Content uploaded by Chris Brownson
Author content
All content in this area was uploaded by Chris Brownson on Apr 23, 2015
Content may be subject to copyright.
New Data on the Nature of Suicidal Crises in College Students:
Shifting the Paradigm
David J. Drum, Chris Brownson, Adryon Burton Denmark, and Shanna E. Smith
University of Texas at Austin
This article presents new data on the nature of suicidal crises in college students. Data were collected
from over 26,000 undergraduate and graduate students at 70 colleges and universities. An anonymous
Web-based survey was designed to provide insight into the full spectrum of suicidal thought, intent, and
action among college students. The authors discuss implications of these data and outline a new,
problem-focused paradigm for conceptualizing the problem of college student suicidality and for guiding
institutional policies and interventions at multiple points along the continuum of suicidal thoughts and
behaviors. The proposed paradigm encompasses and expands on the current model of treating individuals
in crisis in order to act preventively to reduce both prevalence and incidence of all forms of suicidality
among college students.
Keywords: suicide, college students, suicidal ideation, campus suicide prevention
Supplemental materials: http://dx.doi.org/10.1037/a0014465.supp
National attention to the problem of college student suicide has
been growing steadily over the past 25 years. However, recent
high-profile suicide-related events on college campuses, and the
ensuing media coverage of these events, have heightened concern
for mental health issues and campus safety. Campus mental health
professionals, along with administrators and policymakers, are
increasingly preoccupied with the formidable task of not only
decreasing suicide among the student population but also protect-
ing their institutions from liability. Unfortunately, most campus
counseling centers are facing an increasing demand for services
with no corresponding increase in resources. However, one posi-
tive result of the increased attention to college student suicide is
that federal legislation, such as the Garrett Lee Smith Memorial
Act of 2004, is being passed to fund campus suicide prevention
programs. It is also encouraging that recent attention to student
safety has prompted administrators to initiate campus dialogues
about student mental health issues. Unfortunately, lack of data on
the nature of students’ suicidal crises limits the effective imple-
mentation of available funds and inhibits dialogue from translating
into action.
In the absence of guiding knowledge, college mental health
practitioners have largely utilized the individual-focused paradigm
of suicide prevention and intervention that is common in commu-
nity settings. The success of this paradigm on college campuses
rests on the ability to detect students at risk for suicide, effectively
refer them to a professional, and have them meaningfully and
actively participate in treatment. For a number of students, this
treatment paradigm will be familiar, because many students have
had suicidal urges and ideation that may have required intervention
prior to attending college. Although the individual-focused para-
digm helped these students survive the earlier suicidal crisis, they
are once again on the suicidal pathway, and prior experience with
suicidality may have habituated them to suicidal ideation and
behaviors, thereby increasing future risk of completing suicide
(Joiner et al., 2005). The current study suggests a need to go
beyond exclusive reliance on this individual-focused model.
The individual-focused paradigm also fails to provide guidance
for the development and implementation of effective institutional
policies regarding students with mental health needs. Some col-
leges, seeking to protect themselves against lawsuits such as Shin
v. Massachusetts Institute of Technology (2005), are adopting
forced leave policies for students who admit to suicidal thoughts or
behaviors. At best, such policies fail to improve treatment for
students; at worst, they leave colleges and universities vulnerable
DAVID J. DRUM received his PhD in counseling psychology from the
American University. He is a professor of educational psychology at the
University of Texas at Austin. His research interests include college
student suicide and integrated health care.
CHRIS BROWNSON received his PhD in counseling psychology from the
University of Texas at Austin. He is director of the Counseling and Mental
Health Center at the University of Texas and director of the National
Research Consortium of Counseling Centers in Higher Education. His
research and practice interests are in suicide prevention and primary care
psychology.
ADRYON BURTON DENMARK received her BA in psychology from Stanford
University and is currently a doctoral candidate in counseling psychology
at the University of Texas at Austin. Her research interests include help-
seeking behaviors and suicidality among college students.
SHANNA E. SMITH received her PhD in Human Development and Family
Science at the University of Texas at Austin, with an emphasis on meth-
odology and statistics. She manages the statistical consulting center at the
University of Texas at Austin, where she focuses on the practical applica-
tion of advanced statistical techniques.
CORRESPONDENCE CONCERNING THIS ARTICLE should be addressed to David
J. Drum, University of Texas at Austin, Department of Educational Psy-
chology, Counseling Psychology/Counselor Education Program, 1 Univer-
sity Station, D5800, Austin, TX 78712-0383. E-mail: ddrum@mail
.utexas.edu
Professional Psychology: Research and Practice © 2009 American Psychological Association
2009, Vol. 40, No. 3, 213–222 0735-7028/09/$12.00 DOI: 10.1037/a0014465
213
to lawsuits claiming discrimination against students with disabil-
ities, such as Nott v. George Washington University (2005). It is
possible that such policies also deter suicidal students from seek-
ing help. Difficulties in developing appropriate institutional poli-
cies may be aggravated by the narrow scope of the individual-
focused paradigm of campus suicide prevention, with its exclusive
reliance on identifying and either treating or removing suicidal
students.
Currently, attempts by mental health professionals to prevent
suicides and to restore people to their premorbid condition are
based largely on research that has examined the role of various
factors in suicide attempts or completion. Because these are rela-
tively low incidence events, it is both inappropriate to focus
exclusively on factors associated with these outcomes and nearly
impossible to draw meaningful conclusions about helping suicidal
students (Haas, Hendin, & Mann, 2003; Silverman, 1993). Fur-
thermore, the restricted scope of this research contributes to a
disproportionate emphasis on assisting those who are in crisis
rather than focusing efforts on preventing people from entering
and progressing along a continuum of suicidality.
Episodes of suicidal thoughts and behaviors vary across many
factors, including duration, intensity, frequency, associated mood
states, consideration of methods, communication of distress, and
help-seeking behaviors. This study provides an in-depth examina-
tion of the entire suicidal continuum, from passive suicidal think-
ing to multiple attempts. The knowledge thus generated illumi-
nates a new paradigm for conceptualizing the problem of college
student suicidality. Our proposed paradigm, which is described in
detail in the discussion section of this article, is problem-focused
in that it defines suicidality in the population as the problem to be
addressed rather than focusing solely on suicidal individuals. Fur-
thermore, the problem-focused paradigm both encompasses and
expands on the current mode of treating individuals in crisis in
order to decrease both prevalence and incidence of all forms of
suicidality among college students.
