Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

Self-directed violence encompasses a range of violent behaviors, including acts of fatal and nonfatal suicidal behavior, and nonsuicidal intentional self-harm (i.e., behaviors where the intention is not to kill oneself, as in self-mutilation). Suicidal behavior is a complex health problem with fatal and nonfatal components. Suicide, the fatal behavior, is according to the proverbial epidemiological paradigm is only the “tip of the iceberg” of self-directed violence. Suicide, nonfatal suicidal behaviors and suicidal ideations lead to considerable utilization of health care resources, specifically emergency department visits and hospitalizations, with variable outcomes from these encounters (Fig. 2.1). An understanding of the epidemiology of suicide and suicidal thoughts and behaviors, namely the pattern of their distribution in populations and the factors which could be attributed to their occurrence, provide an insight into those at risk for whom preventive efforts may be targeted. Concepts, sources of epidemiological data, and key findings are presented in this chapter.

Fig. 2.1
figure 1

Public health burden of suicidal behavior among adults aged ≥18 years—United States, 2008 (Source Crosby et al. 2011). *Source CDC’s National Vital Statistics System. All rates per 100,000 population. § Agency for Healthcare Research and Quality’s Healthcare Cost and Utilization Project--Nationwide Inpatient Sample. ¶ CDC’s National Electronic Injury Surveillance System--All Injury Program

Full size image

2 Suicide

Suicide is defined as the fatal event of self-directed violence. The death certificate is the source document for the identification of cases of suicide in epidemiologic studies and in vital statistics. Suicide, including accidental poisonings, on the death certificate in the “cause of death” field can be identified through the ICD-10 codes displayed in Table 2.1. In the UK, the definition of suicide includes deaths given an underlying cause of intentional self-harm or an injury/poisoning of undetermined intent. However, it cannot be applied to children due to the possibility that these deaths could have been attributed to neglect, abuse, or unverifiable accidents. The risk suicide is defined as the number of events of death from suicide in a defined population occurring at the same time the population was enumerated and multiplied by a factor of 10. Risk of suicide is represented as a rate measure when referring to the frequency of its occurrence in a general population, where:

Table 2.1 International classification of disease codes (ICD-10) for suicide—including accidental poisonings
Full size table

Rate = Number of events in a specified time interval/Average population during the same time interval × 10k

In 2010, the population of the U.S. was 308,746,000 and there were 38,364 deaths from suicide. Using the above formula and applying a power of 10 multiplier (10k) to convert the fraction into a rate (usually 100,000 for vital statistics), the suicide rate was 12.4 per 100,000 population. This means that for every 100,000 persons, there were about 12–13 suicides in this one year in the U.S.

2.1 Data Sources for Suicide Statistics

National vital statistics systems are the primary sources of country-specific suicide mortality statistics. These statistics are derived from a death certificate, the format and procedures for completing differ by regional laws and practices. The listing of a cause of death on a death certificate is influenced by many factors, including cultural factors, availability of an informant, provisions regarding confidentiality of the cause of death, and local laws governing forensic investigation. Death rates may also be affected by the age structure of a population and any temporal trends within age subgroups. In general, any suicide rate may be underestimated. For example, suicide rates generally exclude those aged <10 years (aged <15 years in the UK) because intent for self-harm is typically not attributed to young children. Thus, when interpreting suicide rates, it is important to keep in mind how these factors could influence the rate and use caution in interpreting rates across countries.

The World Health Organization (WHO) compiles suicide statistics by country. Data, search tools, and reports can be obtained through the Global Health Observation (GHO) at http://ww.who.int/GHO/. Data from the WHO reveal that suicide is a worldwide health problem with more than 800,000 people worldwide dying every year from suicide.

In the U.S., WISQARS TM (Web-based Injury Statistics Query and Reporting System) is an interactive database system that provides customized reports of suicide and other injury-related data. Suicide was the 10th leading cause of death among all ages and the leading cause of injury death in 2012. In Canada, CANSIM is a data base updated daily which provides the latest statistics for Canada, including health conditions. Suicide mortality data by age, gender, and year are available as downloadable tables.

