Curve Confusion: 1918 ‘flu’ to cause deaths in 2020?

The most popular phrase in addressing Covid-19 quickly became ‘flattening the curve’. I suggest the unintended consequence of this phrase has been a dangerous misunderstanding of the nature the pandemic. With containment measures often dictated by politics, this misunderstanding may lead to a huge number of unnecessary deaths.

In the US, this confusion is already likely to increase what could have been a death toll well below 100,000, to at least over over 300,000. If the confusion and 1918 thinking still prevails, then 1918 deaths per 100,000 could be the result, with well over 1 million Americans losing their lives if a vaccine does not come to the rescue.

Outside the US, the same misunderstanding is spreading and causing loss of life.

Solving the problem requires forgetting the notion of ‘the curve’ and realising this curve represents the behaviour of an out of control pandemic which need not apply at all in 2020.

The Confusion.

National Geographic

The confusion is the belief that the outbreak runs as a curve. It actually did back in 1918, right? Well, close enough. Sometimes the presence of a ‘second wave’ made the curve look like a two hump camel, but even when there were lockdowns, the curve climbed until a point of herd immunity. In 1918 there was no blocking lockdowns that completely disrupt the ‘curve’. The war-torn, no internet, no tv, no home refrigeration, almost car free world did not make lockdowns effective enough to stop the curve happening.

In 2020, almost every country in the world has managed to stop the curve through a blocking lockdown. While ‘flattening the curve’ in 1918 meant a longer wait for the eventual curve peak, when at least 1/4 of the population has been infected, ‘blocking lockdowns’ in 2020 mean the curve will normally peak as soon as the lockdown takes effect even if less than 1% of the population has been infected.

The number of people infected at the peak is the big difference.

In 1918 with the less effective lockdowns, the peak was reached by achieving a level of herd immunity. This ‘level of herd immunity’ came at a cost, at cost in terms of deaths per 100,000 that was beyond what any country has yet paid in 2020.

However, what has not changed, is that the only way to reach the true end of an outbreak where all can ‘return to normal’, without a vaccine, is to pay the full cost. In 1918 a peak meant most of that cost was already paid. In 2020 a peak means nothing in terms reaching the end of the outbreak.

In 1918 for cities like Philadelphia, New York and San Francisco, the peak meant the worst was over. In 1918, cities with a flattened curve like St Louis, Columbus, Milwaukee etc.

Generally, he better the ‘flattening’, the less the ‘curve’ resembles the classic pattern, and the more there is still to come after that first ‘peak’.

2020: No Philadelphia, No St Louis!

There is continued cherry picking of analogies from from 1918, with ‘be a St Louis (with a flattened curve), and not a Philadelphia as one of the favourites. These represent the ‘curve’ choices of 1918, but neither is representative of Covid-19 in 2020.

This pandemic killed over 50 million people worldwide between 1918 and 1922 and infected roughly one third of the world’s population. In the US, about 675,000 people died while 22 million caught the disease. Pennsylvania, one of the states that was hit the hardest, faced over 60,000 deaths. Philadelphia lost about 12,000 people and had about 47,000 reported cases in just four weeks. In just six months, there were about 16,000 deaths and half a million cases of influenza in Philadelphia.

University of Pennsylvania Archives.

Considering that Philadelphia at the time had around 1/15th of the population of metropolitan New York today, the equivalent would have been 180,000 deaths in a four week period. Clearly, bad as things were, even New York was no replay of Philadelphia in 1918.

Now consider the poster child of 1918, St Louis. At 358 deaths per 100,000, this rate would still translate to 71,000 deaths in metropolitan New York alone. Before cases started to decline 1 in 4 citizens has been infected. For the worst outbreak in the US in 2020, New York, has outperformed the poster child of 1918 in terms of how early cases started to initially decline. By official statistics, at the time of ‘the peak’ of over 6,000 infections in one day on April 6th, less than 200,000 people, or 1/100th of the population had been infected. Even using the far higher infection estimates from surveys, less than 1 in 10 citizens had been infected at the time of that peak.

Projections by New York modelling as at April 24

So far, everywhere in the US has flattened the curve beyond even the most ‘flattening’ back in 1918. The impact is that everywhere faces a far bigger ‘second wave’ if every thing was to open as soon as the curve reaches a peak.

The reality of 21st century blocking lockdowns is that a peak and declining cases typically mean that only two things that have changed since the outbreak began are:

  • some form of lockdown conditions apply
  • there are more cases than there were when the outbreak first started

These two points combined should make it clear that simply exiting lockdown to go back to how things were on the basis of a ‘curve peak’, will simply repeat the previous growth of cases, but now starting from a larger initial caseload.

Australian comparison of 1918 and 2020. Note 2020 is NOT a curve and ‘peaks’ are meaningless

Many people want to believe ‘the curve’ passing a peak has the significance it did in 1918. There is not a clear enough message ‘the curve of 1918 does not apply’.

A peak can be just ‘noise’. No danger has passed by, and confusion that danger has moved on may cause the next phase of this outbreak to be worse than what has been seen so far.

Confusion Example: US projections suggest the outbreak is almost over

The USA has huge influence over the rest of the world, so what happens in the US has an impact on how people think elsewhere. This is worrying because the message from the USA seems to be that if cases start falling (i.e. a peak) then then the end of the outbreak is near.

Symptom 1: Projections Deaths will Stop.

I think the most recent prediction I have seen by the US president was on May 3rd for up to 100,000 fatalities (in fact on May 8th the prediction was for 100,000, maybe 110,000). The previous prediction I had seen was the April 29th prediction of 70,000 when at the time the death toll was already at least 63,000. The trend for all these predictions is extreme optimism. With deaths averaging over 1,000 per day and many states relaxing isolation measures, it seems extremely likely that the latest prediction will be exceeded by month end.

Personally I find it hard to believe with current attitudes there will be less than 300,000 deaths in the US. Even scientific estimates tend to be low, and in the US political climate, where what action will be taken to lockdown is highly uncertain, predictions are very difficult.

Symptom 2: Lockdown Easing linked to ‘a Peak’

There is a federal guideline that states should not ease lockdowns prior to 14 days of declining cases number. In other words, once there has been a peak, lockdowns can be relaxed.

Conclusion: What hope for Science

We live in a world facing a battle between ‘belief’ and science. It certainly does not help when science itself is not speaking out again use outdated terminology from 1918 being quoted as applicable to today.

Analysis reveals the whole assumption of ‘the curve’ can be misleading in 2020.

The image of the curve is everywhere, suggesting that ‘a peak’ means it is all about things getting better from here. In 1918 for cities the did ‘not flatten the curve’ is was the start of the end, while for cities that did flatten the curve, it was more like the half way point.

In 2020, a peak basically means nothing.

If in 2020 the goal is ‘herd immunity’, as it always was in 1918,, then the measure is ‘when will half the population have been infected?’

In 2020 if the goal is containment, then the measure is ‘when will case numbers be low enough to allow track and trace to be effective’.

‘Cases have started to fall’… means almost nothing.

If this message to ignore peaks does not cut through, and political attitudes don’t change, we could see a repeat of 1918 which would mean far in excess of 1 million deaths in the US alone. Hopefully as the toll rises, people will start to listen and perhaps 300,000 will be realistic. At least, provided there is a vaccine.

One thought on “Curve Confusion: 1918 ‘flu’ to cause deaths in 2020?

Add yours

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Blog at WordPress.com.

Up ↑

%d bloggers like this: