This year we kill it: Corona Virus Thread Mark II

[pendell]

Before I review the earlier thread, I would like to present a pair of dueling studies on the effectiveness of Non-Pharmaceutical Interventions (NPIs) or "lockdowns" as they are colloquially known.

The first is from Imperial College and was conducted in June.

By comparing the deaths predicted under the model with no interventions to the deaths predicted in our intervention model, we calculated the total deaths averted in our study period. We find that across 11 countries 3.1 (2.8–3.5) million deaths have been averted owing to interventions since the beginning of the epidemic; Extended Data Fig. 5 compares the actual total deaths to the counterfactual total deaths. The counterfactual model without interventions is illustrative only, and reflects the assumptions of our model. We do not account for changes in behaviour; in reality, even in the absence of government interventions we would expect Rt to decrease and therefore would overestimate deaths in the no-intervention model. Conversely, we do not consider the effect on the infection fatality rate as a result of an overwhelmed health system in which patients may not be able to access critical care facilities, which would underestimate the number of counterfactual deaths. In the Supplementary Information, we show further counterfactual estimates under differing assumptions of the generation distribution and onset-to-death distribution and all scenarios broadly show the same trends. Given this agreement across differing scenarios, we believe our estimates for the counterfactual deaths averted to be plausible.

Now let's look at this one just released from Stanford

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 non‐significant 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, e.g., the effect of mrNPIs was +7% (95CI ‐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.

I've linked both studies directly.

So .. what to make of this?

Are they really in disagreement? Imperial seems to be saying that any NPI is more effective than the 'herd immunity' strategy, while Stanford seems to be addressing a more fine-grained difference in the kind of NPI employed. In other words, getting people to mask and social distance gives you most of the benefit while additionally going so far as to restrict travel etc. gives little additional benefit.

Or is there a fundamental difference in their modeling and assumptions which brings about a disparate result? If so, can anyone here put an easy finger as to what the difference between the two studies is that generates such different interpretations between the two?

Or, as in my first paragraph, is this not an apples to apples comparison? Are the studies actually in agreement that A) NPI is beneficial but B) that doesn't mean you need a stay-at-home-essential-travel-only NPI to get the benefit?

Another possibility is the factor of time in the studies: Perhaps if you intervene at the very start of an infection you can save millions of lives, but if you wait until you're in the middle of one there's little benefit to locking the barn after the horse is gone. Is that a reasonable conclusion to draw from the two studies?

Respectfully,

Brian P.