Demographic factors (e.g., sex, age, marital status, race, geographic location) provide a general indication of those groups in the general population that are at the highest risk of suicidal behaviors (American Association of Suicidology [AAS], 2004a, American Association of Suicidology [AAS], 2004c and Garrison, 1992). It has been identified that demographic profiles of suicide attempters and completers are different.

In the United States, males complete suicide at a rate four times that of females (in 2001: accordingly, 17.6/100,000 and 4.1/100,000). Suicide rates are the highest among the elderly (age 75 and older), the divorced, the separated, and the widowed. In regard to race and ethnicity, whites die by suicide twice as often as nonwhites, and Native Americans have the highest overall suicide rate, although there are differences between tribal groups. Blacks and Hispanics exhibit low risk of suicide; however, the suicide rate is increasing faster among African-American youths than among Caucasians. White men over the age of 85 are at the greatest risk of all demographic groups. Statistics reveal a suicide rate in this population of 15.9 per 100,000, which is almost six times the overall national suicide rate in the United States. Geographically, suicide rates are the highest in the Mountain states (in 2001: 16.2/100,000) (CDC, 2006).

Nonfatal suicidal behaviors are more frequent among the young and among females, who make three to four times as many attempts as males. African-American females are more likely to attempt suicide, but males in this ethnic group are more likely to die by suicide. Although the elderly make suicidal attempts less frequently than individuals in other age groups, their attempts are more lethal. For example, in this group, the ratio of attempted to completed suicide is 4:1, while for all ages combined the ratio is 20:1.

Although demographic factors reflect high-risk groups in the general population, there is little reliability in the predictive probability that a particular person will engage in suicidal behavior. Therefore, the consideration of other types of information is necessary to improve the sensitivity and specificity of suicide risk assessment.

Psychological autopsy studies (i.e., a postmortem examination of decedent presuicide risk factors, behaviors, and method of death) indicate that more than 90% of persons completing suicide have one or more mental disorders. These results are supported by data obtained using other methods of research (i.e., prospective follow-up studies and retrospective reviews of medical records). Suicide and nonfatal suicidal behaviors occur more frequently than expected among individuals with the diagnosis of mental disorder, and when coexisting disorders are identified, the risk is even greater (Jacobs et al., 1999 and Tanney, 1992).

Harris and Barraclough (1997) conducted a meta-analysis of 249 reports (published between 1966 and 1993) and found that 36 of 44 diagnoses (according to the DSM-III-R or ICD systems) had significantly raised standardized mortality ratios for suicide. These data led them to the conclusion that “if these results can be generalized, then virtually all mental disorders have an increased risk of suicide excepting mental retardation and possibly dementia, and agoraphobia” (p. 222). Although persons with practically any mental disorder engage in suicidal behaviors more often than individuals in the general population, the completed suicide risk is the highest among individuals diagnosed with affective disorders (major depression, bipolar disorder) and schizophrenia. The risk for nonfatal suicidal behaviors is significantly increased in case of depressive neuroses (dysthymic disorders) and personality disorders (especially Axis II diagnoses of borderline, antisocial, and narcissistic personality disorders). Many researchers and clinicians point out that comorbidity of mental disorders (e.g., panic disorder and affective disorder, schizophrenia and comorbid depressive disorder/substance abuse, co-occurrence of personality disorders and depression/schizophrenia) makes the risk of suicide even greater. The suicide risk factors in affective disorders and schizophrenia are discussed next.

Lifetime risk of suicide in affective disorders is 15%, and patients with these diagnoses constitute 50% to 70% of suicides. Among factors associated with an increased suicide risk in patients with the diagnosis of major depressive disorder are the severity of depression (the more severe clinical depression, the more acute suicide risk), increasing agitation and worsening melancholic symptoms, early course of illness before diagnosis and treatment, the recovery period and the period following hospitalization, as well as the co-occurrence of other psychiatric and substance abuse disorders.

Long-term suicide risk factors connected with the diagnosis of depressive disorders are high hopelessness, suicidal ideation, and previous suicide attempts. There is no consensus among researchers concerning increased suicide risk in delusional versus nondelusional depression (Jacobs, et al., 1999).

Significant risk factors for suicide in the manic-depressive illness include the increased severity of symptoms, family history of suicide, and history of patient’s previous suicide attempt. The suicide risk is raised early in the course of the illness, in mixed states (the combination of morbid depressive thoughts with high energy and agitation), in the depressive phase, during the recovery period, and following hospitalization (Jamison, 1999).

