Why People Die By Suicide
Page 21
6.69:1 in Belarus). China is clearly an exception to this rule, but in-
terestingly, so are many Asian countries, as indicated by the following
male:female ratios for suicide rates in Asian countries: 1.34 in India;
1.47 in the Philippines; 1.95 in Singapore; 2.27 in South Korea; 2.33
in Thailand; 2.59 in Japan; and 2.65 in Sri Lanka. The factors that
have evened out the male:female ratio of suicide rates in China may
be generalizable to other Asian countries as well.
Race and Ethnicity
A striking fact about U.S. suicide demographics is that African-
Americans are, in general, protected from suicide as compared to
Caucasians. Among African-American and Caucasian men, the sui-
cide rates per 100,000 are 9.8 and 19.1, respectively; corresponding
figures for women are 1.8 and 4.5.44 Theorists have often explained
this difference with regard to social support and religiosity, argu-
ing that African-Americans experience more social support and are
more religious, and as a function thereof, are protected from sui-
cide.45 On average, African-Americans do report more religiosity
than Caucasians.46 On measures like frequency of church attendance,
frequency of prayer, closeness to God, and self-ratings of spirituality,
African-Americans usually outpace Caucasians in the United States.
Also, there is evidence that religiosity can buffer against suicidality.47
In a national survey, it was found that African-Americans are more
likely to attend church, pray, and feel more strongly about their reli-
What Do We Mean by Suicide? ● 159
gious beliefs than whites.48 Regarding social support, there is little
doubt that its absence constitutes a suicide risk factor.49 However, it is
less clear that African-Americans enjoy more overall social support
than Caucasians, although there is some evidence to this effect.50 Im-
portant in context of the current model, the possibility that African-
American people in particular derive protection from suicide via
contact with religious institutions and perhaps support from others
is consistent with the assertion that the need to belong, when satis-
fied, can buffer from suicidality.
Another striking fact about suicide in African-Americans is the in-
creased rates of death by suicide among African-American men in
the last thirty years or so. This increase is accounted for mostly by the
rise in suicides among young African-American males. During the
1980s and 1990s, African-American boys aged ten to fourteen dem-
onstrated a 283 percent increase in suicide. A 165 percent increase
was observed for adolescents aged fifteen to nineteen. While the risk
for suicide increased among African-American men, the rate of sui-
cide decreased for African-American women (19 percent decrease).51
My former student Rheeda Walker, who now teaches at the Uni-
versity of South Carolina, proposed a model geared toward explain-
ing the increase in African-American male suicide. She focused on
the idea of acculturative stress, which is the stress that someone ex-
periences as they move from one cultural framework to another.
She predicted that African-American men in particular would expe-
rience high levels of acculturative stress, because over the last thirty
years or so, barriers to entry into mainstream education, employ-
ment, and the like have been reduced, perhaps more so for African-
American men than African-American women. Further, she ex-
pected that this stress would be a risk for suicidality, especially if
African-American men leave behind reliance on traditional suicide
buffers among African-Americans; namely, close ties to extended
family and to church. In her questionnaire study on 270 African-
160 ● WHY PEOPLE DIE BY SUICIDE
American young adults, she found empirical evidence to support her
expectations. Here again, this finding is consistent with this book’s
emphasis on failed belongingness as an important risk factor for se-
rious suicidal behavior.52
In the months before his death by suicide, my dad (a Caucasian)
occasionally attended African-American churches. I don’t know for
sure, but my suspicion is that feelings of failed belongingness were
overtaking him, and that usual sources of solace, like his own church,
were not adequate. I believe he reached out to African-American
churches because he sensed the closeness that many African-Ameri-
can congregations enjoy. That this did not work for him is not sur-
prising—my impression is that the churches were gracious to my
dad, but that he never felt like he really belonged. Changing from one
subculture to another and feeling a real part of the new subculture is
always difficult—as Walker’s study showed—and my dad was not up
to this challenge, tragically.
Hispanics in the United States have fairly low rates of suicide—
around 5 per 100,000, as compared to the national rate of approxi-
mately 10 per 100,000. Close contact with extended family is a
plausible explanation for this low rate. However, it should be noted
that Hispanics in the United States are a very diverse group, includ-
ing Mexican Americans, Cuban Americans, Puerto Rican Americans,
and others. People from Puerto Rico have higher rates of suicidal
ideation and suicide attempts as compared to Mexican Americans
and Cuban Americans.53 More research is needed on differences
among Hispanic groups in suicidality.
