The rate of suicide in the elderly population has been increasing since 1980, reversing a 40-year declining trend in the suicide rate for this age group (1). From 1980 to 1992, the rate of suicide among elderly persons increased 36%(1). In 1992, suicide was the third leading cause of injury-related deaths among older U.S. residents, following deaths from unintentional falls and unintentional motor vehicle crashes (2). In 1994, this population had the highest rate of suicide of any age group (1). Young adults characteristically experience job, financial, and relationship problems in the weeks to months before suicide, whereas elderly people experience loss and physical illness (3). To assist the forensic investigator and pathologist to more accurately assign cause and manner of death in alleged suicides, it is important to be aware of the risk factors and other causative features of suicide among the elderly. The authors report an analysis of such variables to further enhance the understanding, investigation, and possible prevention of elderly suicide.
MATERIALS AND METHODS
We retrospectively reviewed all cases referred to the Medical University of South Carolina Forensic Pathology Section at the Charleston County Medical Examiners’ Office (Charleston, SC) from January 1988 through December 1997. The cases were primarily from Charleston County, with the remainder from the various 46 counties in the state. South Carolina has a coroner system, and referral of cases is at the discretion of the coroner. We examined age, race, and sex of the victim; method of suicide; time of year; and toxicologic results for victims 65 years of age and older. The files also were reviewed for medical history, psychiatric history, location of the suicide, previous attempts, personal events before suicide, and whether the suicide victim left a note.
A total of 678 suicides were referred to our office during the 10-year period examined. Of all the cases, 78 victims were 65 years and older, accounting for 11.5% of all the suicides. Homicide and suicide scenarios accounted for 6.4% (5 cases) of elderly suicides. The ages ranged from 65 to 94 years, with an average age of 73 years. Men accounted for 85% of the cases, and whites, 94%, giving a male-to-female and white-to-black ratio of 6:1 and 15:1, respectively (Table 1). Demographic data for the year 1998 in South Carolina is as follows for the total population: whites, 69%; nonwhites, 31%; males, 48%; females, 52%; and individuals 65 years and older, 12%.
Gunshot wound was the leading method of suicide, accounting for 80.7% (63 cases). Other methods consisted of overdose, 6.4% (5 cases); hanging, 3.8% (3 cases); incised wounds, 2.6% (2 cases); drowning, 2.6% (2 cases); fall from height, 2.6% (2 cases); and carbon monoxide poisoning 1.3% (1 case). The time of year for suicide cases was evenly distributed between months, ranging from 5% to 14% (Table 2). For the holiday season, extending from late November to early January, the number of cases was not increased. Of all cases, 41% (32 cases) were autopsied and 59% (46 cases) were examined externally.
Examination of the 63 gunshot wounds revealed that 84.1% (53 cases) were to the head; 14.3% (9 cases) to the chest; and 1.6% (1 case) to the abdomen. Handguns were used by 90.48% (57 cases) of the victims; shotguns, 4.76% (3 cases); and rifles, 4.76% (3 cases). The most common type of handgun used was the .38 caliber revolver, which was used by 28% of the victims (16 cases).
Eighty-five percent (66 cases) of the suicides occurred in the home. The bedroom was the most common site in the home, accounting for 29% (19 cases). For two of the cases that occurred away from the home, one victim committed suicide while an inpatient at a local hospital and another, in the parking lot of a local hospital.
Review of the victims’ medical histories revealed that 26% (20 cases) had a premortem diagnosis of a malignancy. Adenocarcinoma of the prostate accounted for 40% (8 cases) of the cancers, and various lung malignancies accounted for 20% (4 cases). Forty-six percent (36 cases) of the victims had a chronic or debilitating disease other than a malignancy that significantly altered their lifestyle. Review of psychiatric histories showed that 26% (20 cases) had a documented psychiatric illness, with depression accounting for 90% (18 cases); bipolar affective disorder, 5% (1 case); and anxiety disorder, 5% (1 case). A suicide note was left by 23% of the victims (18 cases). Only two of the victims (2.6%) had had a documented previous suicide attempt. In 2.6% (2 cases), the victims had had an immediate family member who committed a documented suicide.
Toxicologic analysis was obtained for 86% (67 cases) of the victims. Of the victims that received toxicologic evaluation (Table 3), 27% (21 cases) had a prescription medication identified. In 42% of the victims (28 cases), the toxicologic results were completely negative. The bodies of 24% (16 cases) of the victims contained ethanol; in 6 of these cases (37.5%), the victim was legally intoxicated. South Carolina defines legal ethanol intoxication as greater than or equal to 100mg/dl. An illegal substance was not identified in any of the victims.
Variables designated as life stressors were evaluated and consisted of a recent family death, divorce, argument, financial problems, nursing home placement, becoming a burden on the family, involvement in crimes, property damage, and eviction. Review of case histories revealed that 35% (27 cases) of the victims experienced one or more of the aforementioned life stressors.
A great deal of research has been focused on suicide in the young, but the elderly are at the highest risk of suicide in many countries (4). More than 6000 elderly adults commit suicide each year, and in 1988, it was estimated there was an elderly person dying by suicide every 83 minutes (5). After a 40-year declining trend, the suicide rate in the 65-year and older age group has increased since 1980 (2). In 1992, this age group comprised about 13% of the population but accounted for approximately 20% of all suicides (2). The study of suicide in a population makes possible general associations between suicide scenarios and the personal characteristics of those individuals who kill themselves (6).