Current Knowledge Regarding Campus Suicide
Suicide is the third leading cause of death for youths between
the ages of 15 and 24 years, following accidental injury and
homicide, and is believed to be the second leading cause of death
for college students because of the low rate of homicide in this
population (Centers for Disease Control and Prevention, 2007;
Suicide Prevention Resource Center, 2004). Although the rate of
completed suicides for college students is estimated at between 6.5
and 7.5 per 100,000, or approximately half that of the nonstudent
matched cohort, nearly all of this reduction in suicide completions
may be attributable to the reduced access to firearms on college
campuses (Schwartz, 2006a; Silverman, Meyer, Sloane, Raffel, &
Pratt, 1997). Additionally, the apparent decline in college suicide
rates that has been noted over the past 3 decades (Schwartz, 2006b)
correlates almost perfectly with the decreasing proportion of men
in college, as the suicide completion rate for male students is over
twice that of female students (Silverman et al., 1997). One might
therefore conclude that campus prevention efforts are either non-
existent or ineffective, especially because nearly 80% of those
students who die by suicide never participate in counseling ser-
vices (Gallagher, 2004; Kisch, Leino, & Silverman, 2005). How-
ever, for those students who do receive help from their college
counseling centers, these services are effective. An analysis of the
prevalence and effect of known risk factors among counseling
center clients suggests that these students would complete suicide
at a rate 18 times greater than the general student suicide rate,
when in fact the actual rate of suicide among counseling center
clients is only 3 times greater than in the nonclient student popu-
lation (Schwartz, 2006a).
Although the benefits of seeking counseling may be due, in part,
to the self-selection of those who seek help, efforts to raise
awareness about the mental health services available on campus
remain important. This is particularly important considering that
only 26% of college students are aware of their campus’s mental
health resources (Westefeld et al., 2005). Yet, even if it were
possible to increase the sensitivity and accuracy of existing referral
systems so that the majority of students at risk of committing
suicide were seen at their college counseling centers, meeting the
needs of these students would be extremely costly and could
require up to a 75% increase in counseling staff (Schwartz, 2006a).
An effective approach to suicide prevention cannot continue to
rely entirely on individual-focused counseling services. Further
reductions in suicide ideation, attempts, and completions must
derive from a thorough understanding of all aspects of the suicidal
spectrum and from use of that information to plan more robust
programs of prevention and intervention.
The National Research Consortium Survey of College
Student Suicidality
The National Research Consortium of Counseling Centers in
Higher Education, founded in 1991, brings together counseling
centers in the United States to study various college mental health
topics. This collaboration allowed for the implementation of the
current study, which is a large-scale survey designed to provide
comprehensive information about college student mental health
and to contribute uniquely to current knowledge of students’
suicidal experiences. To meet these aims, we developed a Web-
based survey with questionnaire items reviewed by the directors
from each campus counseling center in the consortium, as well as
by two experts in suicidology.
Sample Demographics
We selected a stratified random sample of 108,536 students
across 70 participating U.S. colleges and universities, using a
sampling strategy that allowed each campus to yield a sufficient
number of students from its own population to allow for individual
campus-level analyses. For campuses with more than 5,000 un-
dergraduates, 1,000 were randomly selected; for those with enroll-
ments between 500 and 4,999 undergraduates, 500 were randomly
selected; for those with enrollments of fewer than 500 undergrad-
uates, all students were selected. The same sample size guidelines
were used to select graduate students. The undergraduate and
graduate response rates were 24% (15,010 out of 62,000) and 25%
(11,441 out of 46,536), respectively, for a total sample of 26,451
students.
Table 1 presents the demographic makeup of our sample, which
is comparable to samples obtained by other large-scale national
surveys of college health issues, such as the spring 2006 National
College Health Assessment (American College Health Associa-
214 DRUM, BROWNSON, BURTON DENMARK, AND SMITH
tion, 2006) survey of almost 95,000 students. Because both our
study and the National College Health Assessment included only
4-year colleges and universities, it is likely that both samples
overrepresent White students compared with the broader popula-
tion of students in higher education settings in the United States. In
the current study, because students were randomly sampled within
each participating school, it was critical to include a representative
sample of schools. The size of the participating institutions ranged
from 820 students to 58,156 students, with a mean of 17,752
students. Of the institutions in our sample, 38% were private and
62% were public, which is representative of the proportion of
degrees awarded from public and private institutions nationwide
(U.S. Department of Education, National Center for Education
Statistics, Integrated Postsecondary Education Data System,
2004). Of the institutions, 94% included both graduate and under-
graduate students, with only four campuses reporting entirely
undergraduate student bodies. The institutions sampled represent
geographic diversity, with 20% of the schools located in the
Northeast, 20% in the West, 30% in the Midwest, and 30% in the
South.
Lifetime Prevalence of Mental Health Concerns, Suicidal
Ideation, and Attempts
To fully understand the context from which suicidality emerges,
we considered it necessary to inquire about students’ prior expe-
riences with psychological help, psychotropic medication, and
suicidal thoughts and behaviors. Among our sample, 44% of
undergraduates and 49% of graduate students reported that they
had sought mental health services at some point during their lives.
Percentages of lifetime psychological service use among smaller
samples of students have ranged from 35% to 42% (Deane &
Todd, 1996; Kahn & Williams, 2003), suggesting that students in
our sample were somewhat more likely to have used mental health
services. In addition, 19% of undergraduates and 21% of graduate
students in our study reported having received help from their
campus counseling center at some point during their college ca-
reer. Among our sample, 17% of undergraduates and 22% of
graduate students reported having taken medication for mental
health concerns at some point in their lives. These numbers may
reflect a trend of increasing use of psychotropic prescriptions for
children and adolescents, with the result that more students with
persistent and often severe psychological problems may now be
able to attend college (Benton, Robertson, Tseng, Newton, &
Benton, 2003; Rudd, 2004). In fact, 92% of counseling center
directors believe that students are increasingly arriving on college
campuses already taking psychiatric medication (Gallagher, 2006).
One important finding from students’ self-reports is that suicidal
thinking is far more common than has been previously recognized.