Individuals with the diagnosis of schizophrenia account for 10% to 15% of completed suicides, and the lifetime suicide risk in this population is 10%. The risk of suicide is greater in case of comorbidity (e.g., depressive disorder or substance abuse) and in paranoid schizophrenia with numerous positive symptoms, whereas it is lower in patients with negative (deficit) subtypes of the illness. The majority of researchers point out the fact that suicide in patients with the diagnosis of schizophrenia is most often related to the painful awareness of the deterioration of their abilities and the discrepancy between the future envisioned in the past before the onset of the illness and the likely degree of chronic and incurable disability in the future (especially in young males with good intellectual functioning and good premorbid school or work progress). Additionally, the suicide risk is heightened during early stages of the illness, in periods of clinical improvement after relapse, and during periods of hopelessness and depressed mood (Jacobs, et al., 1999).

The causal relationship between the diagnosis of a mental illness and suicidal behaviors is not clear. Tanney (1992) suggested several mechanisms, ranging from direct causes or consequences (e.g., command hallucinations and depressive delusions) and indirect complications (e.g., iatrogenic toxicity of medication and hopelessness of chronic disorder) to additive and releasing effects (e.g., alcohol abuse in depression leading to psychological disinhibition) and common etiology (e.g., isolation or loneliness may lead both to suicide and depression).

Numerous studies have found a strong relationship between suicidal behavior and substance abuse (Jacobs et al., 1999 and Lester, 1992). There is consistent evidence of elevated suicide rate among alcoholics. Of note, approximately 18% of alcoholics commit suicide, and 21% of all suicides involve patients diagnosed with a substance abuse disorder involving alcohol. The evidence for the association between alcohol abuse and nonfatal suicidal behavior is less consistent, although some studies have found a high incidence of suicide attempts (up to 24%) in alcoholics (Lester, 1992).

Several factors have been linked to suicide in alcohol abusers. The prominent risk factors are comorbid depression, communication of suicidal intent, continued drinking (suicide in alcoholics usually is related to late stages of the addiction), serious medical illness, unemployment, living alone, and poor social support, as well as a recent loss of an important interpersonal relationship.

The pathways between suicidal behavior and alcohol abuse are numerous and variegated. For example, substance abuse can be considered a “chronic suicide” indicator, and abused substances can be used as means of suicide (e.g., cocaine and heroine overdoses, a lethal concoction of alcohol and medications). Addictions and abuse often disrupt individuals’ interpersonal networks and their professional performance, which may cause isolation and social decline. Alcohol and drugs may lower restraints against suicide and impair judgment and increase impulsivity and risk taking, as well as increase an individual’s self-depreciation and depression. Some people planning suicide use alcohol and other drugs to achieve such mental state to “get the courage to die.” There is also a possibility that substance abuse and suicide stem from the same predisposing factors (e.g., personality disorder, mood dysregulation), and chronic alcohol abuse may directly change the brain neurochemistry through its impact on the serotonergic system. Paradoxically, in the early stages of abuse, when alcohol and other drugs are used as self-medication, they may lower the suicide risk in depression or replace direct suicidal behavior.

Much more research has been done on alcoholics than on drug abusers; however, the available data point out that individuals addicted to drugs have a higher incidence of suicide, nonfatal suicidal attempts, and suicidal ideation than do individuals in the general population. An increased suicide rate was found in narcotic and opioid addicts; for example, results of studies showed that up to 7% of cocaine abusers and 11% of methaqualone users died of suicide, and more drug abuse was noted in a sample of military trainees who attempted suicide than in the control group (Lester, 1992).

Medical disorders are associated with suicide in various ways (Harris and Barraclough, 1994 and Kelly et al., 1999). For example, some medical disorders may be caused by self-injury or substance abuse stemming from preexisting mental disorders, and a medical disorder and treatment (e.g., medication) may affect brain functioning, leading to personality disorders and mood disturbances. Additionally, disfigurement or disability caused by medical illness may result in mood dysregulation, and stigmatized diagnoses may contribute to social isolation and withdrawal of diseased individuals.

Numerous medical diagnoses have been related to increased suicide risk. For example, Harris and Barraclough (1994) noted that suicide risk increases with the following diagnoses (note that the number in parentheses means increased suicide risk in patients with the given diagnosis over the general population risk): HIV/AIDS (6.6), Huntington disease (2.9), malignant neoplasms (all sites, 1.8; head and neck, 11.4), multiple sclerosis (2.4), peptic ulcer (2.1), renal disease (hemodialysis, 14.5; transplantation, 3.8), spinal cord injuries (3.8), and systemic lupus erythematosus (4.3). Suicide risk is increased in epilepsy (fivefold) and in chronic pain syndrome and traumatic brain injury, which are associated with depression and suicidal ideation (Kelly, et al., 1999).

