How to use methodology and technology to make the benefits of masks understandable to (political) decision makers
As of Monday (27.04.2020), facial masks will be compulsory to wear throughout Germany when shopping and on public transport. Until three weeks ago, the benefits and thus the necessity to wear masks were clearly questioned. But why the change of mind and why now and what does this have to do with Field 33?
There have been countless discussions for days about the pros and cons of the obligation to wear face masks in public. Some people think that they would definitely help to prevent the spread of the virus. Others think that masks are useless. Also, they would suggest a non existent safety and thus lead to carelessness in other protective measures, such as distancing and hand washing, and thus would not reduce but even increase the infection rate. Moreover, masks are not available and the few that do come into circulation must be reserved for medical staff. An obligation to wear one would result in further shortages through hoarding.
We get the impression that the governments at federal and state level are performing a scientific balancing act and, under a lot of pressure from the media and public debate, are making decisions on a fuzzy basis, with little reliable data and far too late. Suddenly everyone is an expert, but nobody really knows anything for sure.
The problem lies in the complex interrelationships between biological/viral, economic, social, political and media patterns of impact. Many participants of the dialogue are pushed to the limits of their own decision-making abilities by sheer complexity. Neither objective data nor simple cause-effect relationships are widely available. In addition, the uncertainty and economic implications that are inherent in the COVID-19 perspective promote the human trait of thinking in a linear and one-dimensional ways. Simple relationships that can be described with “if, then…” or “yes/no” are much easier for our brains to grasp than complex and non-linear questions that often contain variables that also change often. If the relevant connections then also have a cross-domain effect, we stare at the media and marvel at the dialogues between politicians, virologists, economic experts and practitioners from real life. We call on Markus Lanz and Sascha Lobo for help or simply refer to the uncertainty that lies in the unpredictable and unprecedented circumstances and the given necessity to be lenient with the quality and speed of insights and decisions.
It is actually easier and, from our point of view, better: With the help of modern methodologies and technology, causal relationships can be identified and mapped quickly and easily. They help to create transparency and understanding across domain boundaries. Different angles sharpen the view and allow a holistic and thus qualitatively better representation of the overall system and also a better interdisciplinary discourse. Once provided with data and functions (knowledge), these networks of causes and effects can be configured and continuously optimized. If one observes the characteristics in the structures over time, one can learn from them and thus they are suitable for simulations and predictions of the effects of individual measures or events. With the help of artificial intelligence, gaps in data availability and non-obvious relationships can be identified and closed.
This makes complexity manageable and by embracing it, transformational decisions can create great opportunities.
Let’s assume that the mask question is exclusively about the best possible protection of the population [result] and let’s also only consider the measure of wearing a mask [basic conditions]. Then there are really only a few relevant [domains] and [influencing factors] that we have to take care of in order to be able to control the issue objectively at any time. These include
- the mask with its types, its user and application and its effect on the infection rate
- their availability over time and possible alternatives (e.g. scarves or self-made masks)
- and the decision and communication on how to deal with it and thus the behaviour of the population.
Of course, we at Field 33 are not experts in the wearing of masks, their effects or even their acceptance. But with the help of the cause-and-effect overview assembled from the above-mentioned components and the available scientific findings (4, 5, 6, 7) the perspectives of the various discussion participants and decision-makers can be followed much better and completely orgy-free.
With a simple configuration with population (1), lethality (4), mask effect when 1:1 (3) and infected (2), infection rate and death rate (3) on March 1st, April 1st and April 24th we come to the conclusion, that
If we had introduced compulsory masks at the beginning of the discussion on protective measures, 2,880 fewer people in Germany would have died of COVID-19 by 24.04.; with its introduction on 01.04. there would still have been 1,822 fewer deaths.
And again — of course we are no experts for masks and of course there are much more aspects relevant for a complete view. We have taken into account those we could find from the public debate and the sources listed below. A complete overview can be found below (9).
But even if, due to lack of availability at the time of introduction or because of rejection by the population, the full effect of wearing masks would not have been achieved initially, our model can be used to work on specific measures and to further optimise the structure and its explanatory contribution. Regardless of how conservatively we set the parameters, the number of fewer deaths at earlier introduction is always positive, without disproportionate economic or social restrictions. And that is why we have been wearing masks in public for days.