Over half of college students reported some form of suicidal
thinking in their lives (see Table 2). When asked whether they had
“ever seriously considered attempting suicide,” 18% of undergrad-
uates and 15% of graduate students endorsed this item. Among
those who had seriously considered attempting suicide, 47% of
undergraduates and 43% of graduate students had three or more
periods of this serious ideation, suggesting that by the time stu-
dents undergo suicidal crises in college, they are likely to have
significant previous experience with suicidality. Additionally, 8%
of undergraduates and 5% of graduate students reported having
attempted suicide at least once during their lives. Given that the
clearest and most consistent predictor for attempting or completing
suicide is having made previous attempts (Joiner et al., 2005;
Schwartz, 2006a), it is concerning that so many college students
have already made one or more attempts. Our findings make it
clear that suicidal ideation and attempts present a prevalent and
recurrent problem for the nation’s college students.
Suicidal Ideation Within the Past 12 Months
Querying students about whether they have seriously considered
attempting suicide in the past 12 months is the most commonly
used measure of college student suicidal ideation. Our findings
show that 6% of undergraduates and 4% of graduate students
reported that they “seriously considered attempting suicide” in the
past 12 months (see Table 3). The National College Health As-
Table 1
Sample Demographics: Sex, Race, Sexual Orientation, and Age
Demographic Undergraduates
Graduate
students
Sex (%)
Female 62 60
Male 38 40
Ethnicity (%)
Caucasian/White 79 72
Asian American 6 4
Hispanic American/Latino/a 5 5
Multiracial 4 4
African American/Black 4 4
International/foreign student 2 11
Alaska Native/American Indian ⬍1⬍1
Sexual orientation (%)
Bisexual 2 2
Gay/lesbian 2 3
Heterosexual 95 94
Questioning 1 1
Mean age (years) 22 30
Note. For undergraduates, n⫽15,010; for graduate students, n⫽11,441.
Table 2
Lifetime Experience of Suicidal Thoughts and Serious
Suicidal Ideation
Which phrase best describes you?
Undergraduates
(%)
Graduate
students (%)
I have never had suicidal thoughts 45 49
One period in my life of having
suicidal thoughts 24 22
A few discrete periods in my life of
having suicidal thoughts 23 23
Repeated episodes of suicidal thoughts
with periods in between of no
suicidal thoughts 6 4
Suicidal thoughts on a regular basis
for several years 2 2
I have seriously considered attempting
suicide 18 15
Note. For undergraduates, n⫽15,010; for graduate students, n⫽11,441.
215
COLLEGE STUDENT SUICIDALITY
sessment (American College Health Association, 2006) and the
National College Health Risk Behavior Survey (Centers for Dis-
ease Control and Prevention, 1995) each found that approximately
10% of college students endorsed this same item. The difference in
prevalence of suicidal ideation in the current study likely derives
from the fact that our survey explored a continuum of suicidality
rather than a single crisis point. When taking our survey, before
students answered whether they had seriously considering suicide
in the past 12 months, they responded to six prior items about their
various experiences of suicidal thought, including items such as “I
wish this all would just end” and “I wish I were dead.” Answering
multiple questions across several levels of severity prompted stu-
dents to think deeply about their history of suicidality and provided
them with the opportunity to precisely relate their experiences of
suicidal thought. Individuals who experienced low levels of sui-
cidal or presuicidal thinking were able to express this without
endorsing a single item regarding serious suicidal ideation.
Suicide Preparations, Methods, and Attempts
Most clinicians know that when assessing suicide risk they
should be more concerned for clients who report having a plan
than for those who report having suicidal thoughts but no specific
plan for attempting suicide. However, our results indicate that
thoughts of attempting suicide rarely occur without accompanying
thoughts about attempt methods. Among students who seriously
contemplated suicide in the past 12 months, 92% of undergradu-
ates and 90% of graduate students either considered some ways of
killing themselves or had a specific plan (see Table 4). It may be
that students are more likely to acknowledge having considered
specific methods in an anonymous situation than when they are in
the room with a counselor. The most common method considered
by suicide ideators was a drug or alcohol overdose, which was the
primary method considered by 51% of undergraduates and 37% of
graduate students who had a specific plan. Further information
about considered methods is available with the supplemental ma-
terials.
Among those who seriously considered suicide in the past 12
months, 37% of undergraduates and 28% of graduate students
made some preparations for killing themselves, such as gathering
materials, writing a suicide note, doing a practice run, or beginning
an attempt and then changing their mind (see Table 4). To assess
the extent to which preparatory activities predict suicide attempts,
we performed a hierarchical linear model, an extension of multiple
regression that can appropriately handle the sampling framework
and noncontinuous outcomes used in this study. The hierarchical
linear model containing gender, student status, and each of the four
preparatory activities revealed that both gathering material for an
attempt, odds ratio (OR)⫽2.57, t(1243) ⫽4.13, p⬍.001, and
beginning the attempt but changing one’s mind, OR ⫽17.67,
t(1243) ⫽14.06, p⬍.001, significantly increased the odds of
making a suicide attempt. Ultimately, 14% of undergraduates and
8% of graduate students who seriously considered suicide in the
past 12 months actually attempted suicide. These numbers repre-
sent 0.85% of our total undergraduate sample and 0.30% of our
total graduate student sample. The majority of these attempters
(69% of undergraduates and 75% of graduate students) made only
one attempt, with only 9% of undergraduate and graduate student
attempters reporting three or more attempts.
Among those who attempted suicide, 19% of undergraduates
and 28% of graduate students made an attempt that required
medical attention. The most common attempt method was a drug
overdose, which was used by 51% of undergraduate and 50% of
graduate student attempters. Additional information about meth-
ods used is available online with the supplemental materials.
Perhaps the most distressing finding was that 23% of undergrad-
uates and 27% of graduate students who attempted suicide within
the past 12 months reported that they were currently considering
making another suicide attempt. These results again underscore the
highly recurrent nature of this problem, which demands an ex-
panded intervention approach, both to prevent students from orig-
inally entering the repeating cycle of suicidality and to more
effectively prevent relapse.
Characteristics of Serious Suicidal Ideation
Although there is great value in accurately describing the prev-
alence of serious suicidal ideation and attempts, we were most
interested in gaining unique insight into the subjective experience
of being suicidal. We believe that enriched understanding of this
experience holds many keys, both for defining a new and more
useful paradigm for comprehensive suicide prevention, and for
enhancing clinicians’ knowledge of and treatment approaches to
helping suicidal clients. We therefore queried students regarding
Table 3
Suicidality in the Past 12 Months
In the past 12 months . . .
