In recent weeks the vaccination campaign has been rocked by reports of adverse effects related to the Oxford / AstraZeneca vaccine. As of April 4, 169 cases of cerebral venous sinus thrombosis and 53 of splanchnic vein thrombosis had been detected in the European Union and the United Kingdom. Until that date, 34 million people had received this vaccine.
On March 15, the Ministry of Health decided to suspend vaccination with AstraZeneca, but on March 18 the European Medicines Agency issued a statement in which it established that the benefit-risk balance remained positive. Following that report, vaccination was resumed.
In this scenario, health authorities, experts and the media have launched a campaign to improve public confidence in the AstraZeneca vaccine. The messages are focused on reminding us of the small probability that there is of having a thrombus or of dying as a result of it. This strategy assumes that we have an objective measure of risk and that this is perceived by all subjects equally. In other words, it combines the consequences of risk decisions with the probability of their occurrence.
Empirical studies show that subjects do not behave like this.
People do not behave the same in the expectation of losses as of gains. To illustrate the latter, let’s consider a game in which a coin is tossed: if it comes up heads, you lose 50 euros; if tails, you win 75. This is a positive game, since the gains are greater than the losses, and the probability is the same, but the theory of perspectives tells us that the subjective valuation of a loss is greater than that of a profit.
An alternative paradigm is the psychometric theory of risk. It states that people overestimate the risks associated with infrequent, catastrophic, and unintended events; and they underestimate the risks associated with frequent, family and voluntary events.
Specifically, the theory considers nine factors that influence people’s assessment of risk:
Knowledge on the part of science.
Knowledge on the part of the subjects that they are exposed to a risk.
Whether the risk has a catastrophic consequence or not.
Whether the risk is common or not.
The severity of the consequences.
The immediacy of the effect.
In the messages that are launched from the media, experts and health authorities resort to well-known risks so that citizens can assess the magnitude of the risk they run with vaccination.
Thus, the risk of suffering thrombi associated with contraceptive pills (500-1 200 per million), or of suffering thrombi due to the effect of smoking (1763 per million) is discussed compared to the probability derived from the AstraZeneca vaccine (0.5 per million) or having a thrombus as a result of contracting covid-19 (165,000 per million).
These comparisons assume that the risk associated with the different events is comparable, but if we pay attention to psychometric theory, we see that the perception of risk does not work that way.
The risk associated with tobacco is voluntary and long-term, so it will be underestimated by the subjects. The risk associated with the vaccine is involuntary and immediate, therefore, it will be overestimated. The same is true of the birth control pill.
Another factor that influences the perception of risk is the extent to which it is known to science, where unknown risks are overestimated and known ones are underestimated. The novelty of the risk or the knowledge of the exposure to it also influences.
The relationship between thrombi and the AstraZeneca vaccine ticks all the boxes for your risk to be overestimated: it is new, unknown to science, unintentional, and potentially catastrophic.
More plural voices are needed
Throughout this pandemic, a greater plurality in the voices of experts who give advice to the authorities and citizens has been missed. Of course we need epidemiologists, public health specialists, virologists, physicians, and economists. But we also need psychologists who anticipate what the public’s reaction will be to certain measures or messages, or sociologists who advance the behavior of groups. Above all, we need experts to talk to each other.
On the other hand, we consider that it could be interesting to rethink the strategy followed in relation to risk communication in the media. Of course, it is not necessary to hide information from the public, nor to sweeten it, but to measure to what extent, when it is reported for the first time, interpretive biases are taken into account.
Scientists are generally not very aware of how the information is to be read. They have been trained to perceive the real risk, so it can be an effort to try to put themselves in the heads of those who are not used to handling statistical measures and comparing them with others.
The figure of the scientific communicator is becoming more and more necessary, an expert who helps in the relationship between the community of scientists and the public. We might think that this role is already being played by the press, but we see that this is not the case, and that it is instead contributing to the creation of a sense of vulnerability among some people.