Special Article
Risk Factors for Fatal Residential Fires
N Engl J Med 1992; 327:859-863September 17, 1992
- Abstract
- Abstract
Background.
Residential fires are the most important cause of fire-related mortality in the United States. Previous research has concentrated on fatal fires in urban areas; considerably less is known about fatal fires in rural areas.
Methods.
We studied fatal and nonfatal residential fires in predominantly rural areas. Using a case–control design, we compared all 151 fatal fires (cases) in single-family dwellings in North Carolina during a 13-month period with a sample of nonfatal fires (controls). Case fires were identified through the medical-examiner system, and control fires that occurred within a few weeks of the case fires were chosen from the records of randomly selected fire departments statewide. For each fire, fire officials were interviewed about the dwelling, the fire, the people involved, and the fire-response system.
Results.
Although heating incidents were the leading cause of fires, fatal fires were more likely to have been caused by smoking (31 percent of fatal fires vs. 6 percent of nonfatal fires). Mobile homes posed a higher risk of death if a fire occurred (odds ratio, 1.7; 95 percent confidence interval, 1.1 to 2.6), as did the absence of a smoke detector (odds ratio, 3.4; 95 percent confidence interval, 2.1 to 5.6). Smoke detectors were more protective against death in fires involving young children and when no one present was impaired by alcohol or drugs or had a physical or mental disability. The presence of an alcohol-impaired person was the strongest independent risk factor for death in the case of a fire (odds ratio, 7.5; 95 percent confidence interval, 4.4 to 12.7).
Conclusions.
Residential fires are most likely to be caused by heating equipment or smoking materials. The risk of death is greatest in fires in mobile homes, in those involving alcohol-impaired persons, and in those in houses without smoke detectors. (N Engl J Med 1992;327: 859–63.)
Media in This Article
Table 1
Characteristics of Fatal and Nonfatal Fires in North Carolina, January 1, 1988, through January 31, 1989.*Table 2
Likelihood of Death in a Fire When No Smoke Detector Was Present or the Residence Was a Mobile Home, Stratified by Selected Variables.*Article Activity
- Article
IN 1988, 552,500 residential fires claimed 5065 lives and injured an additional 22,600 people in the United States.1 Although residential fires make up 23 percent of all fires, they account for 73 percent of fire-related injuries and 80 percent of fire-related deaths.1 Deaths in residential fires are unevenly distributed among the population: the rate of death due to fire is twice as high in the South as in the West, and rates are higher among persons over 64 years of age, children under 5 years of age, males, members of minorities, the poor, and those who live in remote areas.2 3 4 The rate of fire-related deaths is higher among people who live in mobile homes than among those who live in other types of dwellings.5 , 6
Smoke detectors have been suggested as a means of reducing deaths in fires.7 In residential fires, the risk of death in houses with smoke detectors is estimated to be half that in houses without smoke detectors.8 The protective effects of smoke detectors under different conditions have not been well studied, however.
To determine the relations among smoke detectors, mobile homes, other risk factors, and mortality in residential fires in a predominantly rural southern state, we conducted a case–control study of fatal and nonfatal residential fires in North Carolina, a state in which the rate of fire-related deaths is 1.5 times the national average. We hypothesized that not having a smoke detector and residing in a mobile home would be independent risk factors for mortality in residential fires.
Methods
North Carolina law requires that all deaths in fires be investigated by the county medical examiner, who must file a report of the investigation with the Office of the Chief Medical Examiner. Through this office we identified all fatal fires that occurred in single-family dwellings in North Carolina from January 1, 1988, through January 31, 1989. The criteria for inclusion in this study were as follows: someone was in the house at the time the fire started; the fire was not due to arson; and the fire did not occur in a chimney. We excluded chimney fires after the study was initiated when we discovered that some fire officials were systematically excluding chimney fires when we inquired about the most recent residential fires in their areas. If at least one person (excluding firefighters) had injuries that were fatal within 30 days, a fire was defined as fatal (case); if no deaths occurred within 30 days, a fire was classified as nonfatal (control). A forensic pathologist and the North Carolina chief medical examiner reviewed data on each death and excluded any considered not to be due to the fire. After obtaining the identity of the fire official in charge of the response to each fire from the Office of the Chief Medical Examiner, we interviewed that official by telephone as soon as possible after the fire. The rank of the fire officials varied, because we interviewed only those who were present at the fire.
The fire officials were told that we were conducting a statewide study of fires and were asked about the circumstances of the fire (for example, the cause, extent of damage, and origin), the characteristics of the house (type of structure, age, and value of the house; whether it was rented or owned; the distance to the nearest neighbor; the number of exits; and whether smoke detectors and telephones were present), and the fire-response system (the time of the alarm and the response, whether firefighters were volunteers or paid, and whether a 911 system was available). The structured interview also included questions about everyone present in the house at the time of the fire: their ages, physical disabilities (in vision, hearing, or mobility), cognitive impairments (for example, dementia, retardation, or mental illness), location in the dwelling, and activity at the time of the fire, and whether their ability to function was impaired by alcohol or other drugs (as judged by the fire official). The officials were permitted to consult standard trip reports in answering our questions.
We identified nonfatal fires by randomly selecting 400 fire departments from among the 1255 in North Carolina. Each month we randomly reordered the list of these 400 departments. After completing each interview after a fatal fire, we collected data about two nonfatal fires with use of the same structured questionnaire. We contacted the fire departments on the monthly list in sequence and asked the fire official to identify the most recent nonfatal residential fire within the previous four weeks that met the other study criteria. This procedure was followed to reduce the potential for differences in recall between fatal and nonfatal fires. We did not explicitly attempt to match fatal and nonfatal fires. However, we contacted fire officials from the list as quickly as possible after the fatal fire in order to minimize seasonal variations between fatal and nonfatal fires and continued until we identified two qualifying nonfatal fires for each fatal fire.
Each interview was conducted by one of eight trained interviewers. At weekly meetings, each interview from that week was discussed, and coding decisions were logged to ensure standardization among interviewers and over time. We analyzed the results with use of the SAS statistical software package9 and used the test-based series method to calculate confidence intervals for odds ratios.10 We used techniques of stratified analysis to identify potential confounders and effect modifiers (interacting variables) in the results. Multivariate unconditional logistic-regression models were constructed to assess the relations between the absence of smoke detectors and residence in a mobile home to the fire's being fatal, adjusting for potential confounding and effect-modifier variables identified in the stratified analyses. The final models were constructed by a stepwise backward variable-selection procedure. Decisions to retain or drop variables were based on the magnitude of the change in the beta coefficient for the absence of smoke detectors, rather than on the P value. Regression diagnostic techniques11 were used to establish the lack of interdependence among the independent variables in the models.
Results
There were 151 fatal (case) fires and 283 nonfatal (control) fires during the study period. A total of 322 persons (excluding firefighters) were present during the fatal fires; 186 of them (58 percent) died, and 48 (15 percent) had nonfatal injuries. The mean age of the persons who died was 40 years (range, 2 months to 95 years); 62 percent were male. Seventy-nine percent died of inhalation injuries. Ninety-five percent of all deaths occurred within 24 hours after the fire. At least 652 persons were present during the nonfatal fires, of whom only 33 (5 percent) were reported by fire officials to have been injured; 20 (3 percent) were reported to have been hospitalized. For the fatal fires, 49 percent of the interviews were conducted within the first four weeks after the fire and 77 percent within the first eight weeks. By design, all interviews about the control fires were conducted within four weeks after the fire. The ratio of nonfatal to fatal fires in the study was less than 2:1 because we decided to exclude chimney fires after the study began. Of the state's 100 counties, 82 were represented in the study; 51 counties had at least one fatal and one nonfatal fire, and 7 counties had a fatal fire but no nonfatal fire.
The largest number of fires were attributed to heating equipment, which accounted for 39 percent of fatal fires and 28 percent of nonfatal fires (Table 1Table 1
Characteristics of Fatal and Nonfatal Fires in North Carolina, January 1, 1988, through January 31, 1989.*). Fifty-eight percent of the fatal heating-related fires were started by space heaters, of which 87 percent were kerosene heaters. In contrast, 45 percent of the nonfatal heating-related fires were attributed to wood stoves or fireplaces, whereas 30 percent were started by space heaters (55 percent of which burned kerosene). Smoking caused 31 percent of the fatal fires and only 6 percent of the nonfatal fires, and cooking was responsible for 10 percent of the fatal fires as compared with 23 percent of the nonfatal fires. The odds ratio for fatal fires caused by smoking as compared with all other causes was 7.7 (95 percent confidence interval, 4.1 to 14.6). Seventy percent of fatal fires occurred in the living room or bedroom of the dwelling; 35 percent of nonfatal fires began in these rooms.The variables associated with the fire-response system appeared unrelated to the risk of fatality in a fire; however, the characteristics of the home environment, the fire, and the persons present were related to the risk that the fire would be fatal (Table 1). The risk of fatality was higher for fires that occurred in older houses, rental units, mobile homes, and houses without telephones. Fatal fires were more likely to occur on weekends and during sleeping hours. Higher risks were associated with fires when persons in the oldest age group, persons impaired by the use of alcohol or other drugs (as judged by the fire official), or persons with physical or mental disabilities were present. The odds ratio for the presence of an alcohol-impaired person at the fire varied according to the main cause of the fire, from 2.3 (95 percent confidence interval, 1.2 to 4.4) for fires caused by heating equipment to 4.4 (95 percent confidence interval, 1.8 to 10.5) for fires attributed to smoking.
The absence of a smoke detector was a risk factor for fatality, with an odds ratio of 3.4 (95 percent confidence interval, 2.1 to 5.6) (Table 1). Seventy-seven percent of the fatal fires and half the nonfatal fires occurred in houses with no smoke detectors.
The results of stratified analyses showed that the odds that a fire in a house without a smoke detector would be fatal varied under different conditions (Table 2Table 2
Likelihood of Death in a Fire When No Smoke Detector Was Present or the Residence Was a Mobile Home, Stratified by Selected Variables.*). The absence of a smoke detector was relatively more lethal in the case of fires in which no one was judged by a fire official to be impaired by the use of alcohol or drugs, when children were present, and when no one with a disability was present. Smoke detectors appeared to be less protective in mobile homes than in other types of dwellings.These findings persisted in the multivariate logistic-regression models. Although we also controlled initially for house rental and mobile homes, in the final model we adjusted for the age of the house, the lack of a telephone, the presence of a person older than 64 years of age, and the presence of an alcohol-impaired or disabled person, as well as the one-way interactions of these variables with the absence of a smoke detector. The effect of the absence of a smoke detector varied according to whether an alcohol-impaired or disabled person was present.
A smoke detector was more than five times as likely to be absent in fatal fires when there was no one present who was impaired by the use of alcohol or drugs (odds ratio, 4.5 vs. 0.8) as in similar nonfatal fires, and seven times as likely when there was no disabled person in the home (odds ratio, 4.4 vs. 0.6) (Table 2). When a person with either a cognitive impairment or a physical disability was present, the lack of a smoke detector had no statistically significant relation to whether the fire led to a death.
Although mobile homes account for 11 percent of the housing units in North Carolina,12 31 percent of the fatal fires and 21 percent of the nonfatal fires we studied occurred in mobile homes (odds ratio, 1.7; 95 percent confidence interval, 1.1 to 2.6) (Table 1). The variations in the odds ratios for death in a residential fire in a mobile home are shown in Table 2 under different conditions. Mobile homes were more hazardous for children under five years of age and for persons with disabilities.
We used a similar modeling approach to study the relation of mobile homes to fire fatalities. The potential confounders we examined included the presence of a person more than 64 years of age and the presence of a person with a disability; the effect modifiers were the age of the house, the lack of a telephone, the presence of two or fewer exits, and the presence of a child less than five years of age. Also included in the models were the interactions of the effect modifiers with the variable for a mobile home. The final model adjusted for the age of the house and included the interaction of a mobile home with two or fewer exits. The adjusted effect of mobile homes on the risk of fatality depended on the number of exits the dwelling had; fires in mobile homes with two or fewer exits were 2.6 times more likely to be associated with fatal than nonfatal fires (95 percent confidence interval, 1.3 to 4.9). The age of the house (>19 years) was the only other significant variable (odds ratio, 2.7; 95 percent confidence interval, 1.6 to 4.6).
Discussion
In this statewide study, we assessed the risk factors associated with fire-related mortality by comparing fatal fires with nonfatal fires in predominantly rural areas. We were particularly interested in determining the value of smoke detectors and assessing the role of mobile homes in fire deaths. Two case–control studies of residential fires have been published previously. One study was conducted in Montreal, an urban setting, more than 12 years ago and did not contain data about smoke detectors or deaths from fires.13 The other compared alcohol use and smoking patterns in households where fires had occurred with patterns in the general population.14
Our study has several potential methodologic limitations. Fire officials may remember more details about fatal fires than about nonfatal fires, and fatal fires may be more thoroughly investigated. Information about the characteristics of persons present at fires is particularly difficult to evaluate. For instance, fire officials have an opportunity to smell alcohol on the breath of survivors or to observe behavior suggesting inebriation, whereas persons who die usually undergo toxicologic testing. Only in persons who died were blood alcohol levels measured as part of the routine investigation by the medical examiner. However, the degree of impairment by alcohol of each person involved in a fire was estimated by a fire official, and the results of this assessment were used in our analyses. Among the 149 persons who died whose blood alcohol concentrations were measured, we compared the fire officials' assessments of alcohol impairment with the blood alcohol concentrations in those with values ≥50 mg per deciliter (11 mmol per liter). The fire officials' assessments were highly sensitive (94 percent) and specific (85 percent).
Whether a smoke detector was present was unknown for 21 percent of the fatal fires and 19 percent of the nonfatal fires. However, no other factors varied significantly in relation to whether it was known if a smoke detector was present. When we assumed the two most extreme situations for fires about which data on smoke detectors were unavailable (all case fires and no control fires had smoke detectors, or no case fires and all control fires had smoke detectors), the estimates of the risk of death in a fire in houses without a smoke detector were 1.1 and 6.7, respectively. However, the proportion of fires for which we had no information about the presence or absence of a smoke detector increased among both fatal and nonfatal fires as the level of destruction caused by the fire increased.
In some instances in which smoke detectors were present, they may not have been operational. One researcher has estimated that one fourth to one third of the residential smoke detectors in the United States are not in operation.8 Moreover, in some fires in houses classified as having a smoke detector, adequate coverage might have required two or more detectors. Consequently, our estimates of the risk associated with the absence of a smoke detector may be low.
For both fatal and nonfatal fires, the prevalence of smoke detectors differed depending on whether the house was a mobile home, a rental unit, or a dwelling constructed before 1976, when North Carolina enacted a law requiring that smoke detectors be installed in all new houses. Overall, smoke detectors were present in more mobile homes than other types of dwellings (56 percent vs. 37 percent), but in a lower proportion of rental units than owner-occupied units (18 percent vs. 54 percent).
Similarly, a higher proportion of houses built after 1976 than houses constructed before the law was passed had smoke detectors (74 percent vs. 29 percent). The increased presence of smoke detectors in houses built after the 1976 North Carolina building code was implemented is consistent with other reports of the effectiveness of such policies.15 Because smoke detectors are much less likely to be found in rental units, however, regardless of the time of construction, special regulatory measures are needed to require the installation and maintenance of smoke detectors in rental property.
Although numerous factors are associated with the likelihood that a fire will prove fatal, our results demonstrate that smoke detectors are beneficial in almost every instance. They are especially beneficial when no one in the house has a disability, is impaired by alcohol, or is more than 64 years old. In situations in which smoke detectors have less benefit, sprinklers may be a valuable alternative. However, the use of sprinklers in single-family dwellings, especially in rural settings, may not be realistic because of their cost and the limits of water pressure.
Since mobile homes usually were more apt to have smoke detectors than other types of dwellings, reducing the frequency of deaths from fires in these structures may require other strategies. The relation between the presence of two or fewer exits and the excess risk of fatality in a fire suggests one mechanism that may be particularly useful in reducing fatalities in fires in mobile homes — increasing the number of exits in such units.
As part of primary prevention in clinical settings, people should be counseled about the value of smoke detectors and the need to maintain them. Special encouragement should be given to people who live in rental property to make sure that smoke detectors are present and functional. Even more important would be the involvement of health professionals in advocacy to produce fire-safe cigarettes, ensure that building and housing codes provide uniform and adequate fire protection, and support proper enforcement of these codes.
Supported by a grant (R49/CCR402444) from the Centers for Disease Control to the University of North Carolina Injury Prevention Research Center.
We are indebted to Dr. Jama Gulaid, Dr. Dana Loomis, Dr. Desmond Runyan, Dr. Richard Sattin, Dr. David Strogatz, and Dr. Richard Waxweiller for technical advice; to the interviewers, Ms. Nancy Foreman, Ms. Diana Gray, Ms. Edith Parker, Ms. Elizabeth Sand, Ms. Laureen Tyler, and Ms. Shannon Weigand, for assistance in collecting the data; to Ms. Donna Armstrong, Ms. Hsing-Yi Chang, Ms. Carolyn Crump, Mr. Eric Rodgman, and Ms. Jane Stutts for assistance with the statistical analysis; and to Ms. Janet Heath, Mr. Kevin Moran, and Ms. Deborah Schoenfeld for editorial assistance.
Source Information
From the Injury Prevention Research Center and the School of Public Health, University of North Carolina, Chapel Hill (C.W.R., S.I.B., M.A.L.); the Centers for Disease Control, Public Health Service, Atlanta (J.J.S.); and the North Carolina Office of the Chief Medical Examiner, Chapel Hill (J.B.). Address reprint requests to Dr. Runyan at the Injury Prevention Research Center, University of North Carolina, CB 7400 Rosenau Hall, Chapel Hill, NC 27599.
- References
References
1
Karter 2
Regional distribution of deaths from residential fires — United States, 1978–1984 . MMWR Morb Mortal Wkly Rep 1987;36:645–9.
Medline3
Mierley
CrossRef | Web of Science | Medline4
Baker SP, O'Neill B, Ginsburg J, Li G. The injury fact book. 2nd ed. New York: Oxford University Press, 1992.
5
Fire Analysis Division. Manufactured home fires, 1984–85: final report. Quincy, Mass.: National Fire Protection Association, 1988.
6
Patetta
CrossRef | Web of Science | Medline7
Council on Scientific Affairs
CrossRef | Web of Science8
Hall JR Jr. U.S. experience with smoke detectors: who has them? How well do they work? When don't they work? Quincy, Mass.: National Fire Protection Association, 1989.
9
SAS user's guide: statistics, version 5 ed. Cary, N.C.: SAS Institute, 1985.
10
Kleinbaum DG, Kupper LL, Morgenstern H. Epidemiologic research: principles and quantitative methods. Belmont, Calif.: Lifetime Learning, 1982.
11
Davis CE, Hyde JE, Bangdiwala SI, Nelson JJ. An example of dependencies among variables in a conditional logistic regression. In: Moolgavkar SH, Prentice RL, eds. Modem statistical methods in chronic disease epidemiology. New York: John Wiley, 1986:140–7.
12
Census of population and housing, 1980: public-use microdata sample (a sample). North Carolina. Washington, D.C.: Bureau of the Census, 1983. (Electronic data file.)
13
Ducic
CrossRef | Web of Science14
Ballard
Web of Science | Medline15
McLoughlin
CrossRef | Web of Science | Medline
- Citing Articles (55)
Citing Articles
1
Michael D. Peck. (2011) Epidemiology of burns throughout the world. Part I: Distribution and risk factors. Burns 37:7, 1087-1100
CrossRef2
Michael D. Peck. (2011) Structure Fires, Smoke Production, and Smoke Alarms. Journal of Burn Care & Research 32:5, 511-518
CrossRef3
Elanor A. Sidman, David C. Grossman, Beth A. Mueller. (2011) Comprehensive Smoke Alarm Coverage in Lower Economic Status Homes: Alarm Presence, Functionality, and Placement. Journal of Community Health 36:4, 525-533
CrossRef4
Laura Hendrix, Anthony Charles, Valorie Buchholz, Samuel Jones, Bruce Cairns. (2011) Influence of race and neighborhood on the risk for and outcomes of burns in the elderly in North Carolina. Burns 37:5, 762-769
CrossRef5
Corinne Peek-Asa, Jingzhen Yang, Cara Hamann, Tracy Young. (2011) Smoke Alarm Tests May Not Adequately Indicate Smoke Alarm Function. Journal of Burn Care & Research 32:4, e135-e139
CrossRef6
Samantha Turner, Geri Arthur, Ronan A Lyons, Alison L Weightman, Mala K Mann, Sarah J Jones, Ann John, Simon Lannon, Samantha Turner. 2011. Modification of the home environment for the reduction of injuries. .
CrossRef7
Jonathan Corcoran, Gary Higgs, Angela Higginson. (2011) Fire incidence in metropolitan areas: A comparative study of Brisbane (Australia) and Cardiff (United Kingdom). Applied Geography 31:1, 65-75
CrossRef8
Corinne Peek-Asa, Jingzhen Yang, Cara Hamann, Michael P. Jones, Tracy Young, Craig Zwerling. (2010) Smoke Alarm and Battery Function 42 Months After Installation. American Journal of Preventive Medicine 39:4, 368-371
CrossRef9
C. Bradley Kramer, Frederick P. Rivara, Matthew B. Klein. (2010) Variations in U.S. Pediatric Burn Injury Hospitalizations Using the National Burn Repository Data. Journal of Burn Care & Research 31:5, 734-739
CrossRef10
William J.M. Holmes, Phoebe Hold, Malcolm I. James. (2010) The increasing trend in alcohol-related burns: It's impact on a tertiary burn centre. Burns 36:6, 938-943
CrossRef11
Linda S. Edelman, Lawrence J. Cook, Jeffrey R. Saffle. (2010) Burn Injury in Utah: Demographic and Geographic Risks. Journal of Burn Care & Research 31:3, 375-384
CrossRef12
Dominik Zenner, Tim Allison. (2010) Health of caravan park residents: A pilot cross-sectional study in the East Riding of Yorkshire. Health & Place 16:2, 309-314
CrossRef13
Lawrence R. Berger, L.J. David Wallace, Nancy M. Bill. (2009) Injuries and Injury Prevention Among Indigenous Children and Young People. Pediatric Clinics of North America 56:6, 1519-1537
CrossRef14
Nathaniel J. Bell, Nadine Schuurman, S. Morad Hameed. (2009) A small-area population analysis of socioeconomic status and incidence of severe burn/fire-related injury in British Columbia, Canada. Burns 35:8, 1133-1141
CrossRef15
Ju Ok Park, Sang Do Shin, Jaiyong Kim, Kyoung Jun Song, Michael D. Peck. (2009) Association between socioeconomic status and burn injury severity. Burns 35:4, 482-490
CrossRef16
D. Bruck, I. Thomas. (2009) Towards a better smoke alarm signal - an evidence based approach. Fire Safety Science 9, 403-414
CrossRef17
C. Mulvaney, D. Kendrick, E. Towner, M. Brussoni, M. Hayes, J. Powell, S. Robertson, H. Ward. (2008) Fatal and non-fatal fire injuries in England 1995-2004: time trends and inequalities by age, sex and area deprivation. Journal of Public Health 31:1, 154-161
CrossRef18
Jingzhen Yang, Corinne Peek-Asa, Michael P. Jones, David L. Nordstrom, Craig Taylor, Tracy L. Young, Craig Zwerling. (2008) Smoke Alarms by Type and Battery Life in Rural Households. American Journal of Preventive Medicine 35:1, 20-24
CrossRef19
Linda S. Edelman. (2007) Social and economic factors associated with the risk of burn injury. Burns 33:8, 958-965
CrossRef20
Eric R. Hawkins, Jane H. Brice, Barbara A. Overby. (2007) Welcome to the World. Pediatric Emergency Care 23:11, 790-795
CrossRef21
RA Lyons, A John, S Brophy, SJ Jones, A Johansen, A Kemp, S Lannon, J Patterson, B Rolfe, LV Sander, A Weightman. (2007) Cochrane review: Modification of the home environment for the reduction of injuries. Evidence-Based Child Health: A Cochrane Review Journal 2:2, 899-928
CrossRef22
Basil A. Pruitt, Steven E. Wolf, Arthur D. Mason. 2007. Epidemiological, demographic, and outcome characteristics of burn injury. , 14-32.
CrossRef23
Samantha Turner, Geri Arthur, Ronan A Lyons, Alison L Weightman, Mala K Mann, Sarah J Jones, Ann John, Simon Lannon, Samantha Turner. 2006. Modification of the home environment for the reduction of injuries. .
CrossRef24
Jane H. Brice, Barbara A. Overby, Eric R. Hawkins, Elizabeth L. Fihe. (2006) Determination of Infant-Safe Homes in a Community Injury Prevention Program. Prehospital Emergency Care 10:3, 397-402
CrossRef25
Mohammad Umair Anwar, Sanjib Majumder, Orla Austin, Alan Phipps. (2005) Smoking, Substance Abuse, Psychiatric History, and Burns: Trends in Adult Patients. Journal of Burn Care & Rehabilitation 26:6, 493-501
CrossRef26
Denise Kendrick, Caroline Mulvaney, Paul Burton, Michael Watson. (2005) Relationships between child, family and neighbourhood characteristics and childhood injury: A cohort study. Social Science & Medicine 61:9, 1905-1915
CrossRef27
Michael F. Ballesteros, Mark L. Jackson, Maurice W. Martin. (2005) Working Toward the Elimination of Residential Fire Deaths: The Centers for Disease Control and Prevention??s Smoke Alarm Installation and Fire Safety Education (SAIFE) Program. Journal of Burn Care & Rehabilitation 26:5, 434-439
CrossRef28
Cynthia D. Homer, David A. Engelhart, Eric S. Lavins, Amanda J. Jenkins. (2005) Carbon monoxide-related deaths in a metropolitan county in the USA: an 11-year study. Forensic Science International 149:2-3, 159-165
CrossRef29
L CHEN, S BAKER, G LI. (2005) Drinking history and risk of fatal injury: comparison among specific injury causes. Accident Analysis & Prevention 37:2, 245-251
CrossRef30
J. Sherrard, J. Ozanne-Smith, C. Staines. (2004) Prevention of unintentional injury to people with intellectual disability: a review of the evidence. Journal of Intellectual Disability Research 48:7, 639-645
CrossRef31
C. Peek-Asa, C. Zwerling, L. Stallones. (2004) Acute Traumatic Injuries in Rural Populations. American Journal of Public Health 94:10, 1689-1693
CrossRef32
E. D. Shenassa, A. Stubbendick, M. J. Brown. (2004) Social Disparities in Housing and Related Pediatric Injury: A Multilevel Study. American Journal of Public Health 94:4, 633-639
CrossRef33
R Barrow. (2004) Influence of demographics and inhalation injury on burn mortality in children. Burns 30:1, 72-77
CrossRef34
RA Lyons, LV Sander, AL Weightman, J Patterson, SA Jones, S Lannon, B Rolfe, A Kemp, A Johansen, Ronan Lyons. 2003. Modification of the home environment for the reduction of injuries. .
CrossRef35
Kelly A.Nordyke Messingham, Douglas E. Faunce, Elizabeth J. Kovacs. (2002) Review article. Alcohol 28:3, 137-149
CrossRef36
(2001) Deaths and Injuries from House Fires. New England Journal of Medicine 345:14, 1064-1065
Full Text37
Istre, Gregory R., McCoy, Mary A., Osborn, LindaBarnard, Jeffrey J., Bolton, Allen, . (2001) Deaths and Injuries from House Fires. New England Journal of Medicine 344:25, 1911-1916
Full Text38
Carolyn DiGuiseppi, Cynthia W Goss, Julian PT Higgins, Carolyn DiGuiseppi. 2001. Interventions for promoting smoke alarm ownership and function. .
CrossRef39
RW Byard, J Lipsett, J Gilbert. (2000) Fire deaths in children in South Australia from 1989 to 1998. Journal of Paediatrics and Child Health 36:2, 176-178
CrossRef40
James K. Hammitt, Eric S. Belsky, Jonathan I. Levy, John D. Graham. (1999) Residential Building Codes, Affordability, and Health Protection: A Risk-Tradeoff Approach. Risk Analysis 19:6, 1037-1058
CrossRef41
Patricia Brennan. (1999) Victims and survivors in fatal residential building fires. Fire and Materials 23:6, 305-310
CrossRef42
Peter Leth, Markil Gregersen, Svend Sabroe. (1998) Fatal Residential Fire Accidents in the Municipality of Copenhagen, 1991–1996. Preventive Medicine 27:3, 444-451
CrossRef43
A Waller. (1998) Adult thermal injuries in New Zealand resulting in death and hospitalization. Burns 24:3, 245-251
CrossRef44
ANNE KEANE MARY PICKETT LINDA ROBINSON B. (1998) A MODEL OF SURVIVORS' PSYCHOLOGICAL RESPONSES FOLLOWING A RESIDENTIAL FIRE. Death Studies 22:1, 43-60
CrossRef45
Rivara, Frederick P., Grossman, David C., Cummings, Peter, . (1997) Injury Prevention. New England Journal of Medicine 337:9, 613-618
Full Text46
Craig Zwerling, James A. Merchant, Stephen J. Reynolds, Leon F. Burmeister, Ann M. Stromquist. (1997) Risk Factors for Injury in Rural Iowa. Journal of Agromedicine 4:3-4, 295-302
CrossRef47
Guohua Li, Susan P. Baker, Sophia Sterling, John E. Smialek, Patricia C. Dischinger, Carl A. Soderstrom. (1996) A Comparative Analysis of Alcohol in Fatal and Nonfatal Bicycling Injuries. Alcoholism: Clinical and Experimental Research 20:9, 1553-1559
CrossRef48
Mallonee, Sue, Istre, Gregory R., Rosenberg, Mark, Reddish-Douglas, Malinda, Jordan, Fred, Silverstein, Paul, Tunell, William, . (1996) Surveillance and Prevention of Residential-Fire Injuries. New England Journal of Medicine 335:1, 27-31
Full Text49
D.J. Barillo, R. Goode. (1996) Fire fatality study: demographics of fire victims. Burns 22:2, 85-88
CrossRef50
Tim Squires, Anthony Busuttil. (1995) Child fatalities in scottish house fires 1980–1990: A case of child neglect?. Child Abuse & Neglect 19:7, 865-873
CrossRef51
Paul Williams. (1995) A regulation evaluation system: a decision support system for the Building Code of Australia. Construction Management and Economics 13:3, 197-208
CrossRef52
G Li, G S Smith, S P Baker. (1994) Drinking behavior in relation to cause of death among US adults.. American Journal of Public Health 84:9, 1402-1406
CrossRef53
Guohua Li, Susan P. Baker. (1994) Alcohol in fatally injured bicyclists
. Accident Analysis & Prevention 26:4, 543-548
CrossRef54
Arthur L. Klatsky, Mary Anne Armstrong. (1993) Alcohol Use, Other Traits, and Risk of Unnatural Death: A Prospective Study. Alcoholism: Clinical and Experimental Research 17:6, 1156-1162
CrossRef55
Baker, Susan P., . (1992) What Keeps the Home Fires Burning?. New England Journal of Medicine 327:12, 887-888
Full Text
