The Jordanian government, along with several other governments, rushed to adopt contact-tracing applications to help epidemiological investigation teams identify the chains of infection for COVID-19. There are slight differences in the design of these applications, but they share the same operating mechanism: they rely on Bluetooth or GPS to determine if an infected person has met other people during the infection period, and accordingly, the concerned persons or authorities are informed so they can conduct the necessary checks for these people. Despite how appealing this idea sounds, its benefits are, in fact, very limited, and therefore will most likely prove ineffective. Rather, it may yield negative results that could outweigh its desired benefit. Thus, is this idea worthy of being treated by the government as an essential part of the solution under such circumstances?
The Jordanian government has consistently called on citizens to use the "Aman" app, which was created as a voluntary effort that we highly appreciate. But it is necessary to take a moment to evaluate the desired benefit of this app, and to achieve that, we will highlight two aspects of it. First, the feature that identifies and saves users’ time and place for a period of 14 days with the ability to deliver this information to specialists upon testing positive for the virus. This aspect seems useful if this information is used to help the patient –after being diagnosed– remember the places they visited during the period of infection, and to identify the people they came in contact with, in case they forget. This may assist epidemiological investigation teams in their work and improve the quality of the results, albeit to a limited extent.
The other feature that the application promises is to automatically identify networks/chains of contacts and inform them of the possibility of being infected with the disease, which technically and practically seems fictitious because the technology used cannot achieve the desired goal with sufficient accuracy. According to the World Health Organization, the disease spreads either directly from person to person through droplets transmitted through coughing, sneezing or talking at a close distance, or indirectly by touching contaminated surfaces and then touching the eyes, nose or mouth. The two most effective and frequently recommended protective measures are social distancing (a distance of two meters) and frequent hand sanitation. If we assume proximity as the only criterion for infection risk assessment —which is what these apps assume— then this criterion and its error rate must be examined.
First, we must remember that not everyone who comes close to a sick person is necessarily infected. Since droplet transmission is a necessary condition for direct infection, the likelihood of infection is low when coming close to a patient for only a few seconds because the chance of droplet transmission is minimal. What makes matters worse is that the use of GPS (especially under roofs) or Bluetooth has an error range several times greater than the distance that allows disease transmission, and that -in addition to the above- will greatly distort the results and hinder their credibility. We must also remember that these applications completely neglect the important factor of surface contact, as they cannot identify contaminated surfaces in any way. Thus, the hypothesis of disease transmission proposed by the application is subject to false positive and false negative results on a large scale, in a way that makes us lose confidence in the automatic judgment of the application. If —God forbid— the virus spreads, I do not believe that the epidemiological investigation teams will examine anyone because of the result of the application alone or exclude anyone who was not identified through the app.
But we might ask ourselves: Why not? It is still a plus, isn’t it? Here we must look at and analyze the potential downsides of using the app, the first of which, and as a result of the inaccurate estimation of contact, is that the user may panic when the application alerts them of danger, and this can affect their daily life and work without any logical justification. Also, users who have not been alerted of any danger may acquire a false sense of security and begin to ignore important guidelines such as social distancing and hand sanitizing, which ultimately exposes them to a higher risk of danger. We also can’t ignore the risk of negative exploitation of citizens' sensitive personal information, especially since the information security and privacy guarantees currently presented in the “Aman” app do not seem sufficient or satisfactory.
In this regard, I must stress that I do not see a defect in the existence of the "Aman" application in itself as an independent voluntary effort. I am merely surprised at the government's repeated call to use the application as a substantial part of the solution, regardless of the technical analysis and global experiences that tell us otherwise. In addition to all of the above, a simple comparison with other countries shows us the inability to reach a satisfactory usage rate for contact tracing applications even if we overcome all their other challenges. So why distract citizens with them instead of directing them to more effective solutions?