and Sweden that implemented only lrNPIs (Spain had a nonsignificant effect). After subtracting the epidemic and lrNPI effects, we find no clear, significant beneficial effect of mrNPIs on case growth in any country. In France, for example, the effect of mrNPIs was +7% (95% CI: -5%-19%) when compared with Sweden and + 13% (-12%-38%) when compared with South Korea (positive means pro-contagion). The 95% confidence intervals excluded 30% declines in all 16 comparisons and 15% declines in 11/16 comparisons.
Conclusions: While small benefits cannot be excluded, we do not find significant benefits on case growth of more restrictive NPIs. Similar reductions in case growth may be achievable with less-restrictive interventions.
https://pubmed.ncbi.nlm.nih.gov/33400268/
Originally Posted by PoVG
The studies that look at lock down measures, however, always try to simplify it which is a really bad way to look at it. There are many factors in play:
How widespread lock down measures were applied?
How strict were each lock down measure applied?
How well did the public follow them?
How well was it enforced?
How much does the public try to circumvent the measures?
Simply saying restaurants are closed while people are getting together in their houses more doesn't really tell us the full picture if the latter is not accounted for. This is pretty much what most, if not all, studies, even those that show positive effects for lock downs, fail to satisfy. They're too simplistic.
The “ more restrictive vs less restrictive” study looked at ~50 different NPIs from 11 major countries including Sweden, South Korea, the UK and Germany. Omitted variable bias is a major reason why the efficacy of lockdowns observed in some studies is statistically overstated.
A good example of that is the Flaxman paper referenced earlier and probably in every lockdown discussion since its publication. Professionally or otherwise, it’s been cited hundreds of times as evidence lockdowns work.
Our results show that major non-pharmaceutical interventions—and lockdowns in particular—have had a large effect on reducing transmission. Continued intervention should be considered to keep transmission of SARS-CoV-2 under control.
https://www.nature.com/articles/s41586-020-2405-7
The way the result was determined suggests the conclusion relies on a causation fallacy, which could be why it doesn’t hold when comparing more vs less restrictive NPIs across countries. The reason for this, in the context of the Flaxman study, is their model retroactively uses the number of deaths to predict future deaths under NPIs vs no NPIs and determined that NPIs reduce transmission.
That’s a fairly obvious conclusion, but to stretch this into a claim that “lockdowns in particular” caused it is mostly conjecture. As I said earlier, it’s unsurprising a study, looking at countries in the same region and adopting the broadly similar strategies, found lockdowns were correlated with reduced transmission. What’s odd is that Flaxman et al, determining the efficacy of lockdowns vs a few other categories of NPIs, don’t appear to address the fact that their own model predicts Sweden’s NPIs reduce Rt as much as other lockdown countries, including significant overlap with Denmark and Norway.
Also significant, I think, is that the predicted range of NPI impact is arguably more precise for Sweden than other countries in the model, and yet not much less than the most generous predictions for lockdowns’ impact on Rt. Flaxman and colleagues determine 3-3.5 million lives were saved by NPIs, but don’t mention the caveat in their conclusion this is due to “lockdowns in particular.” The model does not necessarily show this, even under its a priori assumption that all NPIs have the same multiplicative effect on Rt. The latter assumption is significant given it means the effect of any NPI on the model predictions will be equally strong in the first place, and even then, lockdowns don’t appear to have been especially impactful relative to other NPIs.
The reason the authors deduced a particularly strong effect of lockdowns is because most countries in the sample, all except Sweden, employed lockdowns in similar time periods, producing an outsized effect on the model by virtue of being a common factor relative to other kinds of interventions. Because the model attributes 100% of the reduction in Rt to government interventions at the outset, the commonality of lockdowns is predisposed to have a dominant effect, which seems to have led the authors to a correlation/causation fallacy. The results of the model actually suggest less restrictive NPIs did most of the heavy lifting across all countries in the sample, relative to lockdowns. This matches the conclusions of the comparison between mrNPIs and lrNPIs
The flaw in Flaxman’s argument is observed in other research, due in part to the difficulty of accurately weighting the probable effects of individual variables in a Bayesian model.