Rates of suicide for males increase with age and reach their highest levels in the oldest age groupings, whereas rates for females increase with age, peak in middle adulthood, and decline slightly with advancing age (5). Results of this study concur with national gender rates in that 85% of the victims were males. The race differs also in suicide victims. Rates for the white population increase with age and peak in older adulthood (5). The rates for nonwhite suicides peak in younger adulthood and decline to low levels in older adulthood (5). Ninety-four percent of the cases in the present study were white, whereas only five victims were black.
Firearms have become the most common suicide method in both elderly men and elderly women (1). In this study, 80.7% of the suicides were via gunshot wounds; of these, 84.1% were to the head. Review of the current cases in this study revealed that the .38 caliber revolver was the most common weapon used. Hanging, overdose, carbon monoxide poisoning, incised wounds, drowning, and fall from height accounted for only 19.3% of the cases. It has been reported that the elderly use more lethal means of committing suicide and are more successful with any given suicide attempt than are younger persons (1).
Older persons have more physical illnesses than do younger persons, and our results show that 46% of the victims had a chronic or debilitating disease and that 20% had a premortem diagnosis of a malignancy. The most common cancer was adenocarcinoma of the prostate. Studies have shown that physical illness is present in 25% to 75% of all suicide victims (7). In the current study, 26% had a documented psychiatric illness, with the majority suffering from depression. Both depression and many other psychiatric disorders are a significant risk factor for suicide in the elderly (8).
To understand and investigate suicides, it is important to dispel many common myths and falsehoods. Studies have shown that there is no seasonal correlation between suicide and the time of year. Studies do show slight increase in suicides in the spring and fall, but contrary to popular belief, they are not increased around major holidays at the end of the year (7). February and October were the two most common months for suicide in the current study, with January and March containing the fewest.
It is interesting that 42% of the victims had a completely negative result for toxicologic evaluation, although the results may be skewed by the fact that not all cases of elderly suicide are referred to the Medical Examiners’ Office by the coroners throughout the state. Of the 24% of victims who were identified as containing ethanol, 62.5% were not legally intoxicated. The elderly are more likely to be taking medications and probably would have more opportunities to overdose. However, only 6.4% of these victims had overdosed.
It has been reported that suicide notes written by young people are longer, rich in emotions, and often begging for forgiveness. Suicide notes written by the elderly were shorter, contained specific instructions, and were less emotional (9). According to Ho et al. (9), about 20% of the suicide population leave suicide notes, and the current study shows that 26% of the elderly victims left a note. Suicide notes tend to be left more often by young females, by victims with religious beliefs, and by victims without a history of previous suicide attempts.
A review of case histories in the current study revealed that the majority of elderly suicides occurred in the victims’ homes. More specifically, the most common room used in the home was the bedroom, accounting for 29% of the cases. A previous study by Lee et al. (10) showed that the bedroom was also the most common for pediatric suicides.
The fastest growing population is that of older adults, and this trend is expected to continue, accelerating even more in the future (5). As the elderly population continues to grow, suicides in this age group will continue, and possibly will increase in number. Our study shows that the most common elderly suicide victim is a white male with an average age of 73 years, and the most common method is a gunshot wound. The victim is most likely to have a chronic or debilitating disease or malignancy, and the home is the most common location of the suicide. Toxicology is often negative in cases referred to the Medical Examiners’ Office by coroners throughout the state, and a note need not be left by the victim. To help in the understanding of elderly suicide, we must be familiar with its risk factors, victim demographics, methods, and scenarios. We hope that the results of this current study will aid death investigators in dealing with this complex issue.
1. Kaplan MS, Adamek, ME, Rhoades JA. Prevention of elderly suicide
: physician assessment of firearm availability. Am J Prev Med 1998; 15:60–4.
2. The Centers for Disease Control and Prevention. Suicide
among older persons—United States: 1980–1992. MMWR Morb Mortal Wkly Rep
3. Conwell Y, Rotenberg M, Caine ED. Completed suicide
at age 50 and over. J Am Geriatr Soc 1990; 38:640–4.
4. Leenaars AA. Suicide
notes of the older adult. Suicide
Life Threat Behav 1992; 22:62–79.
5. McIntosh JL. Epidemiology of suicide
in the elderly. Suicide
Life Threat Behav 1992; 22:15–33.
6. Allen N. Suicide
statistics. In Hatton CL, Valente SM, eds. Suicide assessment and intervention
. 2nd ed. East Norwalk: Appleton-Century-Crofts, 1984:17–31.
7. Kaplan HI, Sadock BJ. Synopsis of psychiatry
, 6th ed. Baltimore: Williams and Wilkins, 1991.
8. Meehan PJ, Saltzman LE, Sattin RW. Suicides among older United States residents: epidemiologic characteristics and trends. Am J Public Health 1991; 81:1198–1200.
9. Ho TP, Yip PSF, Halliday P. Suicide
notes: what do they tell us. Acta Psychiatr Scand 1998; 98:467–73.
10. Lee CJ, Collins KA, Burgess SE. Suicide
under the age of eighteen: a ten year retrospective study. Am J Forensic Med Pathol 1999; 20:27–30.