In addition to recording death information in a vital records system, some countries and organizations have specialized surveillance systems or registries for obtaining more detailed information about a suicide event and its causes. In the U.S., surveillance data is obtained through the National Violent Death Reporting System (NVDRS). This system links information from death certificates, medical examiners, law enforcement, and forensic laboratories. Circumstances surrounding the suicide by age group, such as depressed mood, declining health, and disclosure of suicidal intent are compiled. The U.S. Department of Defense has a similar system and employs a more standardized and detailed collection of medical, military, and personal history information as a potential model for other organizations (Gahm et al. 2012). The National Suicide Registry in Malaysia also engages psychiatrics and staff members of psychiatric and mental health departments for the collection of the information on the demographics of the decedent, characteristics of the suicidal act, and risk factors (Hayati et al. 2008). The National Poison Data System is a near real-time surveillance database tracking poisonings and their sources in the U.S. Information is collected from calls b the public and health professionals to poison centers nationwide, consolidated, and evaluate Suspected suicide fatal (and non-fatal) cases are among the human exposure cases analyzed The system is maintained by the American Association of Poison Control Centers (www.aapcc.org) with annual reports published in the journal Clinical Toxicology.

2.2 Age, Sex, Race/Ethnicity, Geographical Region, and Other Population Subgroups

There is marked variation in suicide rates by age, sex, race/ethnicity, and geographic region.

By age group, among the highest suicide rates are noted to occur in midlife in Canada, the U.K., and the U.S. Suicide rates by age reveal a peak in adolescence in the U.S. (ONS 2014; Navaneelan 2012) (Fig. 2.2). In the U.S., the rates of suicide exceed homicide rates for all age groups except for the age groups of 18–24, where the homicide rate exceeds the suicide rate and for under age 10, where suicide rates are not reported (Fig. 2.2). In contrast, the highest suicides rates across European nations, the highest suicide rates are reported among people aged 65+ (21.9 per 100,000 population and those aged 45–59 (21.5 per 100,000 population) (WHO 2014). Higher rates have been reported among those aged 85 years and older in China ranging from 68.41 per 100,000 in large cities to 191.74 per 100,000 in rural areas (Simon et al. 2013).

Fig. 2.2
figure 2

Suicide and Homicide Rates,* by Age Group—United States, 2009. MMWR:July 20, 2012:61(28);543. * Per 100,000 population in age group. Suicides are coded as *U03, X60–X84, and Y87.0, and homicides are coded as *U01–*U02, X85–Y09, and Y87.1 according to the International Classification of Diseases, 10th Revision. † 95 % confidence interval. § Suicide data for persons aged 0–9 years are suppressed based on a child’s inability to form and understand suicidal intent and consequences

Full size image

The male predominance of suicide (as much as three times higher than females) is noted for all age groups across all nations with the exception in China, where overall it is higher among females (Table 2.2).

Table 2.2 Suicide rates per 100,000 by country, year and sex, for most recent year available as of 2011
Full size table

Suicide rates are two-times higher among whites than nonwhites. Between 1999 and 2010, the suicide rates among whites increased 20.4 % from 11.3 deaths per 100,000 population to 13.6 per 100,000 (Fig. 2.3) (CDC 2013, April 5, 2013/62(13);257). Ethnicity contributes to suicide risk when those groups are immigrants. Suicide occurs at a disproportionately higher rate among persons in higher income countries than among those in low- and middle-income countries with the exception of low- and middle-income countries of Europe and Southeast Asia (Table 2.2) (Norton and Kobusingye 2013).

Fig. 2.3
figure 3

Annual Age-Adjusted Death Rates*† for Suicide and Homicide, by Black or White Race—United States,§ 1999–2010. April 5, 2013/62(13);257. * Deaths are coded as *U03, X60–X84, and Y87.0 for suicide, and *U01–*U02, X85–Y09, and Y87.1 for homicide, as underlying causes of death, according to the International Classification of Diseases 10th Revision. Rates include deaths related to the events of September 11, 2001. † Rates have been revised by using populations enumerated as of April 1, for 2000 and 2010, and intercensal estimates as of July 1 for all other years. Therefore, the rates might differ from those published previously. § U.S. residents only

Full size image

Other population subgroups at risk for suicide include military personnel and veterans, young parents, and those with various chronic medical conditions (e.g., schizophrenia, certain dementias, bipolar disorder, chronic pain, post traumatic stress disorder, and traumatic brain injury) (Trofimovich et al. 2012; Fazel et al. 2014; Hooley et al. 2014; Niederkrotenthaler et al. 2012).