Please contact us directly at masken-orgie@field33.com, if we have forgotten or misrepresented elements and relationships from your point of view, or if you have input on the configuration mentioned.
More and more we come across complex questions which are handled with the help of inadmissible simplifications and thus do not deliver satisfactory results and decisions. Over time, this destroys the trust of those concerned and the motivation of potential contributors.
Just take the often discussed digital transformation of large organizations and the error rate of such initiatives of over 80%. An approach like ours above would not only have saved the job of several chief digital officers (8) but could also lead to sustainable, share price-relevant success on the basis of productivity increases, increased customer benefit and satisfaction, employee motivation, time to market etc.
What we have observed so far when dealing with complex issues:
1. yes, “it depends”, but that doesn’t mean you have to leave the discussion to the experts.
2. maps of cause-effect relationships show relevant veins from causes to overarching effects. This helps to invalidate incomplete arguments. At the same time, however, they also show gaps (white spots) in the explanatory model or with regard to available data. And thus helpful individual effects and sensors for them can be identified in a targeted manner, the answers to which quickly provide better results in the overall model.
3. visualisations of cause-effect relationships make it easier to get started on a topic and thus allow significantly more participants and thus perspectives to be included in the development (intelligence of the crowd).
With the help of our enterprise meta model and modules for
- Digital transformation including delivery performance, software quality and customer satisfaction, and
- Marketing including Customer Journey, Channel Performance and Customer Lifetime Value
we are able to analyze existing structures of organizations in a short time and to optimize causes and effects against corresponding objectives in dynamic systems.
On the basis of predefined ontologies (profiles or subsystems) we accelerate the exploration. Using methods developed by us, which are connected using machine learning, the creation of models, the completion and the simulation and prediction can be significantly accelerated and improved.
Management is thus once again a task that can be successfully completed by humans.
Please contact us if you find our approach to decision support in complex situations and for your company helpful and would like to learn more.
Sources
1 — Population status, Statistisches Bundesamt, https://www.destatis.de/DE/Themen/Gesellschaft-Umwelt/Bevoelkerung/Bevoelkerungsstand/_inhalt.html;jsessionid=67B196FADCE40AC8D7F5725F9473B622.internet8721, accessed on 04.25.2020
2 — Sorted infection data from John Hopkins via David Kriesel, http://www.dkriesel.com/_media/corona-cases.csv, accessed on 04.25.2020
3 — Sorted death data from John Hopkins on David Kriesel, http://www.dkriesel.com/_media/corona-deaths.csv, accessed on 04.25.2020
4 — SARS-CoV-2 Coronavirus disease profile-2019 (COVID-19) from the Robert Koch Institute, https://www.rki.de/DE/Content/InfAZ/N/Neuartiges_Coronavirus/Steckbrief.html#doc13776792bodyText19, accessed on 04.25.2020
5 — Coronavirus update by Christian Drosten — The podcast episodes as script, German, https://www.ndr.de/nachrichten/info/Coronavirus-Update-Die-Podcast-Folgen-als-Skript,podcastcoronavirus102.html, accessed on 04.25.2020
6 — CDC. Coronavirus Disease 2019 (COVID-19), Prevent Getting Sick Guidelines, https://www.cdc.gov/coronavirus/2019-ncov/prevent-getting-sick/diy-cloth-face-coverings.html
7 — Leung, N.H.L., Chu, D.K.W., Shiu, E.Y.C. et al. Respiratory virus shedding in exhaled breath and efficacy of face masks. Nat Med (2020). https://doi.org/10.1038/s41591-020-0843-2
8 — Sven Clausen and Katharina Slodczyk, conflict with CEO Sewing — Digital CDO leaves Deutsche Bank, https://www.manager-magazin.de/unternehmen/banken/deutsche-bank-digitalchef-markus-pertlwieser-geht-a-1306544.html, accessed on 04.25.2020
9 — The complete picture of the mask question looks like this. The entities, instances and states as well as their relationships are each parameterized and can be optimized in terms of the objective “Best possible protection of the population” with any input. Field 33, 04.25.2020
