Communicating risk in an epidemic: a personal perspective

Updated on March 18th 2020, written by Dr David Owens.

In 1986 I was a junior doctor working in a chest unit. We were seeing an increase in a type of pneumonia caused by the relatively new disease of AIDS. There was a genuine fear in the population. LGBT individuals were stigmatised and blamed for this new epidemic. The “gay plague” as it was prejudicially labelled in the media. At a conference I attended that year, a senior AIDS specialist made a prediction that by the turn of the century, in the year 2000, 25% of the world’s population would be dead of this new disease. Fortunately, he was wrong.

The reasons that he was wrong remain as pertinent today as they were during the SARS epidemic of 2003. The dilemma of predictions based on early epidemic data is explained here.

I have described in another article the importance of understanding new infectious illnesses in the context of both the uncertainties surrounding the disease and the uncertainties surrounding the epidemic. There is a tendency to conflate disease severity with infectivity. These two distinct processes deal with very different questions:

Information about the disease gives information in answer to the question: What happens if this illness infects me or my family?
Information about the epidemic answers the question how likely is this condition to impact me and is that risk changing over time? This is the question that deals with risk.

This concept feeds into a built-in glitch in the software of the human brain. It seems that we are wired to amplify certain fears and mix up severity with risk.

Just think about our fear of shark attacks, highest annual recorded deaths worldwide being 98 in comparison to mosquitoes which kill more than one million people every year (WHO)¹. It is intuitively logical to postulate an evolutionary advantage in our offspring avoiding spiders and snakes. There is likewise something primal about the fear of epidemics. The lessons from historical epidemics may or may not be hard wired but it is certainly deeply embedded in the human narrative.

The media feed off this glitch. Try replacing “killer virus” or “lethal disease” with ”killer car” or “lethal taxi”. Obviously risk changes in epidemic illness but at the time of writing there have been 7,805 deaths from COVID-19 in the last 2 months compared to an estimate of 1.25 million deaths in road traffic accidents globally per year (WHO estimate)². Maybe a better comparison, if we are looking for a baseline against which to measure comparative risk is the 22,000-55,000 deaths from influenza in the USA so far this Winter³. So how do we deal with this processing problem that seems to be hard wired into the human psyche?

The Anxiety Epidemic

Anxiety is highly infectious. It is much more infectious than the COVID-19. The greatest triggers of anxiety are lack of control and uncertainty. A common feature of an anxious states is negative interpretation with catastrophising thought processes. Treatments of anxiety include challenging the prevailing negative interpretations. Essentially this means challenging a way of thinking and asking if there is another explanation. Information and education are powerful antidotes to epidemics of anxiety. They give a paradigm in which we can reestablish a sense of control.

Our understanding of decision making under states of stress has a solid evidence base. The field of behavioural economics has increased our understanding of the predictably irrational nature of our thought processes. Education and information assists in control of anxious thoughts and leads to better decisions.

How is this relevant to the COVID-19?

It is my opinion that as with HIV in the 1980’s and SARS in 2003 given adequate, balanced information that most people will feel less anxious about the current situation.

I have summarised the key evidence of our evolving understanding in terms of:

The disease
The epidemic

This is part of the summary sheet which will be continually updated as evidence accumulates.

The current evolution of the epidemic is suggesting a disease which is likely to continue to occur in clusters. There will be more new cases and there will be more deaths. This is to be expected and must be understood in context as the epidemic unfolds. Try to compare the numbers involved against a baseline. This allows a better assessment of changing risks. For example, the CDC data equates to between 90-220 deaths per day from Influenza in the USA this Winter. The numbers in China may reasonably be expected to be 300-700 deaths per day.

We will accumulate more data and greater certainty over the next few weeks. The paradox is that the more widespread this illness the milder it is likely to be in terms of severity on a case by case basis. The case fatality rates are trending down and it seems likely that this disease is generally not as severe as initially thought except in certain individuals who are vulnerable. The majority of cases are occurring in older patients, more in males and most have co-existing illness.

I wish to emphasize again that COVID-19 is not a type of influenza. We have both vaccinations and treatments against influenza. However, decision making is improved when we have a baseline against which we can anchor our analysis. Therefore, using influenza data as a baseline the COVID-19 is trending in severity and currently seems to be more severe than influenza, at least for some people who catch it. This may change as we accumulate more evidence (depending on the number of milder cases it may end up being the same or even milder). Like influenza it mostly affects older people who also tend to have other illnesses. How likely you are to catch this illness will depend on how widespread the epidemic becomes. At the moment the numbers remain very small 190,152 in comparison to the estimated 36-51 million cases of influenza so far this Winter in the USA³. This is likely to change and the number will grow. How quickly it changes will give you an objective measure of changing risk

I also wish to again emphasize our support for the public health measures. This remains a severe potential global health threat. A disease which is less severe than first thought can still place a huge burden on hospital resources. The early public health measures can significantly delay or prevent the progression of the disease into communities with less developed health systems.

The opinions above are a personal reflection based upon 30 years of experience working as a doctor in Hong Kong in addition to a personal review of the available literature. There are still uncertainties in the evolution of the epidemic. I will be continually updating the summary sheet, including academic references, as evidence accumulates.

By Dr David Owens

To learn more about the differentiation between a disease and an epidemic, click here.