HomeMy Public PortalAboutFire Risk_1 U.S. Fire Administration/National Fire Data Center
Fire Risk
Topical Fire Research Series, Volume 4 - Issue 7
December 2004
�pA�1,Y
FEMA
'HD 5 '
TOPICAL FIRE RESEARCH SERIES
Fire Risk I—A
December 2004 Volume 4, Issue 7
Findings
■ Risk by age:Children under the age of 5 and the population over the age of 54
are at the highest risk of death in fires.The risk of fire injury is greatest(over
1.0) in the 20-44 and the 85+ age ranges.
■ Risk bygender.-Men are 1.6 times more likely to die in a fire than women.
■ Risk byrace:African Americans and American Indians are at much greater risk of
death in a fire than the general population.
■ Risk by region:The risk of dying in a fire in the South is higher than other areas
of the United States.
■ Risk by economic factor.-Populations at the lowest income levels are at a greater
risk of dying in a fire than those with higher incomes.
The risk from fire is not the same for everyone. More than 4,000 deaths and 20,300 injuries in the United
States were caused by fire in 2001.1 These casualties were not equally distributed across the U.S. population, and
the resulting risk of death or injury from fire is not uniform—it is more severe for some groups than for others.
Much can be learned from understanding why different segments of society are at heightened risk from the fire
problem.
Risk is a factor, element, or course of action involving uncertainty. It is an exposure to some peril, and it often
implies a probability of occurrence, such as investment risk or insurance risk. In terms of the fire problem, risk is
the potential for injury or death of a person or damage or loss to property.
This topical report focuses on how fire risk—specifically the risk of death and injury—varies with age and how
other demographic and socio-economic factors weigh upon that risk.
PER CAPITA RATES, RISK,AND FIRE CASUALTIES
When determining fire risk, age, geographic, demographic, and socio-economic factors all come into play.
People in the Southeast, the elderly, the very young, and the poor all are at higher risk of being injured in a fire
than the rest of the population. Males, African Americans, and American Indians have a considerably higher risk of
death or injury from fire than does the population as a whole. These groups have remained at higher risk despite
considerable long-term reductions in fires and fire casualties.
Fire casualties across population groups can be assessed in several ways. The simplest method is to look at the
distribution of the numbers of deaths or injuries across the factor of interest. In the case of age, the number of fire
deaths is greatest for the very young while most fire injuries occur among middle-aged adults.2 Or, in the case of
race, the number of fire deaths is greatest for white Americans and least for Asian Americans.
DepartmentEmergency
Management -
United States Fire Administration* National Fire Data Center
Emmitsburg,
••
1
Although these findings are informative, they do not account for differences in the basic population groups
under comparison. In the case of age, as an age group matures, its population of individuals decreases as a result of
deaths and fatal injuries.3 In the case of race, there are far fewer Asian Americans than white Americans living in
the United States. As a consequence, it is possible for an age group to have greater (or fewer) injuries or deaths
because the sheer number of individuals for whom it is possible to be injured is larger (or smaller) than other
groups.
To account for population differences such as these, per capita rates are used. Per capita rates use a common
population size, which then permits comparisons between different groups.4 Perhaps the most useful way to assess
fire casualties across groups is to determine the relative risk of dying or being injured. Relative risk compares the
per capita rate for a particular group (e.g., females) to the overall per capita rate (i.e., the general population). The
result is a measure of how likely a group is to be affected.
For the general population, the relative risk is set at 1. The relative risk of dying in a fire for the total popula-
tion of females in comparison to the total population is 0.8, which is equivalent to the per capita fire death rate for
females (10.7) divided by the per capita fire death rate for the entire population (14.0). Thus the relative risk of a
female dying from fire is 20%less than that of the total population.
AGE AND RISK OF FIRE CASUALTY
When physical and cognitive abilities are limited, as is often the case for the very young and the very old, the
risk of death and injury from fire rises. The elderly experience large numbers of fire deaths that occur in a small
population group. As a result, the risk of dying in a fire for the elderly is 2.5 times higher than for the population
as a whole and rises even more for the oldest segment (Figure 1). Individuals age 65-74 are 1.8 times more likely
to die in a fire than the general population, while those adults age 85 or older are 4.6 times more likely to suffer
fire-related deaths. Increasing frailty and infirmity accompany aging, and the tendency of higher fire death and
injury risk to rise with greater age is not surprising.5 Approximately 2,200 older adults (65 and older) were
injured and 1,250 killed in fires in 2001.6
FIGURE 1. RELATIVE RISK OF FIRE DEATH BY AGE, 2001
0-4 1.2
5-9 0.6
10-14_ 0.
