Our Vice Chancellor provides us with regular – perhaps overly regular – updates with respect to what the Covid-19 virus means for the University and with respect to what the University is doing in regards to teaching students who have had to remain in China. We also get multiple emails about ensuring the health and safety of staff across our three campuses. In a recent missive the Vice Chancellor let us know that people have all but stopped eating at Chinese restaurants and he opined that the virus is not racist and that nor should we be racist.
The trouble with this perspective is that making the decision not to eat at Chinese restaurants is not necessarily, if at all, a racist decision. Rather, the decision could just as well be a sensible decision that might be made on the basis of the available facts along with taking into consideration facts that cannot be known. The same decision to not eat at particular restaurants would also apply to, for example, Italian restaurants as Italy is now a country rampant with the virus. We might also add Korean restaurants to the list along with Middle Eastern restaurants given the alarming spread of the virus in Iran.
So, in terms of the available facts, we know the countries where the virus is most prolific and we also know the fatality rates in those countries. For example, at the time of writing real time figures indicate that 80,071 people have contracted the virus in China with 3,193 deaths giving a fatality rate of 3.9%. Italy appears second in the Covid 19 data table with 17,660 case of infection and 1,266 deaths giving a fatality rate of 7.2%. Iran has reported 12,729 cases with 611 deaths giving a fatality rate of 4.8%. South Korea has 8086 cases with 72 deaths giving fatality rate of 0.9%. I could continue with examples from other countries but the point is that these are facts that we know.
An article on the closure, and in some cases, liquidation of Chinese restaurants provides figures in Australia for the number of people who had contracted the virus as of March 4th 2020. At that time, Australia had 33 confirmed cases of the coronavirus. 15 of the 33 cases were either Chinese tourists or residents recently returned from China. So 45.45% of cases are accounted for by Chinese people. Notably, given the huge increase in virus figures for Iran, 6 of the cases or 18.18% of all cases were accounted for by people from Iran. So useful data is available with respect to the prevalence and transmission of the disease.
However, it is also the case that a lot of the evidence that we would need in order to make completely rational choices about whether or not to eat at particular restaurants is simply not available. For example, I have reported previously on figures showing that since mid-February 2020 (it is now March 14 2020), 31,000 Chinese students have entered Australia and it seems more than fair to conjecture that a number of these students would be working in the hospitality industry. However, we have no way of knowing whether Chinese staff working at Chinese restaurants have been in contact with people in China who have the Covid-19 virus.
In the absence of being able to make a completely rational, evidence based decision with respect to where to dine, the rational choice is to decide not dine in restaurants offering a cuisine from a country that has a significant incident rates of the virus, particularly Chinese restaurants which may be employing staff who have recently returned from China. This is simple logic. Why take the risk that the waiter at your restaurant of choice has recently returned from China. So if I ever did eat at restaurants, which I do not, there would be no Chinese food.
For those who need to think further on this point, the question of where to eat, and indeed a whole range of decisions with respect to the Covid 19 virus, might be made by engaging in a simple risk analysis exercise. This form of analysis is used extensively by managers and leaders in all kinds of organizations to gauge the probability of a risk eventuating and the impact of that risk in the workplace should it eventuate. The point of carrying out the exercise is to manage the risks that are identified so that they will have minimal impact should they eventuate.
For example, an online shopping service might analyse the risk of its servers crashing in terms of the probability of that risk occurring and in terms of the impact of that risk should it eventuate. The likely outcome of an analysis in this case would be low probability because the servers are crucial to operations and would, therefore, be regularly maintained and tested. There would also be multiple back up servers in multiple locations and the company would immediately switch to these alternate servers if the primary servers crashed. However, the impact of not having any available servers would be high because the shopping service would have no online presence.
So, here’s how it works in detail. For any particular risk, one analyses the likelihood or probability of that risk being realized. The values attached to the probability are low, medium and high. Numerical values are also attached to the three levels of probability. The actual numerical values do not matters so let’s assign the values of 3, 6 and 9 to designate the probabilities from low to medium to high. Next, one attaches values to the perceived impact of the risk ranging from low to moderate to severe. Again, one can attach numerical values to the impacts so we will again use 3, 6 and 9 to designate the impacts from low, to moderate, to severe. Now, all that one needs to do is make an estimate of the likelihood of the risk being realized along with an estimate of the impact of that risk should it be realized.