Undergraduates
(%)
Graduate
students (%)
I thought, “I wish this all would just
end” 37 30
I thought, “I wish I was dead” 11 8
I have seriously considered attempting
suicide 6 4
I have attempted suicide 0.85 0.30
Note. For undergraduates, n⫽15,010; for graduate students, n⫽11,441.
Table 4
Suicidal Plans and Preparations
Answered by those who have
seriously considered attempting
suicide in past 12 months
Undergraduates
(%)
Graduate
students (%)
Plans
Never considered how to attempt 8 10
Thought about some ways, but
not seriously 54 55
Had a specific plan 38 35
Preparations
Gathered material to kill self 19 15
Began to attempt, then
reconsidered 17 10
Wrote suicide note 14 7
Did practice run of suicide
attempt 5 4
Note. For undergraduates, n⫽910; for graduate students, n⫽411.
216 DRUM, BROWNSON, BURTON DENMARK, AND SMITH
features of their periods of suicidal thinking, such as frequency,
intensity, and duration.
The majority of students who seriously considered suicide in the
past 12 months experienced these suicidal thoughts as recurrent,
brief, and intense events. In our sample, 69% of undergraduates
and 63% of graduate students reported having more than one
period in the past 12 months during which they considered at-
tempting suicide. This suggests that serious suicidal ideation is
experienced recurrently even within a relatively short span of time.
Surprisingly, 56% of undergraduates and 58% of graduate students
reported that, on average, their periods of serious suicidal ideation
lasted for 1 day or less. Only 16% of both undergraduates and
graduate students had periods of seriously considering a suicide
attempt for many weeks or more (see Table 5). Despite the relative
brevity of many of these periods of suicidal thought, 50% of
undergraduates and 45% of graduate students reported that their
thoughts of attempting were strong or very strong, and 45% of
undergraduates and 39% of graduate students reported that their
thoughts of suicide greatly interfered with their academic perfor-
mance.
Clinicians who work with suicidal clients benefit from a height-
ened awareness of common patterns of ideation, such as brief but
intense periods of serious suicidal thinking. These findings also
have implications for the traditional model of campus suicide
prevention, which depends on training faculty and staff to look for
behavioral warning signs of suicidal ideation. If, for over half of
suicidal students, a typical period of suicidal thinking lasts less
than 1 day, it is improbable that a referral agent would have the
opportunity to recognize a student in crisis. Also, if the student
were identified, in many cases the suicidal ideation would have
passed by the time he or she could seek mental health treatment.
Contributing Factors and Associated Mood States
Another unique aspect of this survey was the ability to query
students about their perception of the impact of various experi-
ences on their suicidal thoughts in the past 12 months. For both
undergraduate and graduate students, the three factors rated by
most students as having a large or very large contribution to their
ideation were wanting relief from emotional or physical pain,
problems with romantic relationships, and the desire to end one’s
life (see Table 6). An important implication of these findings is
that the incidence of suicidal ideation may be greatly reduced by a
population-oriented preventive approach that seeks to improve
social support networks and engineer a more connected, caring
environment. Problems with academics, problems with friends,
and family problems were also commonly rated as having a large
impact. In particular, university administrators should note that
academic problems were rated as having a large effect on suicidal
ideation by 43% of undergraduates and 45% of graduate students
who seriously considered attempting suicide. This suggests that the
systems that become involved with academically distressed stu-
dents may have an opportunity to interact with these students in a
way that is sensitive to and may decrease their likelihood of
developing suicidal thinking.
Clinicians working with suicidal students should be conscious
of the potential impact of occurrences such as breakups and
academic struggles. These issues are relatively common among
counseling center clients and yet should not be treated lightly. It
should also be noted that less common events, such as interper-
sonal violence or drug and alcohol problems, might nonetheless
contribute profoundly to the development of suicidal ideation.
Population-based interventions aimed at decreasing the occurrence
of these associated factors may effectively reduce the incidence of
suicidal ideation among college students.
Students were also asked about their emotional experiences
during their periods of suicidal ideation. Not surprisingly, sadness,
loneliness, and hopelessness were the most frequently endorsed
moods during students’ typical periods of suicidal ideation. When
looking at all of the emotions together as predictors of placement
on the suicidal continuum, some interesting trends emerge. Hier-
archical linear models including the emotions anger, anxiety, sad-
ness, guilt, loneliness, and hopelessness/helplessness were used to
predict typical length of ideation, strength of suicidal intent, and
whether the person attempted suicide. Those who felt hopeless/
helpless, b⫽0.22, t(1182) ⫽6.37, p⬍.001; angry, b⫽0.07,
t(1182) ⫽3.48, p⬍.001; or sad, b⫽0.08, t(1182) ⫽2.29, p⬍
.05; had stronger suicidal intent. However, only a sense of hope-
lessness/helplessness predicted longer periods of ideation, OR ⫽
1.26, t(1183) ⫽⫺4.27, p⬍.001, and greater likelihood of making
a suicide attempt, OR ⫽1.44, t(1194) ⫽3.21, p⬍.01.
These findings underscore the complex nature of suicide risk
assessment. Feelings of hopelessness/helplessness are associated
with increased risk for attempting suicide, but these feelings may
be difficult for referral agents to identify. Even highly trained
professionals have difficulty accurately assessing the severity of
suicide risk in their clients (Bryan & Rudd, 2006). Wingate, Joiner,
Walker, Rudd, and Jobes (2004) found that clinicians tend to both
overestimate the severity of clients’ suicidal thoughts and under-
estimate the severity of risk that is associated with preparing for a
suicide attempt. If trained professionals have difficulty accurately
assessing suicide risk, then it follows that the laypeople central to
campus gatekeeper training efforts will struggle.
Table 5
Frequency and Duration of Suicidal Ideation Within the
Past 12 Months
Answered by those who have
seriously considered attempting
suicide in past 12 months
Undergraduates
(%)
Graduate
students (%)
No. of periods (in the past 12 months)
of suicidal ideation
a
13137
22625
31612
4–6 14 12
7 or more 11 10
Average length of these periods of
suicidal ideation
1 hr or less 31 32
Several hours/day–1 day 25 26
Many days–1 week 28 26
Many weeks–1 month 11 11
Many months or more 5 5
Note. For undergraduates, n⫽910; for graduate students, n⫽411.
a
Response error accounted for 2% of undergraduate and 4% of graduate
student responses.
