The Epidemiology of Extreme Weather Event Disasters (1969-2018)

Abstract

Introduction
This manuscript summarizes the global incidence, exposures, mortality, and morbidity associated with extreme weather event (EWE) disasters over the past 50 years (1969-2018).

Methods
A historical database (1969-2018) was created from the Emergency Events Database (EM-DAT) to include all disasters caused by seven EWE hazards (ie, cyclones, droughts, floods, heatwaves, landslides, cold weather, and storms). The annual incidence of EWE hazards and rates of exposure, morbidity, and mortality were calculated. Regression analysis and analysis of variance (ANOVA) calculations were performed to evaluate the association between the exposure rate and the hazard incidence rate, as well as the association between morbidity and mortality incidence rates and rates of human exposure and annual EWE incidence.

Results
From 1969-2018, 10,009 EWE disasters caused 2,037,415 deaths and 3,998,466 cases of disease. A reported 7,350,276,440 persons required immediate assistance. Floods and storms were the most common. Most (89%) of EWE-related disaster mortality was caused by storms, droughts, and floods. Nearly all (96%) of EWE-related disaster morbidity was caused by cold weather, floods, and storms. Regression analysis revealed strong evidence (R ² = 0.88) that the annual incidence of EWE disasters is increasing world-wide, and ANOVA calculations identified an association between human exposure rates and hazard incidence (P value = .01). No significant trends were noted for rates of exposure, morbidity, or mortality.

Conclusions
The annual incidence of EWEs appears to be increasing. The incidence of EWEs also appears to be associated with rates of human exposure. However, there is insufficient evidence of an associated increase in health risk or human exposures to EWEs over time.

Full text

For Peer Review
The epidemiology of extreme weather event disasters
(1969-2018)
Journal:
Prehospital and Disaster Medicine
Manuscript ID
PDM-19-0246.R2
Manuscript Type:
Original Research
Keywords:
Extreme weather events, Disasters, Exposures, Climate change, Hazards
Cambridge University Press
Prehospital and Disaster Medicine

For Peer Review
I. ABSTRACT
This manuscript summarizes the global incidence, exposures, mortality and morbidity associated
with extreme weather event (EWE) disasters over the past 50 years (1969-2018).
METHODS
A historical database (1969-2018) was created from the Emergency Events Database (EM-DAT)
to include all disasters caused by seven EWE hazards (i.e. cyclones, droughts floods, heatwaves,
landslides, cold weather and storms). The annual incidence of EWE hazards and rates of exposure,
morbidity and mortality were calculated. Regression analysis and ANOVA calculations were
performed to evaluate the association between the exposure rate and the hazard incidence rate as
well as the association between morbidity and mortality incidence rates and rates of human
exposure and annual EWE incidence.
RESULTS:
During 1969-2018, 10,009 EWE disasters caused 2,037,415 deaths and 3,998,466 cases of
disease. A reported 7,350,276,440 persons required immediate assistance. Floods and storms
were the most common. Most (89%) of EWE-related disaster mortality was caused by storms,
droughts and floods. Nearly all (96%) of EWE-related disaster morbidity was caused by cold
weather, floods and storms. Regression analysis revealed strong evidence (R2 = 0.88) that the
annual incidence of EWE disasters is increasing worldwide. Analysis of variance (ANOVA)
calculations identified an association between human exposure rates and hazard incidence (P-
value = 0.01) No significant trends were noted for rates of exposure, morbidity or mortality.
CONCLUSIONS
Page 1 of 13
Cambridge University Press
Prehospital and Disaster Medicine

For Peer Review
The annual incidence of EWEs appears to be increasing. The incidence of EWEs also appears to
be associated with rates of human exposure. However, there is insufficient evidence of an
associated increase in health risk or human exposures to EWEs over time.
II. INTRODUCTION
Extreme weather events are caused by oceanic and atmospheric hazards that are influenced by the
global climate. Warming of the global climate is predicted to increase the number of extreme
weather events (EWE), (i.e. disasters caused by climatological, hydrological and meteorological
hazards). (1, 2)
Most disease related to EWE disasters occurs as a result of traumatic and/or psychological injury
sustained through: 1) exposure to environmental hazards such as water, wind, fire, smoke, debris,
or heat; or 2) an absence of a life-sustaining requirement (e.g. air, food, and water). There are also
indirect health effects that occur secondary to displacement and loss of health-sustaining services
that (under certain circumstances) can result in outbreaks of infectious disease and/or
exacerbations of chronic disease. (3-5)
These events are predicted to cause catastrophic health consequences for millions of people
worldwide. The world’s poor are disproportionately affected by all disasters. Thus, the most
vulnerable and marginalized in all societies are expected to bear the highest health burden. (6) This
manuscript summarizes the 50-year review of global disaster data to characterize the impact of
seven extreme weather hazards on the health of disaster-affected populations.
II. METHODS
A historical database including seven extreme weather hazards (i.e. cyclones, droughts floods,
heatwaves, landslides, cold weather and storms) occurring globally from 1969 to 2008 was created
from EM-DAT: The Emergency Events Database, an internationally accepted source of aggregate
Page 2 of 13
Cambridge University Press
Prehospital and Disaster Medicine