15-19 0. -Relative risk for general population
20-24 0.6
25-29 0.6
30-34 0.6
35-39 0.8
40-44 1.0
45-49 0.9
50-54 1.0
55-59 M1.3
60-64 1.2
65-69 1.7
70-74 2.0
75-79 2.4
80-84 3.3
85+ 4.6
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Relative Risk
Source:NCHS and U.S.Census Bureau
2
Children also experience large numbers of fire deaths. Though the risk of death for children from birth to age
14 is slightly below the risk to the population as a whole, some groups of children face greater danger. Children
under the age of 5 have a greater risk of death than all other children (Figure 1). As children mature and their cog-
nitive and social abilities develop, the risk of fire death drops sharply. Children between ages 5 and 9 have half the
risk of fire death of their youngest counterparts; children between 10 and 14 have half that of the 5 to 9 group. Of
the youngest, African American, and American Indian children are most threatened. Children in the age 4 and
under group also have a greater relative risk of injury from fire (Figure 2) than their older counterparts.7 Approxi-
mately 2,900 children (14 and younger) were injured and 600 killed in fires in 2001.
After age 19, the risk of fire death increases. By age 55, the risk of death is above average and it continues to
increase sharply as the population ages. Although the overall numbers change, these profiles have remained rela-
tively constant from year to year, according to National Center for Health Statistics (NCHS) and U.S. Census Bureau
data.
The age profile of risk for fire injuries is very different from that for deaths (Figure 2) with a much narrower
range of risk quotients (0.4 to 1.4 vice 0.3 to 4.6 for fire deaths). This difference is thought to be both the result of
cognitive and mobility issues that affect children and older adults. As a result, children and older adults are less
likely to escape the effects of fire and thus suffer fatal injuries. Middle-aged individuals tend to suffer nonfatal
injuries-in 2001, most fire injuries occurred among 30- to 44-year olds.$ In terms of risk, the risk of injury in a
fire is highest for young adults aged 20 to 24, mid-life adults aged 30 to 44, and the elderly over 75. The risk of
injury is below average for children under age 15 (Figure 2).
FIGURE 2. RELATIVE RISK OF INJURY BY AGE, 2001
0-4 MERL__ 1.0
5-9 0.4
10-14 0.6
15-19 1.0
20-24 1.4
25-29 1.2
30-34 1.4
35-39 -1.2
40-44 -1.2
45-49 1.0
50-54 1.0
55-59 0.8
60-64 0.8
< Relative risk for @eneral population
65-69 0.6
70-74 0.8
75-79 0.9
80-84 1.0
85+ - 1.3
0.0 0.5 1.0 1.5 2.0
Relative Risk
Source:NFIRS,NFPA,and U.S.Census Bureau
OTHER FACTORS THAT INFLUENCE RISK
According to USFA's report Socioeconomic Factors and the Incidence of Fire,9 socioeconomic studies repeatedly show
income levels are directly or indirectly tied to fire risk. There is an inverse relationship between fire risk and
income. The poorer population groups have the highest risk of fire injury or death, while the wealthiest have the
lowest. Many elderly people live alone on meager incomes, frequently in substandard housing stock.10 Closely tied
to income is level of education. The groups living in poverty-that is, with income levels below the poverty
line-are often undereducated and lack a basic high school education.
3
Geographic location also comes into effect (Table 1). There is a greater risk of dying in a fire in the South
(30%) than other regions. This is, in part, attributed to the intermittent need for occasional heating. Rather than
including central heating systems as in northern areas, many households in the South use portable heating devices
for heat. By their nature, such heating strategies are more likely to lead to a fire problem. Conversely, the West,
with the exception of Alaska, has a much lower risk of fire death. This reduction may also be due, in part, to the
role of heating (or lack of) in fire deaths.
TABLE 1. RELATIVE RISK BY GEOGRAPHIC AREA, 2001
Death Rate Relative
Region Population Fire Deaths (per million) Risk
Northeast 53,950,802 655 12.1 0.9
Midwest 64,819,817 991 15.3 1.1
South 101,953,947 1,820 17.9 1.3
West 64,592,993 541 8.4 0.6
Overall U.S.L -i I285,317,559 4,007 14.0 1.0
Source:NCHS 2001 mortality data and Population Division,U.S.Census Bureau,Table ST-EST2002-01-State
Population Estimates:April 1,2000 to July 1,2002,Release Date:December 20,2002.
Note:Relative risk may not compute due to rounding.