2.3 Agents of Suicide

Agents of suicide exhibit variation by characteristics of the population and their environmental circumstances. In the U.S., firearm suicides comprised 51 % (6.3 per 100,000 population) of the 38,364 suicide deaths in 2010. Of the drug-induced deaths, 13.1 % were suicidal drug poisoning. Although access to lethal agents are thought to be major risk factors, in nations where firearms are greatly restricted, elevated structures, substances, or other means may prevail. For example, suffocation has become the predominant method for committing suicide among adolescents in Canada, hanging, strangulation and suffocation among men in the U.K., and access to toxic substances, such as pesticides in rural China (Li et al. 2012; ONS 2014; Skinner and McFaull 2012).

Globally, European nations have among the highest suicide rates. The average suicide rate in Europe is 13.9 per 100,000, with the highest rates in the Commonwealth of Independent States (CIS) (21.4 per 100,000), followed by the new EU countries (13.8 per 100,000). Even within the European Union, where the overall average rate is 10.1 per 100,000 population, it rises as high as 30.7 per 100,000 population in Lithuania, 21.5 per 100,000 in Hungary, and 18.5 per 100,000 in Finland, and 18.4 per 100,000 in Slovenia (WHO 2014).

2.4 Risk Factors for Suicide

There are a number of risk factors identified for suicide. The most common ones cited in the literature are displayed in Table 2.3. There is no single risk factor on its own known to be causative, although depression is the most commonly identified from forensic autopsies and other studies of those who have committed suicide. Suicide typically results from a combination and interaction of a number of personal factors, including mental illness, marital difficulties, specifically divorce, financial hardship, deteriorating physical health, fearlessness about death, a major loss, as well as environmental factors such as lack of social support, economic downturns (Europe, Americas) and residence in rural areas (China) (Chang et al. 2013; Fassberg et al. 2012; Hooley et al. 2014; Hor and Taylor 2010; Li et al. 2012; Tang and Crane 2006). More recently, the social environment such as bullying and social internet media providing advice on aspects of suicide methods are playing a more prominent role as potential risk factors.

Table 2.3 Risk factors for completed suicide
Full size table

3 Suicidal Thoughts and Nonfatal Suicidal Behaviors

When obtaining information about suicidal thoughts, plans, or attempts, it is important to take note of the time interval in which the event occurred. Common time periods ascertained are lifetime, recent history (<6 months), baseline (within 30 days), or some prospective measurement (since baseline, since last visit, within one year). Historical measures are subject to recall bias. This means that respondents tend to recall events in the recent past better than those occurring more distantly. This could reflect why certain groups would have a lower prevalence of these behaviors if their event was 12 months or more prior to the survey compared with others, whose event was only one month prior to the survey. Persons in the former group might not report their event because of recall bias, and so the finding might not mean that their rates were actually lower. Prospective measurements are subject to survivorship bias meaning individuals who died from suicide or other means may not have lived and thus might be different from persons available for subsequent interviews.

3.1 Data Sources for Nonfatal Suicidal Events

Population based data on nonfatal suicidal events provide important background context for evaluating the impact of any programmatic or therapeutic intervention or change in public policy.

The World Mental Health (WMH) Survey Initiative is a source of information about suicidal behavior and various mental disorders, risk factors, and their treatment in the general. The WMH surveys were conducted 17 countries in Africa, the Americas, Asia, and the Pacific, Europe, and the Middle East (Nock et al. 2008).

The National Comorbidity Survey Replication (NCS-R) is a probability sample of the United States carried out a decade after the original 1992 NCS (NCS-1) was conducted. The NCS-R repeats many of the questions from the NCS-1 and also expands the questioning to include assessments based on the diagnostic criteria of the American Psychiatric Association as reported in the Diagnostic and Statistical Manual-IV (DSM-IV), 1994. The two major aims of the NCS-R were first, to investigate time trends and their correlates over the decade of the 1990s, and second, to expand the assessment in the baseline NCS-1 in order to address a number of important substantive and methodological issues that were raised by the NCS-1. The NCS-R is part of the Collaborative Psychiatric Epidemiology Surveys (CPES) data collection. Data and documentation for NCS-R can be accessed through the CPES Web site. The NCSR conducted the first national survey of US adolescents (NCS-Adolescent Suppl3ment) assessing a wide range of DSM-IV suicidal behaviors using structured diagnostic interviews (Nock et al. 2013).

The National Survey on Drug Use and Health (NSDUH) is an annual nationwide survey of a representative sample of the civilian, noninstitutionalized U.S. population aged ≥12 years. NSDUH collects data on adverse health consequences related to the use of illicit drugs, alcohol, and tobacco; initiation of substance use; substance use disorders and treatment; health care; and mental health. The suicide-related questions from NSDUH may be correlated with population demographics and the presence of a risk factor (e.g., major depressive episode), its duration and severity.