217
COLLEGE STUDENT SUICIDALITY
Communicating Distress, Help Seeking,
and Protective Factors
Many campus suicide prevention efforts focus on gatekeeper
training, such as the popular Question, Persuade, and Refer pro-
gram, which has shown promise in improving the ability of those
with frequent student contact to notice warning signs and respond
appropriately to students who exhibit emotional distress and may
be contemplating suicide (Wyman et al., 2008). To better under-
stand student patterns of informal help seeking, we asked students
who seriously considered attempting suicide in the past 12 months
if they told anyone about their suicidal thoughts. We found that
46% of undergraduate and 47% graduate students chose not to tell
anyone about their suicidal thoughts. Of interest, after controlling
for student status, gender, and intensity of suicidal thought, a
hierarchical linear model found no relationship between disclosing
suicidal ideation and actually making an attempt. However, 52%
of students who confided in other people reported that telling the
first person was helpful or very helpful in dealing with the suicidal
thoughts. Two thirds of those who disclosed their suicidal ideation
first chose to tell a peer, such as a romantic partner, roommate, or
friend. Further information about choice of confidant is available
online with the supplemental materials. Almost no undergraduates
and not a single graduate student confided in a professor. There-
fore, for gatekeeper training efforts to be more effective, they need
to focus on training students in how to respond to their peers.
Although it is logistically more difficult to train students, who may
be unlikely to voluntarily engage in a training process, it is likely
a more efficient use of limited resources from a population-based
perspective. Not only are students more likely to be chosen as
confidants by peers who experience suicidal ideation, but they also
stand to benefit personally from the increased awareness of mental
health issues that will be facilitated through training, thus resulting
in a deeper and more enduring impact on campus suicide preven-
tion.
Students’ reasons for not telling others about their suicidal
ideation provide valuable insight into the way students perceive
both the seriousness of their suicidal thoughts and the potential
consequences of sharing these thoughts with others. Several prom-
inent themes that emerged from this inquiry were fear of being
stigmatized or judged, not wanting to burden others, knowledge
that the problem was transitory, not having anyone to tell, and fear
of consequences such as expulsion from school or forced hospi-
talization. It is important for university policymakers to note that
wariness of institutional policies like mandatory treatment or med-
ical leave prevents some students from disclosing their distress.
These results also have implications for population-based inter-
ventions aimed at decreasing secrecy and stigma around mental
health issues and increasing students’ social support networks.
Such interventions are likely to both reduce the chance that stu-
dents will initially develop suicidal ideation and increase the
probability that those students who do experience suicidal ideation
will confide in others.
Among students who disclosed their suicidal ideation to others,
only 58% of undergraduates and 50% of graduate students were
advised by the first person they told to seek professional help. A
hierarchical linear model revealed that a student is more likely to
be advised to seek help for suicidal thoughts if the student is
female, OR ⫽1.40, t(546) ⫽1.99, p⬍.05; if the suicidal thoughts
are stronger, OR ⫽1.34, t(546) ⫽2.66, p⬍.01; if the thoughts
create greater interference with academics, OR ⫽1.20, t(546) ⫽
2.81, p⬍.01; or if the student first confides in a family member,
OR ⫽1.81, t(546) ⫽2.48, p⬍.01. Surprisingly, those with
stronger intent to kill themselves were marginally less likely to be
advised to seek professional help, OR ⫽0.84, t(546) ⫽⫺1.69,
p⫽.09. This finding is alarming, because self-reported intent to
kill oneself is highly correlated with actually attempting suicide.
The disconnect between the student’s intent to commit suicide and
the confidant’s perception of the seriousness of the situation may
result from the greater detachment that accompanies serious intent
to kill oneself. It may also indicate a tendency to conceal the
intensity of distress to minimize the likelihood that others would
interfere with a suicide attempt.
Regardless of the precise reasons for the mismatch between
suicidal intent and likelihood of being advised to seek psycholog-
ical services, this disconnect has clear implications for the utility of
the individual-focused paradigm of campus suicide prevention.
Our results indicate that those most likely to serve as referral
agents are less likely to refer higher risk students to treatment.
Increasing the precision of referrals would require considerable
investment of resources, because laypeople would need to be
trained to the level of paraprofessionals. An important supplemen-
tary use of resources would be to create a more connected and
caring campus environment so that fewer students initially enter
the suicidal process.
In addition to understanding students’ patterns of communicat-
ing distress, we hoped to learn how suicidal students experience
the process of seeking professional help. Less than half of students
who seriously considered attempting suicide received any profes-
sional help, and about one third of those who reported receiving
psychological services were already receiving such services before
the onset of their suicidal ideation. Further details about students’
timelines for seeking help are available online with the supple-
mental materials. A hierarchical linear model showed that graduate
students, OR ⫽1.38, t(1263) ⫽2.77, p⬍.01; female students,
OR ⫽1.44, t(1263) ⫽3.51, p⬍.001; those with more intense
thoughts, OR ⫽1.33, t(1263) ⫽3.78, p⬍.001; and those whose
thoughts had a greater impact on their academic performance, OR ⫽
Table 6
Events Rated as Having a Large Impact on Seriously
Considering Suicide in the Past 12 Months
Answered by those who had seriously
considered attempting suicide in past
12 months
Undergraduates
(%)
Graduate
students
(%)
Emotional/physical pain 65 65
Romantic relationship problems 59 53
Impact of wanting to end my life 49 47
School problems 43 45
Friend problems 43 28
Family problems 42 34
Financial problems 31 34
Showing others the extent of my pain 30 27
Punishing others 14 8
Alcohol/drug problems 10 6
Sexual assault 8 6
Relationship violence 6 6
Note. For undergraduates, n⫽910; for graduate students, n⫽411.
218 DRUM, BROWNSON, BURTON DENMARK, AND SMITH
1.05, t(1263) ⫽2.26, p⬍.05; were significantly more likely to seek
professional help. Over two thirds of students who sought help saw
counselors, and over one third of students saw psychiatrists, with
some students receiving help from multiple sources. Among stu-
dents who saw counselors, 53% of undergraduates and 60% of
graduate students rated the counselor as helpful or very helpful in
preventing suicide. Among students who saw a psychiatrist, 39%
of undergraduates and 44% of graduate students rated that person
as helpful or very helpful. These helpfulness ratings were nearly
identical for other medical providers. After controlling for gender,
student status, strength of suicidal thoughts, and strength of intent
to kill oneself, students who sought professional help were less
likely to attempt suicide, OR ⫽0.58, t(1242) ⫽⫺3.20, p⬍.01,
than those who did not seek help.
