ESCP ALUMNI magazine - Special Issue 2022 - 22

Cover Story ❙ Interview
By Florian Lefèvre
" Your graphs help me
understand the situation "
Since the first lockdown, 22-year-old Elias ORPHELIN (MiM 21), a fifth-year student at ESCP
Business School, has been on a mission to inform people about coronavirus in France by
compiling pandemic-related data and making graphs on his Twitter account, but also by making
a risk calculator that estimates how likely you are to be in contact with an infected person.
He has even been hailed by the French health authorities (Santé Publique France).
Back then, each regional health agency
(ARS) published their own reports, and data
was not as structured as it is now. It was a bit
of a stat hunt because some regions would
sometimes fail to publish their own figures,
etc. I had all the ARS websites registered
as favourites on my computer! (He laughs)
Elias ORPHELIN (MIM 21)
Princing Manager Siemens
You use your Twitter account
almost daily to post graphs and
inform people on the evolution of
the coronavirus pandemic. Why do
you do this?
I have loved stats since primary school
because I think they tell a lot of things. Prior
to the pandemic, I used to make graphs and
stats about everything - football scores,
politics, economics, etc. I also did a lot of
" Business Intelligence " , i.e., putting data
into graphs to help the company make
decisions, when I worked for Siemens. As
it is something I am good at and I love, I
thought, in the early times of the pandemic:
" How can I be part of the team effort? " and
I thought graphs could be useful for our
society.
How did it start concretely?
I shared my very first graph mid-March
2020, in the early days of the first lockdown.
22 ❙
How long is it to search and compile
the data?
I would say about 2 or 3 hours a day.
Though the stats are more structured on
Santé Publique France, it still takes a lot of
time. We only had the infection and death
stats to begin with. Now, if you want to get
a thorough perspective of the pandemic,
you must consider the infections, the tests,
the admissions and discharges in hospitals
or intensive care, etc. The stats have always
depended on multiple factors, it is just that
we did not have access to all this data.
Among all these indicators, which
is the most relevant to predict the
evolution of the pandemic, in your
opinion?
I always have an eye on hospital admissions.
It is the most relevant stat regarding
the contamination period, and it is more
reliable than the number of infections,
which depends on the number of tests.
Early September 2020, the number of hospital
admissions increased at about 30%
or 40% per week, but the increase was still
very slow. Consequently, it did not seem
to increase much in absolute value, but it
ESCP Alumni Magazine Special issue ❙ 2022
actually showed that an exponential curve
was in the making. That was the problem:
we struggled to understand how quickly
that virus could spread. Within a few days,
there were twice as many people in hospital.
We would be overwhelmed and make
decisions in a rush... But that's when a new
indicator was found, one that would save us
precious days.
Which one?
The SARS-Cov2 concentration in wastewater.
Finding the virus in our wastewater even
before the first symptoms appeared helped
to know that the virus was spreading in a
town or a neighbourhood. It was a local
thing; for instance, firemen in Marseille
would conduct their own research. A
network of researchers worked on that, but
it took time. Late November 2020, I wrote to
the Minister of Research, Frédérique Vidal,
to ask if she could publish these figures
as open data. I felt that if researchers like
Guillaume Rozier or Germain Forestier
had access to this data, they could analyse
it and make tools to help predict the evolution
of the pandemic. But she said it was
complicated...
You also created a risk calculator
that shows the risks to be in contact
with an infected person, which has
been integrated to CovidTracker...
Engineer Guillaume Rozier created
CovidTracker. I got in touch with him
a few months ago and suggested we

ESCP ALUMNI magazine - Special Issue 2022

Table of Contents for the Digital Edition of ESCP ALUMNI magazine - Special Issue 2022