Let’s take the restaurant example. The most reasonable judgement is that the likelihood of catching the virus from eating at a Chinese restaurant is low. This judgement is made on the basis of the analysis of evidence outlined in the previous paragraph. However, the perceived impact of the risk can vary considerably. If young and in good health then one might rate the impact as potentially low because statistically speaking one would likely fall into the 97.3% category of people who merely develop mild flu symptoms followed by recovery from the virus. However, if advanced in years with respiratory problems then one would likely judge the potential impact of the risk to be severe because older people with health problems are more susceptible to developing the full blow virus and dying as a result.
Numerically, the first scenario – low probability and low impact – would warrant a risk rating of 6. The numerical rating for the second scenario – low probability but high impact – would be 12. And so there it is, an entirely rational way to approach various Covid 19 risk situations. Notice, however, that the decision itself with respect to managing the risk involves an element of judgement. For example, scenario 1 speaks to a low risk situation. However, one may quite sensibly still decide not to eat at Chinese restaurants because there is actually no need to do so. That is, one does not have to take the risk. The judgement in the second scenario is more straightforward. It would be stupid to eat at at a Chinese restaurant given the potentially severe impact of becoming infected with the virus.
So, now that you know the risk analysis strategy – if you did not know it before – you can utilize it more broadly in your Covid-19 decision making processes. For example, is it really worth attending mass gatherings where it is entirely possible that you might become infected with the Covid-19 virus. The likelihood of in Australia of becoming infected is currently low but the impact could be high. A second example. Is it really worth the risk of going to the supermarket everyday to try to stockpile vast quantities of toilet paper or would it be better, that is less risky, to buy 10 sponges on a single occasion in order to practice an alternative form of self cleansing. The answer to that question is, I hope, abundantly clear.
The risk analysis scenario can be applied to stocking up on foodstuffs. Note that this is an entirely different approach from the panic buying that is currently occurring in Australia. My risk analysis looked at three factors. First, what is the risk of Australian supermarkets “running out” of supplies due to supply chain problems. That risk is non existent. Secondly, what is the risk of shoppers panic buying and stripping supermarket shelves of the supplies that I would need to survive for three months. I categorised that risk as high probability and high impact. Third, what is the likelihood that the virus will become rampant in Australia? I could not specify the likelihood but I categorized the impact as high.
Next I set about mitigating the risk of not being able buy what I need at the supermarket. I started by identifying all the foodstuffs and beverages that I would need in order to survive for three months without having to go shopping. As of today, I purchased everything that this article recommends purchasing with the exception of getting hold of fresh meat, chicken and fish, which might then be frozen. The reason for not purchasing frozen food is that my freezer is tiny and so I plan to buy a chest freezer at some point in time and to fill it week by week with meat, poultry and fish.
The other really useful bit of information in the article for taking a planned approach to stocking up on food and beverages, which will likely be ignored, is to take a stock count in one’s pantry to determine what one will actually need should self-isolation become a reality. Whilst useful, this piece of information really does not go far enough. Firstly, one should operate from a worst case scenario basis and assume, for example, that we might all have to isolate ourselves for 12 weeks. Next one should figure out exactly how many meals – breakfast, lunch and dinner – would be required during that time. Having determined the number of meals required one should then plan those meals in terms of the ingredients required for each and every meal. Not really that difficult.
For example, tinned tuna, tinned tomatoes, Italian herbs, black olives, capers and spaghetti to create what would be a delicious dinner. Sure, garlic bread would be delicious but it wouldn’t last very long unless one could put it in a freezer. So, that’s one meal. Now do the maths assuming a single person for each meal. 12 weeks multiplied by 7 days multiplied by 3 meals per day equals 252 meals. With a single person, the pasta delight might last for four meals. However, a family of four would consume the meal in one sitting. Maths again. A family of four would obviously have to cook meals of a sufficient size to satisfy all the members of the family.
To my final point, and to something that I will enjoy doing this evening. Simply stockpiling a ton of supplies is just plain dumb. All the supplies need to be entered into a spreadsheet with relevant information including the expiry date of each product. That way, when the virus is over, it will be easy to ensure stock rotation so that the supplies will still be edible the next time that a virus hits. This fact brings us full circle to risk analysis. Covid-19 is one in a reasonably long line of pandemics and it would be naïve to think that there will not be another pandemic. There will be and the most rational course of action is to plan as though this will be the case. This entails maintaining stocks of food and water that will mean not having to leave the house for at least three months when the next pandemic strikes.
First Published March 14th, 2020