Flatten the curve: Misleading?

The phrase ‘flatten the curve’ has become popular, and perhaps, overused. The term has come to be associated with any strategy to manage Covid-19 that includes some measures to reduce the rate new infections. This page provides analysis of what different things can be done (at least in theory) to manage an outbreak, and the ‘curves’ that result, together with how to achieve the desired curve.

The end result is that textbook ‘flatten the curve’ diagrams do not reflect reality with Covid-19 in the 21st century, and mislead people into complacency.

This post describes four different ‘flattening the curve’ models. However, on review, every curve possible can be viewed as a variation of a ‘containment strategy’. The resulting curve being determined solely by the containment limits. Six scenarios which can be seen as covering the possible scenarios in almost every country are then mapped to the corresponding curves as that county meets the varying requirements and plans to ‘flatten the curve’.

Which countries how seen which scenarios? And what is the road forward in each case.

  • what is ‘flattening the curve’?
  • ‘flattened curves’
    1. the original textbook ‘flatten the curve as shown in popular diagrams: (not known to be applicable to Covid-19)
    2. a more practical flatten the curve result
    3. flattening to stop the curve
    4. stop and go: flattening for a ‘containment curve
  • aren’t all curves containment?
  • reading curve graphs
  • Scenarios, and their path to containment
    1. emergency! flatten the curve! people are dying! A very real Covid-19 Scenario in response to a catastrophe
    2. planned outbreak flattening: we will flatten the curve to have a controlled outbreak
    3. Avoidance lockdown: we will flatten the curve to peak to avoid being another country with a catastrophe
    4. eradication lockdown: stop the curve, flatten it until it is nothing!
    5. flatline: ‘flatten the curve’ with no curve, prevention of an outbreak
    6. flatline lockdown: containment has broken limits, add a lockdown to restore low case levels
  • Conclusion
Continue reading “Flatten the curve: Misleading?”

Coronavirus: Facts, Myths and Surprises (March 20th)

(go here for link to latest update)

This is a collection of facts on the ‘novel’ coronavirus, together with some discussion on myths and surprises such as:

the debunking and fact exploring on topic of ‘the coronavirus’.

My first post on the new coronavirus (the real threat level) back in late January speculated there could be as many as 180,000 cases by end of February if the rate of spread could not be constrained. The rate of spread has been constrained (and it was around half of my estimate), but not sufficiently constrained to prevent a pandemic, or suggest the threat has passed. My second post spoke of tipping points that would lead to the pandemic, and it turned out that the outbreak in northern Italy constituted tipping point3, not the false alarm of the Westerdam cruise ship. Still, the tipping point did occur, as sooner or later seemed inevitable it would. We, now as this post is written on Friday 13th of March, have a pandemic.

This post is on general infromation on the virus and what has been learnt so far, and I should be able to provide updates over time to this post, avoiding the need for new posts.

Continue reading “Coronavirus: Facts, Myths and Surprises (March 20th)”

Coronavirus: Facts, Myths and Surprises (Q&A and toilet paper Update)

(go here for prior update or here for all on covid-19)

I have been asked a few questions…here they are with answers:

Why Toilet Paper?

In many parts of the world, a shortage of toilet paper on shelves has been the first sign of panic buying. Since there appears no logical reason for a specific need for toilet paper as a result of Covid-19, the question arises: “Why Toilet Paper?”

The main reason is that if people is that if people increase their buying of a variety of staples and toilet paper in the mix, toilet paper will be the first to be noticed. Supplies dropping being noticed leads to more panic buying, which leads to empty shelves, which leads to more panic.

So why first to be noticed? Simply because toilet paper is big and bulky, and for that bulk relatively inexpensive. So toilet paper requires a lot of shelf space. All shelf space is expensive, yet toilet paper is not particularly high profit, so great care is required not to allocate too much shelf space. In fact supermarkets take great care to avoid stocking more than a few days of any product as not being efficient with regard to shelf space would eat into profits. The bulkier per dollar of revenue the product is, the more important that product does not have too many days supply taking up valuable shelf space.

But if customers take just a few extra packets of toilet paper, the shelf starts to look empty. A few extra tins of beans simply does not free up the same amount of shelf space, so does not automatically start the snowball rolling.

So in the end, ‘why toilet paper?’: ‘Because it is needed, it is very bulky, and it is relatively inexpensive’.

Why Such Different Mortality Rates?