Suicidal thoughts range from harmless, transient fantasies that may help one to cope with life problems—as exemplified by Nietzsche’s famous words: “The thought of suicide is a great consolation; by means of it one gets successfully through many a bad night” (Nietzsche & Zimmern, 1989)—to recurrent suicidal ideation and concrete plans of self-destruction (Kerkhof and Arensman, 2001 and van Heeringen, 2001).

Adolescent population surveys show that considering suicide as an alternative problem solution is a rather normal and prevalent way of coping in this age group, and a study by Meehan et al. (1992) reported that 54% of college students have thought about suicide, including 26% of subjects who thought about it in the previous 12 months. Although still common, adult subjects report less suicidal ideation. Epidemiological studies have shown that the 1-year prevalence for suicidal thoughts ranges between 2.3% and 5.6%, and the lifetime prevalence is 13% to 15% (Kerkhof & Arensman, 2001).

Although suicidal ideation in the general population may be a frequent phenomenon, it should be remembered that suicidal ideation may evolve into a suicide plan, leading to self-destructive behavior that may result in death. A general population study by Kessler, Borges, and Walters (1999) showed that the transition from ideation to plan occurred in 34% of individuals who thought about suicide and further transitioned into an attempt in 72% (specifically, 26% of subjects proceeded from ideation to an impulsive attempt).

Therefore, an essential part of any suicide assessment procedure should be asking the interviewed individual about her or his suicidal ideation and plans. One of the most serious warning signs of high suicide risk, which calls for an immediate intervention, is a well thought-out and detailed suicide plan (including a place, time, and method), that is to be carried out in circumstances excluding the possibility of discovery and intervention by others. Any activities that show that an individual is preparing for death (e.g., writing a suicide note, making a will, giving away possessions) are other warning signs of suicide (Jacobs, et al., 1999).

Of course, a lack of a detailed suicide plan or denial of any suicidal ideation by a patient does not mean that there is no risk of suicide; in such cases, other means of assessment of suicide risk are recommended (e.g., the clinical judgment, risk assessment scales, and checklists).

Maris (1992) has stated that “any individual with a history of one or more prior nonfatal suicide attempts is at much greater risk for suicide than most of those who have never made a suicide attempt” (p. 11). Results of his psychological autopsy study showed that 30% to 40% of suicide completers had made at least one prior nonfatal suicide attempt, and about 15% of suicide attempters eventually died by suicide.

Several factors associated with risk of suicide after attempted suicide have been identified. These include older age (only females), male gender, unemployment or retirement, marital status (widowed, divorced, or separated), living alone, poor physical health, psychiatric disorder (especially depression, alcoholism, schizophrenia, and sociopathic personality disorder), high suicidal intent in current episode, violent method in current attempt, leaving a suicide note, and previous attempt(s) (Hawton, 2000).

A history of a suicide attempt is also correlated with the risk of repetition of nonfatal suicidal behavior. A classification of suicide attempters based on the history of repetition of their behavior has been proposed by Kreitman and Casey (1988): “first evers,” “minor repeaters” (lifetime history of two to four attempts), and “major/grand repeaters” (five attempts and more). A study conducted in Great Britain indicated that 48% of males with a history of nonfatal suicidal behaviors were first evers, 36% minor, and 16% grand attempters (for women, accordingly, 53%, 35%, and 12%). Factors that are associated with risk of repetition of attempted suicide have additionally been identified. These include a previous attempt, previous psychiatric treatment, personality disorder, substance abuse, unemployment, lower social class, criminal record, history of violence, age 25 to 54 years, and marital status (single, divorced, or separated) (Hawton, 2000).

Marzuk and colleagues (1997) have described a category of an “aborted suicide attempt” in which an individual has intent to kill himself or herself, changes his or her mind before making the attempt, and there is no physical injury. Their study on the prevalence of suicidal behavior among psychiatric inpatients showed that 46% of subjects made a suicide attempt, and 29% had a history of at least one aborted suicide attempt (which, in the case of almost one third of the patients, would be of high lethality, i.e., gunshot, jumping from heights). The authors concluded:

Although Marzuk and colleagues admitted that the relationship between aborted attempts, actual attempts, and suicidal ideation is not clear, they suggested that suicide risk assessment should include inquiries about aborted attempts and reasons for abandoning the suicidal behavior.

Although the acts of self-mutilation (defined by the lack of conscious suicidal intent) are usually distinguished from suicidal behaviors, they should also be considered as suicide risk factors. Research shows that more than 50% of self-mutilators attempt suicide by a drug overdose, usually as a result of demoralization related to their inability to control self-mutilating behaviors (Favazza & Simeon, 1995).

