What we can expect over the next couple of weeks with COVID-19

by | Mar 24, 2020 | Coronavirus, COVID-19

I understand that many people doubt that COVID-19 is as serious as the media is making out. That it is a beat up. I thought it is worth going through a few facts. Predictions I wrote in my My last post are coming true.

At my time of writing now (7:30 am Tuesday 24 March 2020), Australia has 1,717 known cases of COVID-19, 11 in intensive care and 7 deaths.

These reported numbers become complicated by a few factors

  • Medical advice is that it takes about 10-14 days for people infected with COVID-19 to show symptoms sufficient to identify as needing testing.
  • Chinese research showed that even their robust testing found only 14% of those infected.
  • We don’t have enough test kits. Because of that, we don’t test enough people and so don’t know how many are actually infected.
  • Tests take too long to provide a result. So, infected people get to infect quite a few more before we ‘know’ of their infection.

We have a large number of people in our community who unknowingly have COVID-19 and can and do wander around infecting more people.

I’m going to talk through a few graphs and tables. I keep comparing Australia with what has happened in Italy and France – both countries with medical/health systems that you might imagine could cope. The graph above shows that on 1 March, Italy had 1,710 known cases of COVID-19 (compared with our 1,717 as of today) and 41 deaths.  Just 22 days later, Italy now has almost 64,000 identified infected, 3,204 in intensive care or on ventilators and 6,077 deaths – that includes many medical people. Italy is overwhelmed.
The graph above shows that on 10 March, France has 1,784 known cases of COVID-19 (compared with our 1,717 as of today) and 33 deaths. Just 12 days later, France has almost 20,000 identify cases, 2,082 in intensive care/ventilators and 860 deaths. France is close to being overwhelmed.
You could argue that Australia has fewer deaths (just 7). So we are on a different curve. I hope so. On 12 March, Germany had just 6 deaths. Just 10 days later, Germany has 94 deaths. It looks as though Germany has done a few things that work in slowing down COVID-19. I hear politicians for all countries (including ours) stating ‘we are the best in the world’. Yes, we are all the best in the world.
The table above is complex, but worth a study. The columns need to be explained. ‘France Date’ = the day this data was recorded for France. ‘Days Oz Lag’ = my estimate of the number of days Australia is lagging France. ‘Expect Australia Date’ = my estimate of the date that Australia will have the same data that France had earlier. ‘Infected’ = number of identified COVID-19 cases. ‘Active’ = number of those infected that have not been ‘resolved’. ‘Mild’ = an ‘active case’ with a mild infection. All percentages are a percent of ‘Infected’. ‘Severe’ = number of cases in intensive care or on ventilators. ‘Mild’ and ‘Severe’ are all ‘Active’ cases. ‘Recovered’ = the number (and percentage) that have been classified as completely recovered. ‘Dead’ = dead. ‘Recovered’ and ‘Dead’ combined are the cases that are ‘resolved’.
For example, on 15 March, France had 5,423 infected, 5,284 of who were still Active cases, 4,884 were classed as mild, 400 as severe, 12 had recovered and 127 were dead. I predict that Australia will probably have those numbers on about 28 March. Read down and notice the rapid increase in cases in ‘Severe’ (400 to 2,082 in just 8 days) and ‘Dead’ (127 to 880 in just 8 days). Historic data for Severe and Recovered are difficult to get. I just happen to have kept those few data shown.
The Italy table is similar. For example, on 15 March, Italy had 24,747 infected, of which 20,803 were still active cases, of which 18,931 were classed as mild, 1,672 as severe, 2,335 were classed as recovered and 1,809 were dead from COVID-19. By my estimate, Australia will have those numbers on 5 April.
Looking down that table, notice the rapid increase in severe cases (needing intensive beds or ventilators) (from 1,672 to 3,294 in 8 days) and dead (from 1,809 to 6,077 in just 8 days). Also notice that the estimates from France and Italy do not neatly line up. There is a 1,000 difference in my estimate for number of dead on 5 April. I’m sure that difference will work itself out over the next week. I think France might be on a slightly flatter curve. Also, compare the percentage of dead between Italy and France.
I hope we are following a curve closer to Germany – which unfortunately I have not tracked. It is difficult to get historic data for severe and recovered. We will see.
It is worth looking through the percentages. I know that some parts of social media are reporting that most people ‘only have a mild infection’. True. However, look at the very small percentages that are classed as ‘recovered. ‘Mild’ cases are still ‘Active’ and appear to ‘progress’ to ‘severe cases’ (where they overwhelm ICUs) or ‘Dead’.
BTW, those people in the ‘severe’ columns, they are now in intensive care. Their path from ICU is either to death or a long and difficult recovery because their lungs have been partially destroyed by their infection and subsequent pneumonia. They may never be able to function or live as they did before their infection.
We know from reports from all countries with more than about 1,000 in ‘severe’ that smaller ICUs are quickly overwhelmed and triage very soon ensures that older people (or people without dependants, prisoners, wrong colour, sex or other arbitrary categories) are very quickly classified as ‘black tag’ – not worth saving – ‘do not treat’ except maybe palliative care.
We don’t have enough staff with ICU training to look after those people in ICU. Nor do we have enough Personal Protective Equipment for those staff. So, the few ICU staff we do have also become infected and die. Exacerbating the problem.
What to do?
My last post ‘Australia needs to lock down now. Close shops. Restrict people to their residences. Introduce methods of early testing. We must slow this down before we are overwhelmed.’
Stay inside. We can see that full isolation works – so isolate yourself. Do not go out except for food, medical. Do not meet or allow into you home anyone who does not live with you. Do not be in a group larger than 2 – and then 2 metres away. Else, you will soon become a number in one of my tables. This is what we mean by ‘slow it down’.

