NFPA Journal - Winter 2022 - 33

of gray, and depending on how broadly a country
defines a death caused by fire, the more deaths will
be counted in its national numbers.
" One country might only count a fire death if the
victim is found dead in the building. Another might
also count it if they died in the ambulance. Some countries
also count it if they die from complications a year
after the fire. Some count people who die in car fires, "
said Messerschmidt, who led NFPA's involvement
with the EU FireStat project. " A weird discussion we
had during the project was whether we should count
somebody who sets fire to themselves as a fire death
or as a suicide. This is why having specific definitions
is so important. If everyone is doing their own thing,
suddenly you have all kinds of different numbers that
aren't comparable. "
with estimates based on statistical extrapolation.
Countries that mistakenly assume that they are collecting
all fires and don't take that step will end up
with a total that is far fewer than the actual number.
When comparing apples to apples, it all matters.
" Multiply these challenges by 27 countries and
about 20 different languages and you get a sense of
the enormity of our task, " El Houssami said.
Instead of apples and oranges, the consortium
found something more like a cornucopia that was
overflowing with fruits of every shape and size.
At one end, countries like Sweden, Finland, and
Estonia have robust collection efforts and already
collaborate on shared data efforts through an online
platform. Other countries, such as Italy, focus
instead on collecting a small set of data variables,
but do it very well. France, meanwhile, does a good
job of collecting fire data, but has very few definitions,
so its overall accuracy is suspect.
" Then there are other countries that don't have any
As the world becomes more
connected, more urban, and
more globalized, there is
perhaps a greater benefit than
ever to sharing fire experiences
and lessons between
countries-and a greater cost
for not doing so.
Exploring all of the nuances took project participants
down many different rabbit holes. If a
building has three stories underground and seven
above, does it have 10 stories, or seven? If a person
looks injured at a fire scene but refuses care, is that
a fire injury? Even deciding what a " fire incident " is
can get tricky-is it any incident no matter how we
learn of it, or only those where the fire department
is called and responds?
Definitions were only the start, however. The
FireStat team also wanted to know how data is
collected, compiled, and reported, and how countries
account for missing data in their models. In
the United States, for instance, which only collects
data on an estimated 70 percent of known fires,
researchers at NFPA fill in the remaining details
national data collection, or where it's a completely
different situation from one region of the country to
another, " El Houssami said. This includes Germany,
where each region of the country utilizes a different
data collection strategy, and Spain, which has no
public system and instead relies almost exclusively
on private insurance to collect data. Portugal and
some Eastern EU countries have virtually no fire data
collection at all, El Houssami said.
As all this was happening, other members in the
research consortium examined data collection practices
in eight countries outside of the EU-the US,
Canada, Norway, Switzerland, Russia, Australia, New
Zealand, the UK-to see what lessons might emerge.
NFPA worked primarily on preparing information on
the US data collection system, known as the National
Fire Incident Reporting System, or NFIRS.
Armed with this trove of information, the FireStat
group set out to create a data collection system that
could serve as a baseline for all of the EU-and, perhaps
one day, for countries around the world. In its
final report, the group identifies 14 variables " necessary
to be collected as a priority in all EU member
states in a harmonized way " and includes precise
definitions for each. It also offers detailed guidance
to officials on each step of the fire data journey,
from the data collection process at the departmental
level all the way to describing how countries should
report their national data to the EU.
While acknowledging its ambitious scope, the
report makes clear that achieving its aims will be
a long process, and that few countries will be able
to meet all of its recommendations right out of the
gate. According to FireStat's research, only seven
of the fire data variables it identified as priorities
are already collected by more than half of the EU
countries. The other seven are currently collected
by only a handful of nations and will likely take
longer to implement across the continent. Likewise,
countries with limited fire data collection efforts
NFPA .ORG/JOURNAL * NFPA JOURNAL | 33
The South African
cities of Johannesburg
(far left) and
Cape Town record
dramatically different
numbers of
" fire incidents " each
year, likely a result of
differences in how
those incidents are
defined and how
associated details
are collected.
ALLAN BAXTER/GETTY IMAGES
https://www.nfpa.org/journal
NFPA Journal - Winter 2022

Table of Contents for the Digital Edition of NFPA Journal - Winter 2022