Although practically all methods used by individuals engaging in suicidal behaviors may lead to death or serious injuries, the statistical lethality of different means ranges from high to low risk of death (McIntosh, 1992). The likelihood of death is the highest in case of gunshot, carbon monoxide, hanging, drowning, suffocation by plastic bag, physical impact (jumping from heights, in front of a train, etc.), fire, poison, drugs, gas, and self-cutting. At least two factors contribute to the probability of death when using a particular method: the amount of time between the initiation of the suicidal act and death (e.g., drugs and poisons allow for the possibility of changing one’s mind and seeking help and allow for detection and intervention by others) and availability and effectiveness of medical intervention related to the method.

The choice of means of suicide depends on several factors: availability of the method and familiarity with it, suicidal intent and motivation of the individual (although there is no direct relationship between the medical lethality of the method and the desire to die, it is mediated by the attempter’s knowledge of the lethality of the method), and the cultural/ethnic factors (e.g., the gender socialization, social acceptability of suicide, the symbolic meanings of particular methods, and sites of suicide).

A specific suicide plan including an available and highly lethal means of suicide is a major risk factor of suicidal behavior and calls for a prompt and decisive intervention. Reducing access to means of suicide (e.g., legislations limiting access to firearms, careful dispensing of over-the-counter and prescription medications, detoxification or reduced toxicity of domestic gas and car exhaust, reduced access to high buildings, bridges, and legendary “suicide sites”) has been proved to reduce the incidence of suicide on both individual and population levels. It has to be kept in mind that many suicidal behaviors are impulsive and involve ambivalent attitudes toward life and death; in such cases, limited access to highly lethal methods increases the probability of survival and the effectiveness of medical intervention.

Roy (1992) observed that “suicide, like so much else in psychiatry, tends to run in families” (p. 578). Since the mid-1980s, results of numerous studies have led to the conclusion that there may be familiar or genetic determinants of suicidal behavior. Different lines of evidence point to this possibility (Roy, 1992). For example, clinical and follow-up studies (including Amish studies) show that individuals with a diagnosis of a mental illness (mostly affective disorders, especially depression) and a history of (fatal or nonfatal) suicidal behavior and affective disorder among the first- and second-degree relatives have increased risk of engaging in suicidal behavior themselves. These data are supported by results of twin studies showing the statistically significant higher incidence of suicide and psychiatric disorder in monozygotic pairs than among dizygotic twins.

Several explanations concerning the familial vulnerability to suicide have been offered. Genetic factors related to suicide may mostly represent a genetic predisposition to psychiatric problems associated with suicide. These include affective disorders, schizophrenia, and alcoholism, as well as the inability to control impulsivity. Besides, the mechanism of social modeling may play an important role: the family member who dies by suicide may serve as a role model, pointing to suicide as the best “solution” to life problems (Krysinska, 2003).

Biochemical studies of individuals with a diagnosis of depression, schizophrenia, and personality disorders show that a reduced metabolism of serotonin (5-TH; 5-hydroxytryptamine) and a lower concentration of its main metabolite, 5-hydroxyindoleacetic acid (5-HIAA), in the cerebrospinal fluid (CSF) are linked with disturb-ances in regulation of anxiety and inward- and outward-directed aggression (van Praag, 2001). Suicide attempters with a low CSF 5-HIAA level show an increased risk of repeated suicidal behaviors. On the basis of these results, van Praag concluded:

Although suicidal behaviors are present in all occupational groups and social classes, epidemiological data show that certain economic factors are correlated with high suicide risk (Stack, 2000). Poverty increases the risk of suicide through its association with financial stress, unemployment, fear of job loss, family instability, and mental problems (e.g., alcoholism, depression, crime victimization). Sociological studies have consistently found a negative correlation between socioeconomic status and suicide rates (for example, in 1985 the U.S. suicide rate for laborers was eight times higher than the overall national rate; Stack, 2000), although there are some high-status occupations with increased suicide risk (e.g., dentists, physicians, veterinarians) stemming from high job stress and easy access to lethal medication and other means of self-destruction. Another factor connected with the increased suicide risk among the disadvantaged economic groups is the relative deprivation related to the income gap between the rich and the poor, making the latter more frustrated and suicide prone.

Unemployment has often been mentioned as a major suicide risk factor, especially among men. Although the nature of the relationship between those two phenomena has not been fully explained, many correlational or causal pathways have been described. These explore how unemployment may heighten the suicide risk directly through eroding an individual’s income, economic welfare, and self-esteem, or it may affect dependent family members by lowering their financial capabilities. Psychologically disturbed persons may be at risk of both losing their jobs (and not being able to find another one) and being suicidal. High unemployment rates in the general population may lead to one’s fear of losing his or her job and may be related to smaller wages and underemployment.