Shield Shrimp

When it rains across Australia’s vast inland region, temporary pools crop up all over the arid ground, giving life to a strange desert crustacean known as the shield shrimp (Triops australiensis).

Named after the formidable carapace that shields its head and upper body, T. australiensis can grow up to 7.6 cm long, and it uses its long, segmented tail and mass of 60 or so legs to propel itself through shallow water.

It also breathes through these legs – its sub-class Branchiopoda means ‘gill-legged’ – and in the females these legs bear ovisacs for carrying their tiny eggs.

Several pix in the Photo Gallery and a movie.

Acacia peuce

A rare and endangered plant. The tree grows up to 15 to 18 metres (49 to 59 ft) high, with short horizontal branches and pendulous branchlets covered in needle-like phyllodes adapted for the arid dry climate. It has a distinctive habit more similar to a sheoak or a conifer.

Although speculated to have been widespread across central Australia during wetter climates 400,000 years ago, the population is now mostly restricted to three sites, separated by the encroaching Simpson Desert. In the Northern Territory, the species is restricted to the Mac Clark (Acacia peuce) Conservation Reserve which is surrounded by a pastoral lease, Andado Station. The other two sites are near Boulia and Birdsville in Queensland. The tree is found in open arid plains that usually receive less than 150 millimetres (5.9 in) of rain per annum. They grow on shallow sand aprons overlaying gibber or clay slopes and plains and between longitudinal dunes or on alluvial flats between ephemeral watercourses.

 

Owen Springs Reserve on Hugh River

Owen Springs was a station on the Hugh River. The Hugh River flows into the Finke (when it actually flows). Both cut through the Western MacDonnell Ranges. The image above shows Owen Springs Reserve as a dot at lower right. The river it is next to is the Hugh. Hermannsburg, our next town, is near middle left edge. Hermannsburg is almost on the Finke River. You can see both Hugh and Finke Rivers cutting through sections of MacDonnell Ranges.

Palm Valley

Palm Valley is within the Finke Gorge National Park southwest of Alice Springs. Palm Valley has a smallish population of Red Cabbage Palms (Livistona mariae). The nearest related species is 850 kilometres away in Katherine NT. The average rainfall for Palm Valley is just 200 mm per year. Small pockets of semi-permanent spring-fed pools allow the unique flora and fauna (desert fish, shield shrimps tadpoles and frogs) to survive.

It had been assumed that the cabbage palms were remnants of a prehistoric time when the climate supported tropical rainforest in what is now the arid inland of Australia. Genetic analysis published in 2012 determined that Livistona mariae at Palm Valley is actually the same species as Livistona rigida from samples collected near Katherine and Mount Isa, both around 1,000 kilometres (620 mi) away. It is now thought that aboriginal people brought the palms to here from Mataranka.

Mound Springs

Mound Springs occur around the Western edge of the Great Artesian Basin and represent a natural discharge of Artesian water that was captured many hundreds of kilometers away from rain falling along the Great Dividing Range and New Guinea. This article provides details. Dalhousie is an excellent example of a mound spring.

Great Artesian Basin map Great Artesian Basin diagram