Misleading Data

Looking at the chart of Covid-19 Mortality rates, they vary enormously! It would seem you are almost 40 times more likely to die if you contract Covid-19 in Italy than if you contract the disease in Germany. Right now, you are more likely to die in Italy, but nothing like 40 times more likely. The biggest different is in testing. When the panic is on, only the most serious cases get tested and confirmed, so the percentage of people infected who appear in the ‘cases’ statistic is far lower. With at smaller percentage of infections counting as cases, even with the same actual mortality rate, the figures will look worse.

In short, the more testing is done, the lower the mortality rates statistics, even when actual mortality rates are the same.

So even when comparing countries like Germany Vs UK, where in both countries medical caseloads are not yet sufficient to block access to intensive care when required, and actual outcomes will probably be similar, the statistics still vary by a factor of 20x. Lies, dam lies and statistics. Even though to percentage of infections detected varies with government policy, none knows the real number of infections because so may infections are asymptomatic. The UK government Chief Medical officer stated ‘there could be as many as 10,000 infections’ at time when the verified number of infections was around 600. Use the larger number, and you get a mortality rate similar to Germany.

Actual Variations – treatment.

Although statistics and data available do not reveal true mortality rates, it is clear than in a country with an overloaded medical system and the reality that an intensive care bed may not be available when needed. There is a real difference in outcomes between a country with a manageable case load, and a country with an overloaded medical system. Far more cases need intensive care than die- but if there is no intensive care available..what happens?

Mortality and Stage.

Now consider this data for Italy. It is updated daily, so here is a snapshot, which could have old data by the time you read this, but the principle will still apply. On the chart here, there have been 31,500 cases and 2,500 deaths. So almost 8% mortality. But this is a distortion, because it is no known what the outcome will be for 26,000 cases…they are still infected. For those with an outcome decided, it is 2503/(3503+2941) = 46% deceased. Scary…but again inflated because figures in Italy do exclude most cases as unless the case is already extremely serious, it does not get confirmed. So of people infected, most likely at least a 10x or 16x times larger group, the survival rate is far better. But consider …recent infections do not yet have an outcome. So if most of the infections are very recent, of course there are no fatalities as no one has had time to die yet. As figures climb in the early stages, fatalities will look lower.

No Real Data: How Many Infections?

If asked how many cases in country XXX, you can go to the web and find a number. I even made a table of some of that data. Every country produces stats on how many cases of Covid-19 have been confirmed. But this data is misleading, and is almost certainly a much smaller number than ‘how many people are infected’. Problems with the data are:

Different Case Types Are Combined

There are:

  1. ‘foreign source’ cases
  2. ‘local transmission’ cases

These not only represent very different situations, the data is also collected in entirely different ways.

Foreign source cases are where there has been no uncontrolled spread in country. Usually, these do not even represent situations where infection occurred within the country reporting the statistic. Further, the rules for ‘who gets tested’ are very different from the rules for local transmission cases. So different, that often the rules say, ‘if you are a potential foreign source case, get tested, otherwise do not get tested’.

Local transmission cases are those where the source of infection can not be identified. In most countries, due to the shorting of testing kits and/or facilities, the official rules say ‘just self isolate if in this group’. This means in many places, every local transmission case actually confirmed means someone broke the rules, in order for the case to be confirmed.

Focusing testing on foreign source cases may yield a higher percent of the sample (those tested) returning a positive test, but all cases require the same result in

data is misrepresented, and as a result misleading

So often ‘confirmed cases’ is presented ‘the number of cases’. If you want real data while not testing everyone, you collect samples from each group, and extrapolate to the entire population. These stats are more like ‘we will test the sample and present that as the total number with no calculation based on what percentage was tested’. Data is so often presented as ‘there are now XXX cases’. Not ‘XXX people tested positive from a sample of YYY tested’. The data implies the raw number is the total number of infected people, when it is not. Different testing rules means far different numbers from different countries, should how little the raw numbers may convey.

sample rates are not stated and vary from country to country

Look at the section on different mortality rates, as this gives example of how different sample collection rules radically change the number of confirmed cases. For example in South Korea, testing was free for anyone who wanted it. In most countries, testing is only available in specific circumstances.

distorted collection of data as a policy

Consider the rules for testing in Australia. The same rules also apply in many other countries, because if there is anything ‘herd’ related to Covid-19, it seems to be government behaviour. So this example is quite typical.

In Australia, as with many other countries, there are not enough testing kits to cope with anticipated demand, so the rule is if believe you have symptoms self isolate. You can see a doctor, but unless you have had contact with an already known case, you have travelled internationally, been in contact with people who have travelled internationally, or your case is extremely serious and will require hospitalisation, you cannot be tested.

