Suicide

In the absence of adequate prospective studies, the psychological autopsy (PA) method has been used to reconstruct a detailed picture of the victim’s psychological state prior to death, including psychiatric symptomatology, behavior, and life circumstances during the weeks or months before death. This includes interviewing knowledgeable informants, reviewing available clinical records, and comprehensive case formulation by one or more mental health professionals with expertise in postmortem studies. The PA method has been used to provide an inclusive, well-defined sample of all persons who die by suicide within a defined catchment area, region, or population. One of the most striking and consistent findings of the PA method is that psychiatric disorder and/or substance use is present in about 90 percent of all suicides, with affective disorder as the most common psychopathology, followed by substance use and schizophrenia (Conwell and Brent).

When compared to younger suicide victims, older victims are more likely to have had a physical illness, and to have suffered from depression that is not comorbid with a substance disorder (Conwell and Brent). The type of depression found in the majority of later life suicides is usually a first episode of depression, uncomplicated by psychoses or other comorbid psychiatric disorders, and, ironically, is the most treatable type of late-life depression. Such age-related patterns have appeared in reports from a number of countries including the United States, Finland, and the United Kingdom.

Although substance use is less frequent among elderly suicides, there is some evidence that among the ‘‘young old,’’ alcohol may be a correlate. For men with early onset alcoholism who have survived to their fifties and sixties, the combination of continued alcohol abuse and burn out among their social support network may be lethal. Murphy and his associates described that for older male alcoholics, loss of the last social support can be a pivotal event in suicide risk (Murphy, Wetzel, Robins, and McEvoy). How current, as well as past alcohol abuse, lowers the threshold for suicidal behavior in later life requires further systematic examination. Although it is often assumed that medication misuse (e.g., benzodiazepine dependence, psychotropic medication with alcohol abuse) is a risk for late life suicide, there is little published information on this topic.

Despite high rates of dementia and delirium in later life, few studies have found these diagnoses to be risk factors for suicide (Conwell and Brent). Controlled PA studies are needed to determine what other factors in combination with mental and physical disorders are related to risk for later life suicide.

The PA has been used to explore possible personality traits that may increase risk for later life suicide (Duberstein). Duberstein used an informant-based personality inventory measure to examine possible personality traits among older and younger suicides, relative to age- and sex-matched controls. The inventory measured five general personality traits: neuroticism, extroversion, openness to experience, agreeableness, and conscientiousness. Suicides were found to have higher neuroticism scores than normal controls, and older suicide victims had lower openness to experience scores than both younger suicides and normal controls.

Hopelessness, a set of beliefs related to lack of anticipated positive outcomes about the future, has also been examined in the context of later life suicide. A prospective investigation of a retirement community found a single item asking about hopelessness was related to later completed suicides (Ross, Bernstein, Trent, Henderson, and Paganini-Hill).

Suicide intent has also been examined in older adult suicide victims. Using the PA method, older adults were found to be more intent compared to younger suicide victims (Conwell, Duberstein, Cox, Herrmann, Forbes, and Caine). That is, older adults were more likely to have avoided intervention, taken precautions against discovery, and were less likely to communicate their intent to others. Moreover, older men, in particular, were less likely to have had a history of previous attempts.

The PA method has also been used to examine patterns of health services use among suicide victims. Health services for older adults who later suicided was typically available, and used. A number of reports indicate that approximately 70 percent of older suicide victims had seen a primary health care provider within a month (Conwell). In contrast, few older adult suicide victims have had a history of mental health care.