3.2 Age, Sex, Race, and Geographical Region Variations in Suicidal Thoughts and Nonfatal Suicidal Behaviors

There is marked variation in the prevalence of suicidal thoughts, plans, and attempts by age, sex, gender, and other demographic characteristics across nations and within nations (Casey et al. 2006; Crosby et al. 2011; Nock et al. 2008; Simon et al. 2013). Demographic, social, and access to health care could explain most of the cross-country variation, although there are some similarities in patterns. Results from the World Mental Health Surveys display the prevalences of these behaviors for individuals 18+ years are displayed in Table 2.4. Across all countries studied, the lifetime prevalence (standard error) of suicidal ideation, plans, and attempts is 9.2 % (0.1), 3.1 % (0.1), and 2.7 % (0.1). The prevalence of each of these varied widely by country. Although these prevalences appear small, when translated on a population basis, the extent of the impact becomes profound. For example, in the U.S., during 2008–2009, 0.5 % of the adult population reported having made a suicide attempt or an estimated 1 million (annual average) persons. Sixty percent of transitions from ideation to plan and attempt were estimated to occur within the first year after ideation onset in both adults and children (Nock et al. 2008, 2013). Further details on the prevalences of these according to more detailed population characteristics for the U.S. adults are listed in Table 2.5. By comparison, among adolescents in the U.S., the lifetime prevalences of suicide ideation, plans, and attempts were 12.1, 4.0, and 4.1 %, respectively, rates higher than in the overall adult population (Nock et al. 2013). Noteworthy in these statistics combined are that suicidal thoughts (or ideas) are not uncommon in the population ranging from 15.9 % in New Zealand to 3.0 % in Italy. Accordingly, plans are less frequently reported, ranging from 5.6 % in New Zealand to 0.9 % in China. The prevalence of attempts is of similar magnitude to prevalence of those making suicidal plans, ranging from 0.5 % in Italy to 5.0 % in the U.S. (Table 2.4).

Table 2.4 Lifetime prevalence of suicide-related outcomes in the world mental health surveys (Knock et al. 2008)
Full size table
Table 2.5 Annual average estimated number* and percentage of adults aged ≥18 years who had suicidal thoughts or engaged in suicidal behaviors during the previous year, by selected demographic characteristics—National Survey on Drug Use and Health, United States, 2008–2009†
Full size table

3.3 Risk and Protective Factors for Suicidal Thoughts and Behavior

There are a number of risk factors for nonfatal suicidal thoughts and behavior many of which are also associated with a fatal suicide (Table 2.6). Similarly, these also vary by age, gender, race, and ethnic group. Consistently among countries, the strongest association with suicidal thoughts and behaviors is depression, particularly severe depression, despite variability in the prevalence of this factor across nations (Casey et al. 2006; Nock et al. 2008). Other populations at risk for self-harm include those with dementia or psychiatric disorders, particularly schizophrenia. In these populations, the risk is further increased with the presence of comorbidites such as depression (Haw et al. 2009; Radomsky et al. 1999). An increased risk of suicide attempts after benzodiazepine and/or antidepressant use has been reported (Neutel and Patten 1997). Caution should be used in attributing prescription drug use and suicidal thoughts and behavior among persons with psychiatric behaviors as the association may be from a confounding by indication. Protective factors should also be considered as these are associated with a lower risk of suicidal behaviors. Both are listed in Table 2.6.

Table 2.6 Risk and protective factors for suicidal thoughts and nonfatal suicidal behaviors
Full size table

4 Conclusion

There are a number of challenges in assessing suicide risk and suicidal behavior in populations. Among these are changing environmental factors and their differential distribution among population subgroups. But uniform in concern is that suicide rates are trending upwards in many countries and population subgroups despite a recognition that suicide data may be under-reported due to the difficult nature of classifying suicide and the time lag in determining this as a cause of death. The epidemiological concepts underlying suicide and suicidal behavior illustrate a complex pathway to either fatal or nonfatal self-harm and present a challenge in developing preventive strategies. The factors that give rise to these events are highly subject to personal and environmental characteristics, whose influences are changed over time because of social, cultural, and economic conditions. Thus, the interpretations of any measures of such must consider to what degree biases enter into how their measurement and reporting could affect the underlying true rates.