We asked students who seriously considered attempting suicide
but did not ultimately make an attempt within the past 12 months
to rate the importance of various factors in preventing a suicide
attempt. Among both undergraduate and graduate students 77%
said that disappointing or hurting their family had a large or very
large impact on their decision to not attempt suicide, whereas 56%
and 49%, respectively, said the same of disappointing or hurting
their friends (see Table 7). Knowledge of protective factors may
assist clinicians who are working to reduce a client’s risk for
attempting suicide. In addition, 40% of undergraduates and 35% of
graduate students reported that wanting to finish school was an
important reason for not attempting suicide, implying that judicial
or involuntary withdrawal procedures for suicidal students may
remove an important protective factor.
Prior research suggests that aspects of campus life that increase
students’ sense of belonging to a caring social network, such as
sports participation, are associated with decreased suicidal behav-
ior (Brown & Blanton, 2002). After controlling for gender and
student status, we found that students who participated as either
leaders, OR ⫽0.71, t(25854) ⫽⫺5.16, p⬍.001, or members,
OR ⫽0.75, t(25854) ⫽⫺3.84, p⬍.001, of student organizations
were less likely than those who did not participate in organizations
to have seriously considered attempting suicide in the past 12
months. These results underscore the importance of creating and
maintaining an inclusive campus environment and providing op-
portunities for students to connect to others as a possible means for
decreasing the numbers of students who enter the suicide contin-
uum.
Implications for Developing a Problem-Centered
Paradigm
This study has contributed to the understanding of college
student suicide in several important ways. First, it provides a view
into the mind of suicidal individuals, from initial thoughts of
suicide through multiple attempts. Second, it shows that suicidal
ideation can be a recurrent phenomenon: Those who have one
episode of suicidal ideation or urge are reasonably likely to have
more. Third, it suggests that periods of suicidal ideation for some
people are quite brief, yet ideation seldom occurs in the absence of
some form of contemplation or planning for how one would act on
the suicidal urge. Fourth, it manifests the profile of a spectrum of
suicidality in a large population of college students as clusters of
ideation, contemplation, and planning as well as attempts, multiple
attempts, and completions. Fifth, it highlights some limitations of
the prevailing individual-focused paradigm for suicide prevention
and intervention.
Our study does not discredit the prevailing paradigm; in fact, it
fortifies the knowledge needed to increase the success of interven-
tions characteristic of this paradigm. For example, clinicians are
likely to benefit from an enhanced understanding of the variable
risk associated with different subjective emotional states, common
patterns of ideation, students’ perceptions of the impact of various
risk-conferring and protective factors, and students’ experiences of
seeking professional help. Increasing the efficacy of crisis inter-
ventions through this new knowledge is particularly beneficial to
the private practitioner whose responsibility is to the patient ac-
cepted into treatment and not to the health status of a larger
community. These interventions are also appropriate to the college
mental health practitioner when treating a client in suicidal crisis.
However, college campuses have the additional responsibility of
not only protecting the student in suicidal crisis, but also consid-
ering the public health goals of reducing the incidence of suicid-
ality and enhancing the health and well-being of the larger popu-
lation of students.
A difficulty of the prevailing paradigm is that it blinds and
binds. To use the well-worn analogy of an iceberg, it blinds college
mental health practitioners’ vision to the submerged behaviors that
are part of a spectrum of expression of suicidality and culminate in
the crises that lead to eventual death by suicide (Joiner et al.,
2005). Our contention is that when suicidality is viewed from the
perspective of a population of individuals, there is a discernable
profile of experience and expression of symptoms of suicidality in
that population. Those symptoms can then be ordered into a
progression reflecting increasing intensity of suicidality. Ideally, to
maximize the opportunity to reduce incidence and prevalence of
suicidality, each symptom cluster would be examined to identify
factors contributing to elaboration of the symptom. Once contrib-
uting factors are known, then interventions precisely tailored to
that factor can be applied.
The individual-focused paradigm, rather than suffusing inter-
vention efforts to the entire spectrum of symptoms and their
related contributing factors, is place specific on the spectrum; that
is, it microfocuses on the individual in crisis. In doing so, this
paradigm binds intervention behavior to that single point on the
Table 7
Important Factors in Preventing a Suicide Attempt
Answered by ideators who had not
attempted suicide in the past 12 months
Undergraduates
(%)
Graduate
students
(%)
Disappointing/hurting my family 77 77
Disappointing/hurting my friends 56 49
Hope/plans for the future 42 35
Wanting to finish school 39 32
Support of my friends 38 28
Support of my family 35 33
Disappointing/hurting my partner 34 46
Religious/moral beliefs 34 33
Support of my partner 26 28
My pet(s) 19 20
Relationship with mental health professional 10 14
Note. For undergraduates, n⫽761; for graduate students, n⫽370.
219
COLLEGE STUDENT SUICIDALITY
profile and contributes to college mental health practitioners’
failure to capitalize on countless opportunities both to prevent
development of the symptom and to reduce forces routinely con-
tributing to life-threatening intensification of suicidal intent. The
result is an overallocation of resources to crisis intervention efforts
at the expense of other forms of intervention. Additionally, by
focusing solely on the individual in crisis, the current paradigm
obscures the reality of how common a phenomenon it is for
students both prior to and during the college years to engage in one
or several behaviors considered part of a continuum of suicidality,
which, with repeat episodes, contributes to lowering the threshold
for acting on suicidal urge.
To enable the problem of college student suicidality to be seen
and addressed in its fullness, it is necessary to step back from
microfocusing on just those students experiencing a suicidal crisis
to see and address the entire spectrum of suicide-related thoughts
and expression. Practitioners must do the equivalent of raising the
suicide iceberg to reveal its submerged elements, magnitude, and
internal dynamics, and they must learn how the milieu in which
this iceberg developed has shaped and influenced its features. The
knowledge gleaned from this more complete view of suicidality in
college students must be matched with a more complete interven-
tion paradigm. Such a paradigm must capitalize on the strengths of
the current model but must also encase it in a broader framework
capable of reducing the incidence of suicidality by preventing
occurrences of its most mild and often transient manifestations all
the way through posttreatment interventions to reduce relapse
through recovery strengthening programs. Our proposed problem-
focused paradigm incorporates some of the concepts and interven-
tions espoused by community psychologists as well as by public
health specialists engaged in primary prevention.
A problem-focused paradigm requires the entire campus com-
munity to share responsibility for reducing student suicidality.
