Caregiver Stress Increases Risk of Homicide-Suicide
by Donna Cohen, Ph.D.| Geriatric Times |  |
November/December 2000 | |
Vol. I | |
Issue 4 |
Homicide-suicides are lethal events involving a perpetrator who kills one or more people-usually one-and then commits suicide within a very short period. It is not known exactly how many homicide-suicides occur in the United States because there is no national surveillance system. Experts estimate, however, that homicide-suicides account for about 1,000 to 1,500 deaths annually, a mortality similar to meningitis and tuberculosis (Marzuk et al., 1992).
Homicide-suicides are emerging as a public health challenge, especially in the older population. Descriptive epidemiological research in Florida has shown that rates in the population aged 55 years and older are higher than rates in younger age groups (Cohen et al., 1998). This study also found that homicide-suicides account for about 3% of all suicides and about 12% of all homicides in the older population. Applying these Florida figures to United States data, it is estimated that about 200 homicide-suicides occur every year among people 55 and older (Malphurs et al., in press).
Most of these acts involve older men killing their spouses or lovers. Our Florida studies also suggest that for each unsuccessful homicide-suicide (i.e., one person, usually the perpetrator, survives), there are five that are successful. As a result, there are clinical and social issues that need to be addressed to improve our capabilities to intervene and prevent these tragedies.
Although many papers have been published about homicide-suicide since 1900, it was not until 1995 that rates and clinical patterns were empirically studied in older people (Cohen, in press; Cohen, 1995). Berman (1979), who was one of the first to conduct psychological autopsy studies of a small series of couples, reported the important role of depression in the perpetrator. In the few papers where older populations and victims were mentioned, the number of events were simply reported, or the deaths were attributed to sickness and frailty.
Homicide-suicides in the old were presumed to be pacts, mercy killings or altruistic actions (McIntosh et al., 1995), and the two most widely used taxonomy systems, which use age, relationship and presumed motivation to classify homicide-suicides, also have assumed homogeneity of motive: sick older men are killing sick wives as acts of mercy, altruism or mutual consent before committing suicide (Hanzlick and Koponen, 1994; Marzuk et al., 1992). Although the results of our research verify that men are most often the perpetrators, we have found these homicide-suicides are not mercy killings, altruistic events or suicide pacts. They are not acts of love or adoration, and they are not compassionate homicides. They are acts of desperation and depression, other forms of psychopathology, or domestic violence. Suicide pacts occur in older couples, but they are very rare-less than one-tenth of one percent of all suicides in the older population.
Clinical Patterns
About 85% of homicide-suicides involve spouses or consorts, and the remaining victims are siblings or other family members (Berman, 1979; Nock and Marzuk, 1999). Older people rarely kill children and nonfamily members. There are at least three types of spousal/consortial homicide-suicide: dependent-protective, aggressive and symbiotic (Cohen, 2000; Cohen and Eisdorfer, 1999). A common feature of all three is the perpetrator's perception of separation and an unacceptable threat to the integrity of the relationship. Clinicians need to be alert to the warning signs of a homicide-suicide.
Half of all spousal/consortial homicide-suicides are the dependent-protective subtype. The husband, usually two to four years older than his wife, may or may not have a serious illness. In most circumstances, however, he is caring for a wife who is chronically ill. There is evidence of serious depression-including helplessness, hopelessness and vital exhaustion-which in most circumstances has gone undetected and untreated despite frequent medical care contacts. Most of these men have seen a physician within a few weeks of committing the homicide-suicide.
Dependent/protective homicide-suicides are not impulsive acts. The men who commit these acts have usually thought about or planned the dyadic deaths for months or sometimes more than a year. Although illness, age, depression and other life stressors may be predisposing factors, a real or perceived decline in the perpetrator's physical health and discussions about, or a pending move to, a nursing home or assisted-living residence are precipitating factors.