Like age, gender plays a role in the risk of death or injury from fire. For virtually all age groups, males are
more likely to die in a fire-related incident (Tables 2-4). Overall, in 2 0 01, men were 1.6 times more likely to die
in a fire than women. U.S. Fire Administration data from the National Fire Incident Reporting System (NFIRS)
show that men are also likely to suffer more injuries-from 1.5 to 2 times more.11 Why this is so is subject to
speculation. Men may be more willing to take risks than women, and this behavior could account for some of the
difference. Previous NFIRS data indicate that more men than women will try to extinguish a fire. This action alone
could account for much of the difference in injury rates.
Race, which may be related to societal factors, cannot be ignored. African Americans and American Indians
have noticeably higher death rates per capita than the national average. African Americans comprise a large and
disproportionate share of total fire deaths, accounting for 25% of fire deaths-60%more than their share of the
overall population.12 In 2001, African Americans had twice the relative risk of dying than the general population.
For American Indians that year, the relative risk was also elevated; 30% higher than the overall risk. By contrast,
Asian Americans are much less likely than the overall population to die in a fire.
TABLE 2. RELATIVE RISK OF FIRE DEATH BY RACE AND GENDER, 2001
OVERALL POPULATION
Death Rate Relative
Gender/Race Population Fire Deaths (per million) Risk
Total 285,317,559 4,007 14.0 1.0
Male 140,075,610 2,455 17.5 1.2
Female 145,241,949 1,552 10.7 0.8
White 230,664,347 2,908 12.6 0.9
African American 36,283,895 1,006 27.7 2.0
American Indian 2,713,047 49 18.1 1.3
Asian/Pacific 11,602,700 44 3.8 0.3
White Male 113,863,214 1,777 15.6 1.1
African American Male 17,256,399 616 35.7 2.5
American Indian Male 1,357,962 33 24.3 1.7
Asian/Pacific Male 5,609,267 29 5.2 0.4
White Female 116,801,133 1,131 9.7 0.7
African American Female 19,027,496 390 20.5 1.5
American Indian Female 1,355,085 16 11.8 0.8
Asian/Pacific Female 5,993,433 15 2.5 0.2
Source:NCHS 2001 mortality data;Population Division,U.S.Census Bureau,Table NA-EST2002-ASRO-03-
National Population Estimates-Characteristics,Release Date:June 18,2003,
http://www.census.gov/popest/archives/2000s/vintage_2002/NA-EST2002-ASRO-03.html:and Population
Division,U.S.Census Bureau,detail files for Monthly Population Estimates,2000 to 2002,
http://www.census.gov/popest/archives/2000s/vintage_2002/files/2002RESIDENT200l MONTHS07_12.txt
Note:Relative risk may not compute due to rounding.
4
TABLE 3. RELATIVE RISK OF FIRE DEATH BY AGE, RACE, AND GENDER, 2001
CHILDREN (AGE 0-14)
Death Rate Relative
Gender/Race Population Fire Deaths (per million) Risk
Total 60,482,119 599 9.9 0.7
Male 30,958,761 361 11.7 0.8
Female 29,523,358 238 8.1 0.6
White 46,280,132 380 8.2 0.6
African American 9,487,444 210 22.1 1.6
American Indian 740,542 8 10.8 0.8
Asian/Pacific 2,343,120 1 0.4 0.0
White Male 23,748,525 228 9.6 0.7
African American Male 4,814,914 126 26.2 1.9
American Indian Male 376,449 6 15.9 1.1
Asian/Pacific Male 1,192,485 1 0.8 0.1
White Female 22,531,607 152 6.7 0.5
African American Female 4,672,530 84 18.0 1.3
American Indian Female 364,093 2 5.5 0.4
Asian/Pacific FemaleRk 1,150,635 11 . - ' - ML -
Source:NCHS 2001 mortality data;Population Division,U.S.Census Bureau,Table NA-EST2002-ASRO-03-
National Population Estimates-Characteristics,Release Date:June 18,2003,
http://www.census.gov/popest/archives/2000s/vintage_2002/NA-EST2002-ASRO-03.html:and Population
Division,U.S.Census Bureau,detail files for Monthly Population Estimates,2000 to 2002,
hftp://www.census.gov/popest/archives/2000s/vintage_2002/files/2002RESIDENT2001 MONTHS07_12.txt
Note:Relative risk may not compute due to rounding.