Rather than focusing on the suicidal student as the institution’s
problem, the new paradigm defines the problem as how to reduce
suicidality in all its forms of expression among the entire popula-
tion of students. This shared campus responsibility broadens the
personnel and resources available to reduce the percentage of
students in a given institution who engage in suicidal thinking,
contemplating how to attempt suicide, writing suicide notes, doing
practice runs, attempting suicide, engaging in multiple attempts,
and relapsing postrecovery. The problem-focused paradigm calls
for the activation of a cadre of stakeholders (administrators, stu-
dent leaders, advisors, faculty, parents, counselors, student affairs
specialists, and other relevant constituencies) to engage in
problem-solving activities. Furthermore, it illustrates how the cur-
rent intervention focus must be diffused to include primary pre-
vention, proactive assistance, early intervention, and recovery
maintenance programs and how it must continue to improve the
effectiveness of helping students resolve their suicidal crises.
Primary prevention efforts, which are designed to precede the
origination of an observable need, can be understood as having
four main foci: (a) to refashion the environment so that it is both
more supportive and more protective, (b) to increase awareness
and promote help seeking through the dissemination of educational
materials and self-assessments, (c) to reduce the incidence of
traumatic negative life events, and (d) to increase the available
sources of internal resilience among the population. For example,
academic programs such as living-learning environments and
freshman interest groups foster the development of supportive
relationships. Web-based mental health screeners may engage
students in thinking about their personal distress levels and ways
of coping, whereas information and advocacy campaigns can raise
awareness and reduce the frequency of highly damaging interper-
sonal events, such as sexual assault and relationship violence, in
the lives of college students. Furthermore, the protective benefits
of campus prohibitions against firearms suggest that other methods
of environmental engineering, such as restricting access to campus
rooftops and securing potentially lethal laboratory chemicals, will
save the lives of many unidentified individuals.
The rationale behind this population-level focus is that the same
expenditure of resources results in greater improvements in health
status than the exclusive targeting of at-risk individuals. Frohlich
and Potvin (1999) explained,
when many people lower their risk, even a little, the total benefit for
the population is larger than if few people at high risk experience a
large risk reduction. This is consistent with the notion that groups of
individuals function collectively and, as such, are affected by the
average functioning of the individuals around them. (pp. 213–214)
By shifting the entire distribution of individuals to a lower risk
level, not only are those at high risk being shifted to a lower risk
status, but the overall prevalence of suicidality in the population is
also decreasing and the overall population health is increasing.
The potential benefits of preventive policies are clear, but en-
acting such policies will require universities to reevaluate many
existing beliefs that contribute to a dual-role conflict regarding
students with emotional or behavioral problems. Institutional atti-
tudes and policies may promote detachment from those students
who struggle academically or express extreme emotional distress.
To increase the health and well-being of the student population, it
is necessary to enhance the supportive aspects of the university
environment so that the institution engages with, rather than de-
taches from, students in distress, thereby communicating to all
students a message of connection and caring. This type of sup-
portive and inclusive campus community is a beneficial end in
itself and will fortify the resilience and coping of its members
through reducing isolation and enhancing social support.
Following prevention on the continuum of need and response
are early interventions, for which the focus shifts from the entire
student population to an identified subpopulation with elevated
known risk factors. The majority of individuals in these subpopu-
lations will not be on the extreme end of the distress spectrum
indicating a suicidal crisis but may be experiencing less severe
suicidal thoughts or may simply be at increased risk for entering
the suicidal continuum. Early interventions include treatment mo-
dalities such as group and workshop formats, as well as the use of
Web-based interventions. For example, the American Foundation
for Suicide Prevention’s College Screening Project invites students
by email to participate in anonymous online screenings, with
follow-up contact and encouragement to seek treatment if a certain
level of distress is indicated (Haas, Hendin, & Mann, 2003). Other
early interventions include the creation of support and therapy
groups for students who are experiencing problems with romantic
relationships, family and friends, or academics. The purpose of
such interventions is to boost recovery from negative life events
that correlate highly with suicidality and thereby to proactively
counteract the worsening of suicidal thoughts among these groups.
220 DRUM, BROWNSON, BURTON DENMARK, AND SMITH
The next phase of treatment along the temporal continuum, that
of treating the individual in a state of suicidal crisis, needs no
explanation because it is currently the ubiquitous mode of dealing
with college student suicidality and has been elaborated on at
length elsewhere in this article. This phase of treatment, which if
used alone will do little to decrease the incidence of suicidality
among college students, is nonetheless necessary to protect those
who have reached the extreme end of the suicidal continuum, that
of intense crisis or suicide attempt. This individual-focused form
of intervention can also benefit from the shift in institutional
attitudes and increased environmental support advocated previ-
ously. Currently, the dual-role conflict that establishes the univer-
sity as the adversary of the suicidal student appears to discourage
many students from seeking the help they need because of fears of
academic consequences. By changing the university’s perception
of the suicidal student as a problem for campus mental health
services to deal with and by reframing this perception to focus on
suicidality as the problem to be shared among all members of the
university, institutions can continue to engage with rather than
detach from students in distress.
According to the current paradigm, if the individual treatment
phase is deemed successful, then the student is released from
treatment, typically into the same environment in which the sui-
cidal crisis originally emerged. The problem-focused paradigm
recognizes that suicide is a highly relapsing condition akin to
substance abuse, depression, eating disorders, and a variety of
states that fuel detachment from oneself. Simply treating and
resolving the current crisis does not necessarily reduce future
occurrences, and these individuals are placed at increased risk of
eventually completing suicide by virtue of having entered the
suicidal continuum. A final phase of treatment focused on relapse
prevention is needed and may include the use of support groups that
incorporate coping and problem-solving skills and mindfulness-based
practices. Relapse prevention may also consist of programs such as
disease management or recovery support centers, many of which are
currently aimed at reducing relapse for substance abuse but may be
expanded to include other highly recurrent conditions, including sui-
cidality. The inclusion of this final relapse-prevention phase of treat-
ment is crucial for achieving the goal of reducing both the incidence
and prevalence of student suicidality.