Just as it is important to differentiate among subgroups of His-
panics, it is necessary to couch findings on suicide in Native Ameri-
cans in the context of tribal differences. Overall, Native Americans
die by suicide at higher rates than other people in the United
States—about 1.5 to 2 times the rate, depending on the year. Just as
close family contacts appear to buffer African-American and His-
What Do We Mean by Suicide? ● 161
panic people from suicide, some have suggested that social disinte-
gration related to the plight of Native American people is a factor in
high rates of Native American suicide. Variations in social cohesion
may play an important role in varying tribal suicide rates. Native
American cultures of the Southwest, for instance, have a greater sense
of social cohesion as compared to Native Americans of the Northern
Plains; suicide rates are higher in the latter group, which is consistent
with the current emphasis on belongingness as key in explaining sui-
cidal behavior. Within individual southwestern tribes, social cohe-
sion may also be explanatory. Higher suicide rates have been found
in Apache as compared to Navajo or Pueblo people, and this differ-
ence may be attributable to higher tribal social integration among
the Navajo and Pueblo.54
Within various ethnic groups and cultures, perceived burden-
someness might be more or less painful than failed belongingness.
An important dimension on which cultures differ is the way their
members construe the self. There are cultures in which an inter-
dependent self-construal is normative (e.g., many As
ian cultures);
that is, people in these cultures see themselves as part of a larger
whole, and do not emphasize their own personal autonomy and
independence. People in cultures with autonomous self-construals
take the opposite stance. They prioritize personal agency, control,
and independence more than being part of a larger whole. The United
States is an example of a culture with relatively autonomous self-
construals.55 One might speculate that, in cultures in which an inter-
dependent self-construal is normative, failed belongingness may be
particularly painful, whereas in cultures in which an autonomous
self-construal is normative, perceived burdensomeness may be more
painful. To my knowledge, no previous work has examined this pos-
sibility with regard to suicidality—a potential direction for future re-
search.
Finally, it is interesting to note that pain tolerance appears to re-
162 ● WHY PEOPLE DIE BY SUICIDE
late to race. In a study of chronic pain patients, Caucasian patients
showed higher pain tolerance than African-American patients on a
variety of experimental and questionnaire measures.56 Other studies
have affirmed this result in more general population samples.57 In
Chapter 2, I asserted that increased pain tolerance may be implicated
in suicidal behavior. In this context, it is interesting that a relatively
high-risk group, Caucasians, have higher pain tolerance than a rela-
tively low-risk group, African-Americans.
Age
A very important factor in serious suicidality is the learned capability
to, in Voltaire’s words, “surmount the most powerful instinct of na-
ture.” The acquisition of this capability requires time and experi-
ence—it thus stands to reason that it would increase with age. Ac-
cording to the logic of my theory, if the acquired ability to enact
lethal self-injury increases with age, so then should suicide.
In fact, in the vast majority of countries and cultures, suicide does
increase with age. In the United States, suicide is most common
in those who are sixty-five years old or older,58 and this extends to
virtually all countries with reliable suicide rates.59 The ratio of at-
tempted to completed suicide among adolescents is quite high (more
than a hundred to one), whereas this ratio is around four to one
among older people.60 With age, then, suicidal behavior becomes
increasingly lethal. Of course, there may be associations with age
of the other two parts of the model as well—burdensomeness and
thwarted belongingness. In our study of psychotherapy patients at
the Florida State University Psychology Clinic, a measure of per-
ceived burdensomeness was correlated with age, such that older peo-
ple reported more perceived burdensomeness (the study did not in-
clude a measure of failed belongingness).
Although speculative, it seems plausible that over the last few de-
cades exposure to violence has become more common through such
What Do We Mean by Suicide? ● 163
means as violence in films, the media, and video games; weapon use;
and drug use. Exposure to violence in general has increased mark-
edly over the last fifty years. If it has become easier to acquire the ca-
pability for suicide, there should be some recent flattening of the age
curve, such that suicides occur, on average, somewhat earlier in life in
more recent cohorts. This does appear to be the case.61
It might be suggested that an important aspect of the current
model, burdensomeness, seems mostly applicable to older people,
and that despite the fact that suicide clearly increases with age, it
remains true that some young people die by suicide. Indeed, in 2000,
suicide was the third leading cause of death for young people in the
United States, whereas it was the eleventh leading cause of death
overall.62 Why would burdensomeness be applicable to adolescents?
