Covid19 Reflections #23

Insight

Today the Australian Government talked about more modelling that would start to inform its strategy for lifting the lockdown. One particular piece of research informing that is well worth reading – this piece from the Doherty centre which talks about two key issues – how many symptomatic cases of Covid10 are being detected, and the transmission rate of Covid19.

As of 9th April 2020, our estimate of the symptomatic case detection rate for Australia is 93% (95% Confidence Interval 77–100%). The corresponding estimates for each state/territory are all greater than 80% (Figures 1 and 2)…

…These analyses produced broadly consistent results showing that the effective reproduction number is likely less than 1 in NSW, VIC, QLD, SA, and WA as of 5 April 2020 (Figures 3–5). It should be noted that these estimates are averaged across the whole of each jurisdiction, and may reflect Reff >> 1 in a number of localised settings and Reff << 1 elsewhere.

First, the authors look at whether we are detecting symptomatic cases. Using the current death rates from Covid19 (and adjusting for the time differential between when you test positive, and your death or recovery from the disease) they compare Australian death rates to Chinese death rates and conclude that we are finding the vast majority of symptomatic cases. While the underlying death rates (technically Case fatality rates (CFR) – those who die from the disease after they have had symptoms of it) are still subject to wide ranges of estimates, this seems a good way of testing how good our testing regime actually is.

Second, the authors look at the effective reproduction number of the virus R(0). For an explanation of this number, you can’t go past Angela Merkel.

This is how Angela Merkel explained the effect of a higher #covid19 infection rate on the country’s health system.

This part of today’s press conf was great, so I just added English subtitles for all non-German speakers. #flattenthecurve pic.twitter.com/VzBLdh16kR

— Benjamin Alvarez (@BenjAlvarez1) April 15, 2020

Basically if the R(0) is less than 1, then each infected person infects less than one person, on average. But if it goes over 1, then the outbreak just keeps getting bigger. And as Angela Merkel explains, the higher the number of people infected from each individual infection, the quicker our health system is overwhelmed.

Right now, in Australia, according to the study I linked above, the R(0) value for each infection is well under 0 in every state except Tasmania (and that outbreak is small enough that the range of possible outcomes is quite wide). In NSW, where I live, the paper suggests that R(0) is around 0.5 -two infected person in the community will between them infect only one other person.

All three methods indicate that the COVID-19 epidemic in Australia is currently being suppressed sufficiently strongly to achieve Reff<1. If current measures were sustained indefinitely, and in the absence of imported cases or localised clusters, local elimination may be achievable. However, this effort could take many months, and would be dependent on the initial number of cases in each jurisdiction.

In NSW, with around 1,500 active cases, we would need around 8 generations of transmission (say a week each) before the number of infected people was below 10.

Eventually, if R(0) is under 1, then the virus will die out, as gradually each person will not infect anyone else. The lower that number the quicker the virus will die out. SARS and MERS were both diseases where the initial R(0) value was above 1, but rigorous non pharmaceutical interventions (social distancing, and very careful use of personal protective equipment in health care settings) reduced the level so that the virus died out.

Right now, according to the respective prime ministers, Australia is adopting a strategy of suppression – effectively keeping R(0) below 1, but hopefully cautiously opening up while keeping that number below 1, so that the economy can restart. New Zealand, though is adopting a strategy of elimination – not opening up the economy until the virus has all but died out, so that full opening up (within New Zealand) is possible, as there will be no virus to infect people.

Link

The insight above looks at what proportion of symptomatic cases we are detecting with our current testing regime. One of my colleagues on the Actuaries Institute Covid19 working group, Douglas Isles, has looked at how many unsymptomatic cases there are out there. His conclusion is that there a lot more than you might expect.

While it relies on crude assumptions, the simple model^[iii] output suggests at 9 April there may have been around 20,000 carriers of COVID-19 in Australia, rather than the 3,000 or so active confirmed cases as reported^[iv].

This means the community rate of infection could be about 6-7x that the media is reporting. While, this is a simplistic exercise, the key point is that the reported data is an order of magnitude less than the actual data.  With access to more granular data, particularly around the demographics of those tested and the dynamics of the viral transmission, these estimates could be improved.

Douglas is advocating random testing, so that we can have a better sense of how much Sars-Cov2 is really out there. While I have read that there is some evidence that asymptomatic carriers do not spread the virus as much as those with symptoms, any spreading still has the potential to restart the exponential growth we have been fearing (and so far avoiding). That is also something the Australian prime minister is advocating – I suspect any actuary reading this is advocating random sampling even more urgently based on our understanding of statistical methods.

Life Glimpses

In the last few days there has been a flurry of confusing information about school opening. Here in NSW, Gladys Berejiklian announced that schools would reopen to students in the third week of Term 2 (that’s May 11 – in three and a half weeks). She said that

The return to classroom teaching will be accompanied by new social distancing measures to protect teachers and students.

I haven’t checked the local parents’ reaction, but based on past reactions, there are two fairly big issues here. One is that our local school (like many public schools in this area with demountable classrooms because there aren’t enough classrooms for all students) cannot practically exercise social distancing if all students are on site. If you go to our high school when the bell goes to change periods, you would be reminded of rush hour on Wynyard station.

The other issue is soap and hand sanitiser. There just isn’t enough. And based on the experience in term 1, I don’t have confidence that the education department will be able to provision all the schools that need it with enough soap and hand sanitiser near every classroom and toilet block for students to maintain hygiene as well as social distancing.

There is a month to go. But the view in this house is that if you think face to face education is really important, then prioritise the year 11 and 12s, and be serious about social distancing. In the meantime, it is still school holidays, and all of us (including the school student in the house) are enjoying the opportunity for our school student to sleep in, and not feel too pressured to complete a full day of school work each day.

Bit of Beauty

Today’s picture comes from our local park, where the cockatoos were quite agitated yesterday (too many locals exercising?) This one was willing to sit still to have her picture taken, though.