Current statistics are almost have of all detected cases are local transmission without an identified source. How do they even get detected under current rules? All are cases so serious hospital is required? If statistics already indicate almost half of all cases are this type of case then we are already missing many cases. Each case must have a source, so the number can at least be doubled. Plus since rules normally mean no testing for this group, there clearly must be more out there. The reality is this must actually be the largest group, although with a lower testing rate it will be under-represented in confirmed cases.

If not tested, you must quarantine. Which is the same outcome as if tested positive. Only those tested have the option of escaping quarantine. So the only people who can avoid quarantine are the smaller group that have the least chance of having the result that will avoid quarantine. What does this achieve?

One outcome is that infections related to international travel represent a far higher percentage of confirmed cases. This means the closure of borders should drop the number of confirmed case, as those meeting to criteria to be even tested will drop.

Of course the cases under-represented in statistics, those without an international connection, will continue to climb but still be largely undetected. So governments can produce a win be closing borders in terms of statistics, even if the situation is not really improved at all.

with distorted data on infection rates, making managing the disease a guessing game

How viable is a ‘herd immunity’ strategy? Without collecting the data of actual infections we just do not know. How do we create rules on what number of groups are safe for meetings without knowing how many people are infected? Governments are making so many decisions blindly.

we could and should collect real sample data, but it seems no one does

Consider this, the rule means 100% testing of two groups, and 0% testing of what is now the largest group. Why 100% test any group, unless it is needed to determine medical care? Why not test a percentage selected randomly, and use statistics to determine the overall data? The use the kits saved, to test a randomly selected number from the otherwise untested group, and learn what is actually happening?

For those who have contact with a known case but not serious symptoms, the only consequence is they may isolate for no valid reason. All people who do not have a contact with a known case have this outcome, why give preferential ‘avoid an unnecessary isolation’ status only to one group?

We could use a statistically approach. And we should because that way we would have data.

The End?

How will it all end. Scenarios:

Virus Extinction: Covid-19 Eradication

Viruses can only reproduce by infecting a host, and can only continue to exist if they can find new hosts before their current host eliminates the infection. This means being left of surfaces by their host, or directly infecting another host.

Tests indicate, the SARS-CoV-2 virus which causes Covid-19 can only survive outside of a host for a 3 days, and inside an infected person until they die or recover, which means less than 3 weeks. So the only way the virus survives beyond around 3 weeks is by person to person infection with the three weeks or by person to surface (within 3 weeks) and then to person (within another 3 days). If the chain of infections is broken completely, every copy of the virus would die within one month!

The problem is ensuring it is every single copy of the virus, and the risk or the virus being reintroduce into an area where successfully quarantine previously eradicated infections.

Herd Immunity

If not eradicated, Covid-19 is expected to move from ‘epidemic’ level, to an experience comparable with ‘the flu’, in any region once there is ‘herd immunity’ in that region. Herd immunity occurs once a threshold percentage of the population (around 60%) has already contracted the disease and recovered. The resulting reduction of potential new hosts significantly disrupts spread of the virus, significantly reducing the case load, and thus hospitalisations and fatalities. The disease is not eradicated, but becomes manageable, at least to the level influenza is manageable.

As opposed to eradication, where Covid-19 being eradicated within an area or country prior still leaves that area or country vulnerable to new infections until(if ever) the virus is eradicated worldwide, herd immunity is lasting solution. As a solution, reaching herd immunity is a far more likely outcome than eradication

The Cost of Herd Immunity with Covid-19

Herd Immunity within a country or region is generally thought to require 18 months of exposure of the population of the that region to the disease.

To reach 60% exposure over 18 months, 3.33% (3 and 1/3 %) of the population will be infected each of those 18 months, assuming of the best case of a perfectly even spread. Infections per day of 3.33 รท 30 = 1.11%.

This means for Italy, with a population of 60 million, 60% of the population for herd immunity equates to 36 million. It requires 2 million cases per month to reach the total in 18 months. 2 million cases per month requires over 66,000 cases per day. Currently the worst day is around 2,900 cases, and that case load has already represented a catastrophe not to be repeated. So 66,000 cases per day for every day over 18 months would be an apocalypse. At least, so it seems….

There are no actual confirmed figures on actual infections, just confirmed cases. In a Even if confirmed cases are 1/10 of actual infections, that is still 6,600 confirmed cases per day… which means every day for the next 18 months worse than the worst day in Italy so far. In other words…. getting to the number of cases for ‘herd immunity’ during the ‘first pass’ or 18 month period will, as far as we know, required overloading the medical system, which mean more deaths.