TABLE 4. RELATIVE RISK OF FIRE DEATH BY AGE, RACE, AND GENDER, 2001
ELDERLY (AGE 65+)
Death Rate Relative
Gender/Race Population Fire Deaths (per million) Risk
Total 35,353,266 1,250 35.4 2.5
Male 14,619,070 638 43.6 3.1
Female 20,734,196 612 29.5 2.1
White 31,182,858 942 30.2 2.2
African American 2,915,629 283 97.1 6.9
American Indian 154,885 8 51.7 3.7
Asian/Pacific 904,344 17 18.8 1.3
White Male 12,972,922 473 36.5 2.6
African American Male 1,112,495 151 135.7 9.7
American Indian Male 66,748 5 74.9 5.3
Asian/Pacific Male 384,994 9 23.4 1.7
White Female 18,209,936 469 25.8 1.8
African American Female 1,803,134 132 73.2 5.2
American Indian Female 88,137 3 34.0 2.4
Asian/Pacific Female Ij 519,350 8 15.4 J&
1.1
Source:NCHS 2001 mortality data;Population Division,U.S.Census Bureau,Table NA-EST2002-ASRO-03-
National Population Estimates-Characteristics,Release Date:June 18,2003,
http://www.census.gov/popest/archives/2000s/vintage_2002/NA-EST2002-ASRO-03.html:and Population
Division,U.S.Census Bureau,detail files for Monthly Population Estimates,2000 to 2002,
http://www.census.gov/popest/archives/2000s/vintage_2002/files/2002RESIDENT2001 MONTHS07_12.txt
Note:Relative risk may not compute due to rounding.
5
CONCLUSION
The very young and the very old are some of the nation's most vulnerable residents. These high-risk groups
merit special attention to reduce their risk of injury or death from fire. With an aging population, the U.S. demo-
graphic profile is changing rapidly. The elderly population is expected to increase from its current 12.5% of the
total population to nearly 20%within a few decades,13 with an assumed corresponding increase in fire deaths and
injuries among older adults.
Because children and older adults account for 45% of fire deaths and one-quarter of fire injuries, the USFA has
been working toward the goal of reducing fire deaths and injuries to children and older adults. A number of
resources to help address the fire problem for children and adults are available. USFA's fire safety campaign for
babies and toddlers at http://www.usfaparents.gov provides parents with home strategies ranging from the control
of matches and lighters to home escape planning to protect young children from fire. For adults, A Fire Safety Cam-
paign for People 50-Plus (http://www.usfa.fema.gov/50plus) addresses lifestyle strategies of safe smoking, safe cook-
ing, and safe heating to reduce the incidence of fires that traditionally affect older adults. For further information,
see the USFA Web site (http://www.usfa.fema.gov) or contact your local fire department.
To request additional information or comment on this report, visit
http://www.usfa.fema.gov/applications/feed back
Notes:
1 Fire deaths are extracted from the 2001 NCHS mortality data using International Classification of Disease(ICD)codes F63.1,
W39-W40,X00-X09,X75-76,X96-97,Y25-26,and Y35.1 where these codes were noted as either the underlying cause of
death or a contributing factor in the chain of events leading to death.Only deaths where age and gender were specified were
used in this analysis.Estimates of fire injuries are from the National Fire Protection Association's (NFPA's)annual survey.
2 Fire in the United States 1992-2001,United States Fire Administration,Federal Emergency Management Agency, 13th Edition.
3 Immigration and changes in U.S.overseas populations have an additive effect on a static population group in the early and mid-
life years.Over the lifespan of the population group,however,death rates have a larger,decreasing effect.
4 Per capita rates are determined by the number of deaths or injuries occurring to a specific population group divided by the total
population for that group.This ratio is then multiplied by a common population size.For the purposes of this report,per capita
rates for fire deaths and injuries are measured per 1 million persons.For example,the per capita fire death rate for the total
female population is computed from the total number of female fire deaths(1,552)divided by the total female population
(145,241,949)multiplied by 1,000,000 persons.This rate is equivalent to 10.7 deaths per 1 million population.
5 See also"The Fire Risk to Older Adults,"Topical Report Vol.4,Issue 9,December 2004.
6 Numbers of fire deaths are extracted from NCHS mortality data using the ICD codes noted previously.Estimates of fire injuries
are calculated by determining the percent of injuries from the NFIRS data and applying the percentage to the NFPA estimate of
fire injuries.
7 See also"The Fire Risk to Children,"Topical Report Vol.4,Issue 8,December 2004.
8 Fire in the United States 1992-2001,loc.cit.
9 Socioeconomic Factors and the Incidence of Fire,U.S.Fire Administration,FA 170,June 1997.
10 Idem.
11 Fire in the United States 1992-2001,loc.cit.
12 Based on U.S.Census Bureau population estimates for July 2001.
13 U.S.Census Bureau population data as noted in"The Fire Risk to Older Adults,"loc.cit.
6