In summary, colleges and universities face increasing pressure
to prevent college student suicide in the context of growing levels
of distress among the student body and diminishing resources with
which to treat that distress. A comprehensive approach to suicide
prevention can come about only through a fundamental shift in the
paradigm for conceptualizing suicidality. The problem-centered
approach that we advocate here calls for the entire administrative
structure of the university to become engaged with the problem of
suicidality and to implement policies and programs to reduce
suicide by intervening at all points along the suicidal continuum,
from prevention through early intervention, crisis treatment, and
relapse prevention. Furthermore, expanding the campus dialogue
around issues of suicide to include all stakeholders will help
involve a greater cross-section of campus personnel and will add
valuable perspectives while facilitating program development, im-
plementation and planning. Finally, it is apparent that all clini-
cians, regardless of whether they serve a college student popula-
tion, will benefit from greater understanding of the entire range of
suicidal thoughts and behaviors that precede the crisis points that
are so often the focus of treatment. This knowledge will contribute
to more targeted risk assessments and treatment approaches. Fur-
ther research is needed to both inform the effective implementation
of these interventions and to develop and test new approaches.
References
American College Health Association. (2006). American College Health
Association—National College Health Assessment: Reference group
data report spring 2006. Baltimore: Author.
Benton, S. A., Robertson, J. M., Tseng, W., Newton, F. B., & Benton, S. L.
(2003). Changes in counseling center client problems across 13 years.
Professional Psychology: Research and Practice, 34, 66 –72.
Brown, D. R., & Blanton, C. J. (2002). Physical activity, sports participa-
tion, and suicidal behavior among college students. Medicine and Sci-
ence in Sports and Exercise, 34, 1087–1096.
Bryan, C. J., & Rudd, D. M. (2006). Advances in the assessment of suicide
risk. Journal of Clinical Psychology, 62, 185–200.
Centers for Disease Control and Prevention. (1995). Youth risk behav-
ior surveillance: National College Health Risk Behavior Survey.
MMWR Surveillance Summaries, 46(SS-6), 1–54. Retrieved Septem-
ber 15, 2007, from http://www.cdc.gov/mmwr/preview/mmwrhtml/
00049859.htm
Centers for Disease Control and Prevention. (2007). Suicide prevention:
Youth suicide. Retrieved January 14, 2008, from http://www.cdc.gov/
ncipc/dvp/Suicide/youthsuicide.htm
Deane, F. P., & Todd, D. M. (1996). Attitudes and intentions to seek
professional psychological help for personal problems or suicidal think-
ing. Journal of College Student Psychotherapy, 10(4), 45–59.
Frohlich, K. L., & Potvin, L. (1999). Health promotion through the lens of
population health: Toward a salutogenic setting. Critical Public Health,
9, 211–222.
Gallagher, R. P. (2004). National Survey of Counseling Center Directors.
Arlington, VA: International Association of Counseling Services.
Gallagher, R. P. (2006). National Survey of Counseling Center Directors.
Arlington, VA: International Association of Counseling Services.
Garrett Lee Smith Memorial Act of 2004, Pub. L. No. 108 –355, 118 Stat.
1404 (2004).
Haas, A. P., Hendin, H., & Mann, J. J. (2003). Suicide in college students.
American Behavioral Scientist, 46, 1224 –1240.
Joiner, T. E., Conwell, Y., Fitzpatrick, K. K., Witte, T. K., Schmidt, N. B.,
Berlim, M. T., et al. (2005). Four studies on how past and current
suicidality relate even when “everything but the kitchen sink” is covar-
ied. Journal of Abnormal Psychology, 114, 291–303.
Kahn, J. H., & Williams, M. N. (2003). The impact of prior counseling on
predictors of college counseling center use. Journal of College Coun-
seling, 6, 144 –154.
Kisch, J., Leino, E. V., & Silverman, M. M. (2005). Aspects of suicidal
behavior, depression and treatment in college students: Results from the
spring 2000 National College Health Assessment Survey. Suicide and
Life-Threatening Behavior, 35, 3–13.
Nott v. George Washington University, No. 05– 8503 (D.C. Super. Ct. Oct.
25, 2005).
Rudd, D. M. (2004). University counseling centers: Looking more and
more like community clinics. Professional Psychology: Research and
Practice, 35, 316 –317.
Schwartz, A. J. (2006a). College student suicide in the United States:
1990 –1991 through 2003–2004. Journal of American College Health,
54, 341–352.
Schwartz, A. J. (2006b). Four eras of study of college student suicide in the
United States: 1920 –2004. Journal of American College Health, 54,
353–366.
Shin v. Massachusetts Institute of Technology, 19 Mass. L. Rep. 570
(Mass. Super. Ct. 2005).
221
COLLEGE STUDENT SUICIDALITY
Silverman, M. M. (1993). Campus student suicide rates: Fact or artifact?
Suicide and Life-Threatening Behavior, 23, 329 –342.
Silverman, M. M., Meyer, P. M., Sloane, F., Raffel, M., & Pratt, D. M.
(1997). The Big Ten Student Suicide Study: A 10-year study of suicides
on Midwestern university campuses. Suicide and Life-Threatening Be-
havior, 27, 285–303.
Suicide Prevention Resource Center. (2004). Promoting mental health and
preventing suicide in college and university settings. Newton, MA:
Education Development Center.
U.S. Department of Education, National Center for Education Statistics,
Integrated Postsecondary Education Data System. (2004, fall). Postsec-
ondary institutions in the United States: Fall 2004 and degrees and
other awards conferred: 2003–2004. Retrieved January 15, 2008, from
http://nces.ed.gov/pubsearch/getpubcats.asp?sid⫽010#
Westefeld, J. S., Homaifar, B., Spotts, J., Furr, S., Range, L., & Werth, J. L.
(2005). Perceptions concerning college student suicide: Data from four
universities. Suicide and Life-Threatening Behavior, 35, 640 – 645.
Wingate, L. R., Joiner, T. E., Walker, R. L., Rudd, D. M., & Jobes, D. A.
(2004). Empirically informed approaches to topics in suicide risk as-
sessment. Behavioral Sciences and the Law, 22, 651– 665.
Wyman, P. A., Brown, C. H., Inman, J., Cross, W., Schmeelk-Cone, K.,
Guo, J., et al. (2008). Randomized trial of a gatekeeper program for
suicide prevention: 1-year impact on secondary school staff. Journal of
Consulting and Clinical Psychology, 76, 104 –115.
Received July 10, 2008
Revision received October 8, 2008
Accepted October 9, 2008 䡲
222 DRUM, BROWNSON, BURTON DENMARK, AND SMITH
- A preview of this full-text is provided by American Psychological Association.
- Learn more
Preview content only
Content available from Professional Psychology Research and Practice
This content is subject to copyright. Terms and conditions apply.