Caregiving responsibilities over time appear to cause significant strain and depression in perpetrators of a dependent-protective homicide-suicide (Malphurs et al., in press). These men are consistently described by surviving family and friends as having dominant or controlling personalities. Thus, depression and helplessness coupled with a perceived inability to fix the situation, make the wife better or protect her appear to be significant risk factors for homicide-suicide.
One of our most distressing findings is evidence that the older women who are killed are not knowing or willing participants (Cohen and Eisdorfer, 1999). Most are shot in their sleep or in the back of the head or chest. It appears that homicide-suicides are unilateral decisions by men with controlling personalities with no evidence from surviving informants that the homicide victim had spoken about wanting to be dead or be killed.
The symbiotic subtype accounts for 20% of older homicide-suicides. In these cases, the male perpetrator tends to be a few years older than the victim, and both the husband and wife are sick. There is no suicide note signed by both parties, but neighbors and/or family members have reported that both individuals had talked about wanting to die or being better off dead. When the homicide victim had Alzheimer's disease, a related brain disease or was unable to communicate, the event would not be classified as a symbiotic homicide-suicide, because this person could not have participated in the decision-making process.
One-third of older homicide-suicides are the aggressive subtype where there is a history of verbal and/or physical conflict and/or domestic violence. The male perpetrators are usually nine to 10 years older than their victims. Neither the perpetrator nor victim has a physical illness. What usually triggers the homicide-suicide is the victim talking about separation or divorce, or making plans for or actively moving out of the home. The action is usually a surprise attack, the homicide is usually violent, and the victim is shot or stabbed multiple times.
Intervention and Prevention
Since older perpetrators have usually thought about the act for a long time, there is a window of opportunity for the prevention of a homicide-suicide. Clinicians should assess the risk for homicide-suicide in all older patients where the following factors exist: 1) the patient has been married for a long time and one or both members of the couple have real or perceived health problems; 2) there is evidence or reports of domestic conflict, discord, violence, the existence of a restraining order, or a pending separation or divorce; and/or 3) there is a history of ideation about suicide, homicide or violence.
In the case of dependent-protective and symbiotic homicide-suicides, the older couple has been married a long time, they appear to be interdependent on each other and the husband has a dominant, commanding or controlling personality. This does not mean that the wife has a submissive personality, but rather that the man has a strong need to control decisions in the relationship. The wife is usually sick, frail, and chronically or terminally ill. The husband is usually the primary caregiver (even when others give some assistance) and is depressed. Results from a recent case control study of older married men who committed only suicide and those who committed a homicide-suicide show that the men who committed suicide were care recipients in contrast to the homicide-suicide perpetrators who were caregivers (Malphurs and Cohen, unpublished data).
Since the male is usually the perpetrator and a caregiver, it is essential to query the wife, if possible, during medical care contacts and to evaluate the husband for depression and other psychiatric problems. Assessment can be complicated since the potential victim rather than the potential perpetrator may be the patient, and the perpetrator may resist an evaluation (Nock and Marzuk, 1999). Discuss your concerns regarding lethal violence with the physician of your patient's spouse, and suggest that an appointment be set with the spouse.
If you have older couples in your practice, consider administering brief depression screening instruments, such as the Center for Epidemiologic Studies depression scale (CES-D), while couples are sitting in the waiting room. Inquire about changes in the husband's behavior-increased anxiety and agitation, giving possessions away, talk of feeling helpless or hopeless or being too exhausted to go on, crying, and/or difficulty sleeping or sleeping too much. Ask the wife whether the couple has been fighting, if there have been discussions about divorce, a history of estrangement or threats to harm her.
Ask whether the husband has guns and ammunition in the house or has considered buying them. Inquire whether the husband has given a house key to a neighbor, friend or attorney. Talk with adult children and other caregivers such as home-health aides about threats. It is not uncommon for the perpetrator to tell children or close relatives, "there is something I must do." Although adult children may suspect suicide, a homicide-suicide is usually not part of the equation for them.
Women almost never commit a homicide-suicide, but the wish to kill a spouse and themselves is not uncommon. Although women caregivers have a greater prevalence of depression, most have little experience with guns, the most common method for a homicide-suicide; and their approach to caregiving is less task-focused than men and more emotion-focused. When older husbands or wives express homicidal and suicidal ideation or talk about a specific method to kill themselves and their spouse, it is essential to take the threat seriously.
There are several situations when the danger of homicide-suicide is especially high: the period prior to and after a move to a long-term care residence; circumstances during which an individual is spending days at the hospital bedside of a terminally ill spouse or in which the spouse is about to be discharged from the hospital to a nursing home, hospice or home; and situations where a person has been spending most of their waking hours for months or more with a spouse who is institutionalized. It is important for staff members in health care facilities to be aware that these are high-risk periods and to talk with the patient, spouse-caregiver and other family members. In a number of cases the men have told one or more adult children of their intentions, and the children were too embarrassed to tell anyone.
Summary
The strong evidence of undetected and untreated depression in almost all perpetrators of homicide-suicides and the existence of domestic violence in one-third of older homicide-suicides emphasizes the importance of careful interviews when one or both spouses are in the acute or long-term care system. Interventions should include intensive treatment of depression and other psychiatric problems when appropriate, removal of guns or other lethal weapons, social support for spouses and families in caregiving situations, and appropriate interventions to deal with marital conflict-especially where the woman is a potential victim of aggressive, lethal behavior.
Intervention is complicated and should be done on a case-by-case basis. Separating the perpetrator and victim is usually appropriate to diffuse the tension and protect the victim. A careful clinical plan is essential, however, since separation is often the trigger for violence. Finally, clinical staff and family members as well as individuals who have continuing contact with older people-clergy, senior center staff, home health care staff and others-should be educated about warning signs and trained to ask questions and get appropriate help.
Dr. Cohen is professor in the department of aging and mental health and the department of psychiatry and behavioral sciences at the University of South Florida in Tampa. She also heads the Senior Violence and Injury Prevention Program.
References
Berman AL (1979), Dyadic death: murder-suicide. Suicide Life Threat Behav 9(1):15-23.
Cohen D (in press), An update on homicide-suicide in older persons: 1995-2000. Journal of Mental Health and Aging 6(3).
Cohen D (2000), Homicide-suicide in older people. Psychiatric Times 17(1):49-52.
Cohen D (1995), Homicide-suicide in the aged: a growing public health problem. Journal of Mental Health and Aging 1(2):83-84.
Cohen D, Eisdorfer C (1999), Characteristics of perpetrators of homicide-suicide in older couples. Session 5-07-1. Presented at the 9th Congress of the International Psychogeriatric Association. August 24; Vancouver, British Columbia.
Cohen D, Llorente M, Eisdorfer C (1998), Homicide-suicide in older persons. Am J Psychiatry 155(3):390-396.
Hanzlick R, Koponen M (1994), Murder-suicide in Fulton County, Georgia, 1988-1991. Comparison with a recent report and proposed typology. Am J Forensic Med Pathol 15(2):168-173.
Malphurs J, Eisdorfer C, Cohen D (in press), A comparison of antecedents of homicide-suicide and suicide in older married men. Am J Geriatr Psychiatry.
Marzuk PM, Tardiff K, Hirsch CS (1992), The epidemiology of murder-suicide. JAMA 267(23): 3179-3183 [see comment].
McIntosh JL, Santos JF, Hubberd RW, eds. (1995), Elder Suicide: Research, Theory, and Treatment. Washington, D.C.: American Psychological Association.
Nock MK, Marzuk PM (1999), Murder-suicide: phenomenology and clinical implications. In: The Harvard Medical School Guide to Suicide Assessment and Intervention, Jacobs DG, ed. San Francisco: Jossey-Bass.