For Peer Review
disaster data. (7) EM-DAT is a global database on natural and technological disasters, containing
essential core data on the occurrence and effects of more than 21,000 disasters in the world, from
1900 to present. It is maintained by the Center for Research on the Epidemiology of Disasters
(CRED) at the Catholic University of Louvain, School of Public Health, in Brussels, Belgium.
CRED was established in 1973 and has collaborative status with the United Nations Department
of Humanitarian Affairs, the European Union Humanitarian Office, the International Federation
of the Red Cross and Red Crescent, the US Office of Foreign Disaster Assistance as well as with
non-governmental agencies such as the International Committee of the Red Cross and Red
Croissant. (7)
The EM-DAT database is comprised of information from various sources, including United
Nations (UN) agencies, non-governmental organizations, insurance companies, research institutes
and press agencies. Priority is given to data from UN agencies, governments, and the International
Federation of Red Cross and Red Crescent Societies. This prioritization is not only a reflection of
the quality or value of the data, it also reflects the fact that most reporting sources do not cover all
disasters or have political limitations that could affect the figures. The entries are constantly
reviewed for inconsistencies, redundancy, and incompleteness. CRED consolidates and updates
data daily. A further check is made at monthly intervals, and revisions are made at the end of each
calendar year. (7)
For a disaster to be entered into the EM-DAT database at least one of the following criteria must
be fulfilled: 1) Ten (10) or more people reported killed; 2) One hundred (100) or more people
reported affected; 3) Declaration of a state of emergency; or 4) Call for international assistance.
Deaths are defined as the “number of people who lost their life because the event happened”. Total
deaths are the “sum of deaths and missing”. The EM-DAT definition of “injury” is actually
Page 3 of 13
Cambridge University Press
Prehospital and Disaster Medicine

For Peer Review
comprehensive of all categories of disease, not merely injury - “persons suffering from physical
injuries, trauma or an illness requiring immediate medical assistance as a direct result of a
disaster”. EM-DAT defines “affected” as those persons “requiring immediate assistance during a
period of emergency, i.e. requiring basic survival needs such as food, water, shelter, sanitation and
immediate medical assistance”. (7)
The annual incidence of disasters related to EWE disasters was calculated over a 50-year period
(1969-2018) for all seven hazards. A regression analysis was then performed to characterize any
trends involving the annual global rates of EWE-related hazard incidence, exposure, morbidity
incidence, and mortality incidence.
Linear regression and analysis of variance (ANOVA) calculations were performed to evaluate the
association between the dependent variable of exposure rate (i.e. global number of persons affected
per 100,000 population) and the independent variable of hazard incidence rate (global number of
EWEs per year).
Separate multiple regression and ANOVA calculations were also performed to evaluate the
association between individual dependent variables of morbidity and mortality incidence rates and
the independent variables of both exposure rate and EWE hazard incidence rate.
III. RESULTS
According to this review, during the 50-year-long period, (1969-2018) 10,009 EWE disasters
caused over 2 million (2,037,415) deaths and nearly 4 million (3,998,466) cases of disease. Over
7 billion (7,350,276,440) persons required immediate assistance during a period of emergency.
Global incidence of EWE disasters
Floods (47%) and storms (30%) were the most common EWE disasters to occur during the 50-
year period studied, followed by landslides (7%), drought (6%), cold (4%), wildfire (4%) and heat
Page 4 of 13
Cambridge University Press
Prehospital and Disaster Medicine