I said “as far as we know” because we still do not know now many infections there are. If we use the ’16x’ multiplier of confirmed cases they calculated for the UK, then Italy is already at the ‘herd immunity’ case load path. If there are actually even more cases, then perhaps herd immunity may be more realistic. We just do not know because we do not collect the data.

Vaccines

Another path to ‘herd immunity’ is by vaccination. People become immune by vaccination, not by catching the disease as part of the outbreak. Some vaccines result in a controlled mild case of the disease to create immunity, while other vaccines boost immunity in other ways and there is no exposure to the actual disease.

For vaccination to be successful, ‘herd immunity’ is required, so the level of around 60% of the population must be vaccinated, or otherwise known to be already immune. The vaccine does not need to be given to those will the most severe health problems, but it must be given to otherwise healthy people. To be giving a vaccine to otherwise healthy people, you must be very certain it is safe.

Newly developed vaccines developed specifically for Covid-19 that have already began trials with still take 12-18 months to be widely available, given the process of proving new medicines safe and bringing them to market. Nothing that could work as a vaccine, even one originally designed for SARS, has been clinically tested to the level required.

Cures or Treatments

There is also the use of medicines already available on the market to treat other diseases and conditions that are being, or may be found to, be successful either in the original form or in combinations against Covid-19. Some research also suggest some success in this area and could result in treatments being available much sooner than would be otherwise as these medicines have already undergone tests and approvals.

Further, cures and treatments do not need to be as safe as a vaccine to start being used. While vaccines need to able to be given to healthy young people who are not ill at the time, a potential cure can selectively be offered initially to those who would otherwise not be expected to live long.

Reality.

Real global eradication is extremely difficult to achieve. While significant work is underway, cures and vaccines take time to get to market and even more time to be available to entire populations. Herd Immunity will take years unless either it comes at the expense of a huge death toll, or infection numbers are higher than any currently accepted estimate.

At this stage, local eradications can keep case counts low but it is yet to be learnt if they will need to be repeated and if so at what interval. Cures and Vaccines can start to play a real roll at best in around 6 months, but it will only be a role.

Herd immunity even at the highest possible level of asymptomatic infections existing in the community but not yet detected will only play a partial role for several years.

Currently it seems difficult to see things settling within a ‘new normal’ in less than 12 months, and even then expect a ‘new normal’, not the old one.

Of course, this insufficient data here for this to be anything but speculation. But it is difficult to see how things can be significantly better in 3 or even 5 months.

Coronavirus: Facts, Myths and Surprises (March 13th)

(go here for link to latest update)

This is a collection of facts on the ‘novel’ coronavirus, together with some discussion on myths and surprises such as:

the debunking and fact exploring on topic of ‘the coronavirus’.

My first post on the new coronavirus (the real threat level) back in late January speculated there could be as many as 180,000 cases by end of February if the rate of spread could not be constrained. The rate of spread has been constrained (and it was around half of my estimate), but not sufficiently constrained to prevent a pandemic, or suggest the threat has passed. My second post spoke of tipping points that would lead to the pandemic, and it turned out that the outbreak in northern Italy constituted tipping point3, not the false alarm of the Westerdam cruise ship. Still, the tipping point did occur, as sooner or later seemed inevitable it would. We, now as this post is written on Friday 13th of March, have a pandemic.

This post is on general infromation on the virus and what has been learnt so far, and I should be able to provide updates over time to this post, avoiding the need for new posts.

Continue reading “Coronavirus: Facts, Myths and Surprises (March 13th)”

Covid-19: Tipping point 3?

The growth of the ‘novel coronavirus’ (there was no official name at the time of my previous post) has been largely as projected in my previous post on the topic. At that time just over two weeks ago the virus had no official name, and while cases have grown from 4,000 to almost 70,000, these cases are still a consequence of ‘tipping point1: the Wuhan wet market‘. Since then there have been reports of two potential tipping points

  • Avoiding Tipping Points?
  • Reaching tipping points
  • tipping point3: Westerdam?
  • tipping point 2: Grand Hyatt Singapore Conference?
  • japan cruise ship
  • tipping point 1 again
  • pandemic?
  • How bad can it get?
Continue reading “Covid-19: Tipping point 3?”

Wuhan Coronavirus: The Real Threat Level

I use the label ‘Wuhan Coronavirus’, but it is also described as ‘novel coronavirus’, and other names including perhaps confusingly just ‘coronavirus’. This is a disease that has managed to raise alert levels before even getting a unique name.

Continue reading “Wuhan Coronavirus: The Real Threat Level”

Blog at WordPress.com.

Up ↑

%d bloggers like this: