Skip Navigation LinksSkip Navigation Links
Centers for Disease Control and Prevention
Safer Healthier People
Blue White
Blue White
bottom curve
CDC Home Search Health Topics A-Z spacer spacer
Blue curve MMWR spacer

Deaths Among the Homeless -- Atlanta, Georgia

Between July 1, 1985, and June 30, 1986, the Office of the Fulton County Medical Examiner in Georgia investigated 40 deaths occurring among the homeless. All of the deaths occurred in Atlanta, 95% of which is located in Fulton County. In 1986, Atlanta had an estimated homeless population of 4,000 to 7,000 (Task Force for the Homeless, unpublished data). Based on these figures, the crude death rate among the homeless for that year was 5.7 to 10.0/1,000.

The medical examiner's (ME's) office identified these deaths by reviewing the 2,380 deaths reported during the 1-year period being studied. A decedent was considered homeless if there was no address available at the time of death or if the available address could not be considered a residence. The city directory was used to verify addresses. The ME's records supplied information on the age, sex, and race of the decedent; the location and date of death; and the results of autopsy (performed in 23 cases) and toxicologic examinations (performed in 35 cases).

Black males accounted for 19 (48%) of the 40 deaths; black females, for three (8%); and white males, for 18 (45%). The age at death was known for 36 of the 40 decedents; the median age for this group was 44 (range = 21-70 years). For black men, the median age at death was 43 (range = 22-56 years), and, for white men, it was 53 (range = 23-70 years). The age at death was known for two of the three women; one was 21, and the other, 63. The address of 11 of the decedents was a shelter.

Twenty-two persons (55%) died or were found dead outdoors; one was in a parked car. Of the 18 persons who died indoors, seven were found in vacant buildings; five, at shelters; three, in houses or apartments; one, in prison; and one, in an alcohol treatment unit; one was killed while in a store during a robbery. Two to five deaths occurred each month except October, when eight persons died--four in a single fire.

Cause of death was determined from the medical history, the scene investigation, circumstances of death, and autopsy and toxicologic studies, when performed (Table 1). The ME categorized the manner of death as either natural (the consequence of a disease or of the aging process), accidental (unintentional), homicidal, or suicidal.

Sixteen deaths (40%) were classified as natural. Six of these were attributed to chronic alcohol (ethanol) abuse. Only one of the six had a measurable level of blood alcohol (198 mg%). The ME determined causes of the other five deaths from the circumstances of death and the medical history. Ten deaths resulted from natural causes other than the direct effect of alcohol. Three of these were from seizures (probably due to alcohol withdrawal); four, from heart disease; and three, from lung disease.

The ME classified 19 deaths (48%) as accidental. Seven of these resulted from acute alcohol toxicity (mean blood alcohol = 498 mg%, range = 296 mg%-610 mg%). Twelve resulted from accidental injuries and included six deaths from fires, two from hypothermia, two from pedestrian-motor vehicle incidents, one from drowning, and one from a fall.

There were four homicides and one suicide. Although blood alcohol was measurable for one of the homicide victims, the ME determined that alcohol was not causally associated with the death.

The ME determined that 28 (70%) of the 40 deaths were alcohol-related. Although blood alcohol was not measured or measurable at the time of death for 11 of these decedents, their deaths were classified as alcohol-related either because they had a history of alcohol abuse or because of the circumstances of death. Three of these 11 decedents died from seizures consistent with alcohol withdrawal. One, who died from a fall, had had measurable blood alcohol in a sample taken from a subdural hematoma sustained in the fall. Another, who died 10 days after being struck by an automobile, had had measurable blood alcohol upon admission to the hospital. One, who was hospitalized for burns prior to death, was clinically judged to be intoxicated when admitted to the hospital. The other five died from the effects of chronic alcohol abuse.

No deaths were attributed to drugs other than alcohol. Of 31 decedents screened for barbiturates, benzodiazepines, phenytoin, and other weakly acidic or neutral drugs, three (10%) were positive (one for barbiturates, one for phenytoin, and one for barbiturates and phenytoin). All of the drugs were present at therapeutic or subtherapeutic levels. Thirteen persons were screened for cocaine or cocaine metabolites in their urine, and one was positive. Four (20%) of the 20 screened for cannabinoids were positive. Reported by: R Hanzlick, MD, Office of the Fulton County Medical Examiner. Surveillance and Programs Br, Div of Environmental Hazards and Health Effects, Center for Environmental Health, CDC.

Editorial Note

Editorial Note: Most deaths among the homeless in Atlanta occurred among men 60 years of age and were alcohol-related. The high proportion of alcohol-related deaths reported in this study underscores the potentially serious health consequences of alcohol use or abuse. These consequences include those resulting from the acute intoxicating effects of alcohol, alcohol withdrawal syndrome, and the effects of chronic alcohol abuse.

Little information is available on the size and composition of the homeless population or the health problems and causes of death among this group. Estimates of the number of homeless persons in the United States vary widely (1,2). However, there is general agreement among health service providers that the number of homeless is increasing and that a growing proportion of these are young and female (3,4).

Homeless persons have been characterized as extremely poor, significantly disabled by mental or physical illness, and socially isolated. Marginal ties to family and others have been identified as a significant contributor to homelessness (5). Forty percent of homeless persons have psychiatric illnesses (6). Physical health problems among the homeless include trauma, respiratory disease, tuberculosis, scabies and pediculosis infestations, peripheral vascular disease, and chronic illnesses, such as diabetes mellitus, that are exacerbated by adverse living conditions and lack of health care (7).

MEs investigate sudden or unexpected deaths, violent deaths, and deaths to persons unattended by a physician. Since the homeless often die suddenly and without a physician, many of these deaths are investigated by an ME. However, an unknown number of homeless persons die while hospitalized, and their deaths are not routinely investigated. Despite this limitation, ME's records are one of the few sources of information available for describing deaths among the homeless.

Additional studies are needed to describe the characteristics of deaths among the homeless more completely. Studies on such deaths in other parts of the country are needed. A better understanding of the causes and circumstances of these deaths would help in developing public health programs to prevent them. For now, this limited study suggests that, although providing shelters might prevent deaths from hypothermia and some fires, this intervention alone will not prevent most deaths among the homeless.


  1. Office of Policy Development and Research, US Department of Housing and Urban Development. Report to the Secretary on the homeless and emergency shelters. Washington, DC: US Department of Housing and Urban Development, 1984:19.

  2. Hombs ME, Snyder M. Homelessness in America: a forced march to nowhere. 2nd ed. Washington, DC: Community for Creative Nonviolence, 1983:xvi.

  3. US General Accounting Office. Homelessness: a complex problem and the federal response. Gaithersburg, Maryland: US General Accounting Office, 1985; publication no. (GAO/HRD)85-40.

  4. Marwick C. The 'sizable' homeless population: a growing challenge for medicine. JAMA 1985;253:3217-9,3223-5.

  5. Rossi PH, Wright JD, Fisher GA, Willis G. The urban homeless: estimating composition and size. Science 1987;235:1336-41.

  6. Lamb HR, Talbott JA. The homeless mentally ill: the perspective of the American Psychiatric Association. JAMA 1986;256:498-501.

  7. Brickner PW, Scanlan BC, Conanan B, et al. Homeless persons and health care. Ann Intern Med 1986;104:405-9.

Disclaimer   All MMWR HTML documents published before January 1993 are electronic conversions from ASCII text into HTML. This conversion may have resulted in character translation or format errors in the HTML version. Users should not rely on this HTML document, but are referred to the original MMWR paper copy for the official text, figures, and tables. An original paper copy of this issue can be obtained from the Superintendent of Documents, U.S. Government Printing Office (GPO), Washington, DC 20402-9371; telephone: (202) 512-1800. Contact GPO for current prices.

**Questions or messages regarding errors in formatting should be addressed to

Page converted: 08/05/98


Safer, Healthier People

Morbidity and Mortality Weekly Report
Centers for Disease Control and Prevention
1600 Clifton Rd, MailStop E-90, Atlanta, GA 30333, U.S.A


Department of Health
and Human Services

This page last reviewed 5/2/01