For Peer Review
wave (2%). Figure one depicts a review of the 50-year trend for the relative contribution from
individual hazards to the total annual incidence of EWE disasters over time. This graph highlights
the scale and consistency of flood and storm incidence over time as the most common EWE
disasters, worldwide.
Calculation of linear regression for EWE disaster incidence revealed an upward trend over time
(with a model equation of y = 6.9931x – 13741 and R2 = 0.88). (See figure two)
Regression analysis of the other factors studied revealed the following values: exposure rate (R2 =
0.07); incidence of mortality (R2 = 0.07) and incidence of morbidity (R2 = 0.02). In 1969, there
were 40 EWE disasters on record. In 2018, there were 286. The mean annual incidence for EWE
disasters during this 50-year period is 200.
Linear regression and ANOVA calculations performed to evaluate the association between the
dependent variable of hazard incidence rate and the independent variable of exposure rate resulted
in an R2 = 0.12; F-significance = 0.01; and P-value = 0.01.
Global mortality from EWE disasters
Most (89%) global mortality during this period was caused by storms (39%), droughts (34%), and
floods (16%); followed by heat wave (8%), landslides (2%), cold (1%), and wildfire (<1%). The
global crude death rate for EWE-related mortality during this period was 0.08 deaths per 100,000
persons. Figure 3 represents the annual global incidence rate of mortality (deaths per 100,000
persons) attributable to EWE disasters over a 50-year period. There are 9 notable peaks that rise
above the mean. Seven of the 9 spikes were mainly attributable to only one hazard per year: storms
(3); drought (3); and heat (1). During the other two years (1999 and 2010), increases were caused
mainly by two hazards per year (floods and storms; and heat and drought), respectively.
Page 5 of 13
Cambridge University Press
Prehospital and Disaster Medicine

For Peer Review
Results of multiple regression and ANOVA calculations performed to evaluate the association
between the dependent variable of mortality incidence rate were as follows for the independent
variables of: exposure rate (R2 = 0.04; F significance = 0.35 and P-value 0.95); as well as for
hazard incidence ((R2 = 0.04; F significance = 0.35 and P-value 0.18).
Global morbidity from EWE disasters
During the 50 years studied, EWE disasters caused nearly 4 million (3,998,466) cases of disease.
Nearly half of global morbidity from EWE disasters was caused by cold (47%); with 31% caused
by floods; and 18% caused by storms; (followed by heat (4%), drought <1%), and wildfire (<1%).
The mean annual incidence rate of EWE-related morbidity (injuries and illness) during this period
was 0.13 cases per 100,000 persons. Figure 4 represents the annual global incidence rate of EWE-
related morbidity over a 50-year period.
There were 8 years when annual morbidity occurred above the mean. Seven of the 8 peaks can be
directly attributed to increases caused mainly by only one hazard floods (4); storms (2); and cold
(1). During the remaining peak year (2013), 55% of the annual morbidity was due to heat and 37%
was due to storms. It should also be noted that according to this data, during 2004, 98% of the
1,800,000 cases of EWE-related morbidity were reportedly due to the cold. This value represents
a 22-fold increase over the annual average for all cold weather disasters during the same timeframe.
Results of multiple regression and ANOVA calculations performed to evaluate the association
between the dependent variable of morbidity incidence rate were as follows for the independent
variables of: exposure rate (R2 = 0.01; F significance = 0.69 and P-value 0.98); as well as for
hazard incidence ((R2 = 0.01; F significance = 0.69 and P-value 0.42).
IV. DISCUSSION
Page 6 of 13
Cambridge University Press
Prehospital and Disaster Medicine

For Peer Review
During the past 50 years (1969-2018), 22,173 disasters (i.e. caused by biological, natural,
technological, extraterrestrial and conflict hazards) were responsible for an estimated 6.2 million
deaths and $3.4 trillion in damages worldwide. (7) Half (50%) of these disaster events were
climate-related and 95% of all people affected by disasters during this time were affected by
climate-related disasters. (7)
Overall, the results of this study characterize a significant source of morbidity and mortality
occurring on a global scale. To place this in context, the number of people affected worldwide by
EWE disasters during the past 50 years (~7 billion) is roughly equivalent to the entire world’s
population of current day.
Disaster-related health risk is caused the joint probability of occurrence for both hazards and
human exposures. (8) In this study, there is an observed trend of increasing EWE hazard incidence,
in the setting of no significant change in the measures (e.g. number of persons affected) or potential
proxies of exposure (e.g. morbidity and mortality).
These results suggest with a high degree of certainty that the annual incidence of EWE disasters
is increasing. There is also a moderate degree of evidence of an association between the incidence
of EWE’s and human exposure rates (i.e. number of persons affected per 100.000 population).
However, there were no significant changes noted for rates of EWE-related exposures, morbidity,
or mortality during the past 50 years.
V. LIMITATIONS OF THIS STUDY
In general, the effects of disaster events are the subject of gross approximations and aggregations
that have a great deal of imprecision. The availability and quality of data has likely increased and
improved over time and the use multiple data sources increased reporting. However, in many
events deaths are unknown or unrecorded; for other outcomes such as injured and affected,
Page 7 of 13
Cambridge University Press
Prehospital and Disaster Medicine