In reply, the sense of being a burden is not limited to situations in
which one feels like a failed breadwinner. One can feel a burden at
any age, whether on one’s family or on society. This is made clear
by conceptual and empirical work on the “expendable child,”63
mentioned in Chapter 3. Feelings of expendability, which explicitly
include burdensomeness, are connected to suicidality in young peo-
ple.64
Suicidal behavior is rare in young children, in part because they
have not had the experiences and time to have acquired the ability to
seriously injure themselves. Though rare, suicidality is occasionally
observed in very young children. One study compared suicidal pre-
schoolers, ages 2.5 to 5, to preschoolers who were not suicidal but
who had serious psychiatric problems. The two groups of children
were matched on age, sex, ethnicity, parental marital status, and so-
cioeconomic status. The suicidal children differed from the compari-
son children in two relevant ways: they had higher pain tolerance, as
indicated by fewer displays of pain and crying on injury; and they
were more likely to be unwanted, abused, or neglected by parents.65
Few models of suicidal behavior would claim to be able to explain
164 ● WHY PEOPLE DIE BY SUICIDE
suicidality in a three-year-old, yet the model developed here is con-
sistent with data from preschoolers in that key aspects of the current
model differentiate suicidal from other psychiatric inpatient pre-
schoolers.
Depression is often viewed as a source for suicidality, with good
reason. Depression exacerbates feelings of burdensomeness and dis-
connection. However, the view that depression is the main source
for suicide does not square well with some epidemiological facts.
Specifically, depression is, if anything, a young person’s disease: the
average age of onset is around twenty,66 and rates of depression are
highest in young adults.67 On average, negative emotions are
higher—and positive emotions lower—in young than in old peo-
ple.68 By contrast, despite some mild flattening of the age curve, sui-
cide is much more a problem in older than in younger age groups. A
simplistic framework that views depression as the main source for
suicidal behavior has trouble grappling with these facts. By contrast,
the model proposed in this book is compatible with the association
between age and serious suicidal behavior.
To my knowledge, the only exception to the rule of increasingly se-
rious suicidal behavior with increasing age occurs in Native Ameri-
can people. The peak for death by suicide among Native Americans
is in young adulthood. Despite generally elevated rates of suicide
among Native American people, older Native Americans are actually
less likely to die by suicide than their U.S. Caucasian counterparts.
This pattern may be attributable to tendencies toward passive accep-
tance by elderly Native Americans and traditional respect of the aged
in Native American culture.69 This seems plausible in terms of ex-
plaining low rates of suicide among older Native Americans, but why
the very high
rates of suicide among younger Native Americans (es-
pecially young men)? Far too little research has been conducted on
this important question, but my prediction would be that young
adult Native Americans confront conditions that drain them of feel-
What Do We Mean by Suicide? ● 165
ings of belonging and effectiveness—experiences that older Native
Americans, relatively speaking, may be buffered from.
In the United States, older, white men are most at risk for suicide.
This stands to reason—on each dimension taken separately (age,
gender, ethnicity), these men are in the at-risk group each time. One
other factor is at play, I believe, and it is the tendency of this group in particular not to replenish their social connectedness as they age.
U.S. men in general and white men in particular seem to form some
close friendships in childhood and early and late adolescence, but the
forming of new and deep friendships in adulthood is relatively rare.
This is less the case for other groups (e.g., women in general; non-
white men). Older white men in the United States thus may be par-
ticularly prone to feelings of failed belongingness as they age and as
early friendships end for whatever reason; they are not buffered by
the replenishment of new adult friendships, at least not to the degree
of other groups.
This was clearly the situation for my dad at the time of his death.
He had close friendships in early adulthood, but as they faded or
failed for whatever reason, he did not form new ones. This was not
the case for another man I knew who died in his eighties from natu-
ral causes. At his memorial service, I was impressed to learn of his
constant social connectedness throughout his life. He made new sets
of friends every decade, it seemed. He was a pretty gruff character,
but he nonetheless had a habit of calling at least one friend a day just
to chat for a few minutes. He worked at initiating and maintaining
friendships, and this seemed to sustain him. If more older white men
did this—indeed if more people did this—I would predict a decrease
in the overall suicide rate.
The Clustering and “Contagion” of Suicide
As I mentioned in the Prologue, from time to time, completed sui-
cides cluster in space and time. For example, in a high school of ap-
166 ● WHY PEOPLE DIE BY SUICIDE