The peculiar aspect of the claim that lockdown accounts for 81% of the reduction in R 0, is that Sweden did not implement any lockdown, but still see a similar decrease in R 0 as the other countries, even though the other NPIs were reported to have no substantial effect on R 0. To solve this problem, as compared with the authors’ earlier work9, which showed a significantly higher R 0 for Sweden, they invoke a country-specific last intervention parameter, which is only implemented for Sweden10 (see equation i). The “last intervention” parameter is multiplied with R 0, and can therefore be seen as a parameter adjusting the model for Sweden independently. As can be seen, when analysing the posterior distributions of the intervention parameters, the “last intervention” parameter for Sweden results in 73.5 % of Sweden’s reduction in R 0 (Figure 1). The last intervention impact on R 0 is not reported or discussed in the Nature publication, possibly misleading decision-makers on the importance of lockdowns.
In conclusion, it is peculiar that the model displays an almost identical change in R 0 in all countries, dropping sharply below one at the final NPI, independently on the nature of that NPI. In reality, all countries had different NPIs implemented at different time points, likely with varying strength and efficiency, and it is quite likely that NPIs such as enforcing social distancing at least had some effects, not seen in the models. Given the importance the initial report had on government policies and the fact that we show here that the conclusions made about the significance of the lockdown are not entirely correct, we do think that we should pinpoint this to readers and policymakers. Correct assumptions on the effects of NPIs are becoming even more urgent as many nations still are imposing different NPIs, and that these might go on for an extended period
https://www.medrxiv.org/content/10.1...40v1.full-text
Because modeling attributes virtually all reduction in Rt to whatever NPI is used last, this indicates not only the existence of confounders but also that the common use of lockdowns is predetermined to have an oversized effect on any estimates.
Our finding in Fact 1 that early declines in the transmission rate of COVID-19 were nearly universal worldwide suggest that the role of region-specific NPI’s imple- mented in this early phase of the pandemic is likely overstated. This finding instead suggests that some other factor(s) common across regions drove the early and rapid transmission rate declines. While all three factors mentioned in the introduction, voluntary social distancing, the network structure of human interactions, and the nature of the disease itself, are natural contenders, disentangling their relative roles is dicult.
Our findings in Fact 2 and Fact 3 further raise doubt about the importance in NPI’s (lockdown policies in particular) in accounting for the evolution of COVID-19 transmission rates over time and across locations. Many of the regions in our sample that instated lockdown policies early on in their local epidemic, removed them later on in our estimation period, or have have not relied on mandated NPI’s much at all. Yet, e↵ective reproduction numbers in all regions have continued to remain low relative to initial levels indicating that the removal of lockdown policies has had little e↵ect on transmission rates.
The existing literature has concluded that NPI policy and social distancing have been essential to reducing the spread of COVID-19 and the number of deaths due to this deadly pandemic. The stylized facts established in this paper challenge this conclusion. We argue that research going forward should account for these facts when assessing how important NPI policy is in shaping the progression of COVID-19.
https://www.nber.org/system/files/wo...719/w27719.pdf
Originally Posted by PoVG
Taiwan is a good case of early strong response that made it possible that they don't even need to consider a full lock down.
Taiwan certainly did respond earlier than most other countries, thanks in part to its network of contacts on the mainland that passed information on the outbreak during the time Beijing was still trying to cover it up. Subsequent miscommunication from the WHO, probably in deference to China, about human transmission also delayed appropriate response by other countries, to the extent Taiwan’s NPIs came a month earlier than most. It would be a misnomer to say acting “early enough” makes lockdowns unnecessary, because lockdowns are ineffective beyond the earliest phases of an outbreak in the first place, which had likely passed by the time China publicly acknowledged the outbreak in Wuhan and announced countermeasures. This fits the findings from earlier research, which found the spread of Covid could have been confined to a regional outbreak had Chinese authorities put in place NPIs that were delayed 3-4 weeks by the coverup effort.
https://www.axios.com/timeline-the-e...l&stream=world
|
Primicerius
POTF 44 - Winner and Runner Up
𐤇𐤂𐤁
Posting Permissions
Reply With Quote