For Peer Review
reporting frequency is even lower which likely contributes to a substantial underestimation of the
impacts of EWE disasters on human populations. While the available data is sufficient for a cursory
analysis of global EWE disaster impacts and trends, improved reporting of EWE outcomes, would
improve the accuracy of the data and the validity of such conclusions.
VI. CONCLUSION
During the past 50 years, floods and tropical cyclones were the most common EWE disasters,
comprising 47% and 30% of all EWE disasters, respectively. Most EWE-related disaster
mortality was caused by hazards related to tropical cyclones (39%), droughts (34%), and floods
(16%). Nearly all EWE-related disaster morbidity was caused by hazards associated with cold
weather (47%), floods (31%) and tropical cyclones (18%).
The results of this study strongly suggest that the incidence of EWEs is increasing, and that the
incidence of EWEs is also associated with the rate of human exposure. However, there is
insufficient evidence of an associated increase in health risk or human exposures to EWEs over
time. In effect, during the past 50 years while the incidence of health hazards has been increasing,
the resultant health risks (in terms of morbidity and mortality) have remained relatively the same.
Further study is necessary to ascertain if this lack of association between hazard incidence and
health risk may be an indicator of human capacity for resilience related to the absorption,
adaptation and transformation of climate-related risk. (8)
VII. FIGURES:
Figure 1: Relative contribution of individual hazards to the total annual incidence of EWE-
related disasters (1969-2018)
Figure 2: Global incidence of EWE disasters (1969-2018)
Figure 3: Annual incidence of global mortality due to EWE disasters (1969-2018)
Figure 4: Annual incidence of global morbidity due to EWE disasters (1969-2018)
Page 8 of 13
Cambridge University Press
Prehospital and Disaster Medicine

For Peer Review
VIII. REFERENCES
1. Intergovernmental Panel of Climate Change (IPCC). Climate Change 2007: Impacts,
Adaptation, and Vulnerability. Cambridge, UK; 2007.
2. (IPCC) IPoCC. Global Warming of 1.5°C.An IPCC Special Report on the impacts of global
warming of 1.5°C above pre-industrial levels and related global greenhouse gas emission
pathways, in the context of strengthening the global response to the threat of climate
change, sustainable development, and efforts to eradicate poverty 2019.
3. Keim M. Extreme weather events. In: G L, editor. Climate and Health. First ed.
Philadelphia, PA: Lippincott, Williams & Wilkins 2016. p. 35-76.
4. Keim M, Abrahams J, Castilla-Echenique J. How do people die in disasters and what can
be done? USA: DisasterDoc LLC; 2016 [Available from: http://disasterdoc.org/how-do-
people-die-in-disasters/.
5. Keim M. Environmental Disasters In: Frumkin H, editor. Environmental health from global
to local. Third ed. San Francisco California: John Wiley and Sons; 2016. p. 667-92.
6. Brouewer R AS, Brander L, et al. Socioeconomic vulnerability and adaptation to
environmental risk: a case study of climate change and flooding in Bangladesh. Risk
analysis : an official publication of the Society for Risk Analysis. 2007;27(2):313-26.
7. Center for Research on the Epidemiology of Disasters (CRED). EM-DAT: The
International Disaster Database Brussels, Belgium: Ecole se Sante Publique, Universite
Catholique de Louvain; 2019 [Available from: www.emdat.be/.
8. Keim M. Defining Disaster-Related Health Risk: A Primer for Prevention. Prehosp
Disaster Med. 2018;33(3):308-16.
Page 9 of 13
Cambridge University Press
Prehospital and Disaster Medicine

For Peer Review
Figure 1: Relative contribution of individual hazards to the total annual incidence of EWE related disasters
(1969-2018)
373x278mm (72 x 72 DPI)
Page 10 of 13
Cambridge University Press
Prehospital and Disaster Medicine

For Peer Review
Figure 2: Global incidence of EWE disasters (1969-2018)
220x165mm (120 x 120 DPI)
Page 11 of 13
Cambridge University Press
Prehospital and Disaster Medicine

For Peer Review
Figure 3: Annual global incidence rate of mortality due to EWE disasters (1969-2018)
228x167mm (120 x 120 DPI)
Page 12 of 13
Cambridge University Press
Prehospital and Disaster Medicine

For Peer Review
Figure 4: Annual global incidence rate of morbidity due to EWE disasters (1969-2018)
225x164mm (120 x 120 DPI)
Page 13 of 13
Cambridge University Press
Prehospital and Disaster Medicine