Accepted Manuscript

Family history of suicide and high motor impulsivity distinguish suicide attempters
from suicide ideators among college students

Yong-guang Wang, Shen Chen, Zhi-ming Xu, Zhi-hua Shen, Yi-quan Wang, Xiao-yan
He, Ri-fang Cao, David L. Roberts, Jian-fei Shi, Yi-qiang Wang

PII: S0022-3956(16)30300-4
DOI: 10.1016/j.jpsychires.2017.02.006
Reference: PIAT 3061

To appear in: Journal of Psychiatric Research

Received Date: 30 August 2016

Revised Date: 6 February 2017
Accepted Date: 7 February 2017

Please cite this article as: Wang Y-g, Chen S, Xu Z-m, Shen Z-h, Wang Y-q, He X-y, Cao R-f, Roberts
DL, Shi J-f, Wang Y-q, Family history of suicide and high motor impulsivity distinguish suicide attempters
from suicide ideators among college students, Journal of Psychiatric Research (2017), doi: 10.1016/
j.jpsychires.2017.02.006.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

Family history of suicide and high motor impulsivity distinguish suicide

attempters from suicide ideators among college students
Yong-guang Wangabcd*, Shen Chend, Zhi-ming Xua, Zhi-hua Shena, Yi-quan
Wanga, Xiao-yan Hee, Ri-fang Caoa, David L. Robertsf, Jian-fei Shiabc**,
Yi-qiang Wangabc

PT
a
Department of Brain Functioning Research, The Seventh Hospital of Hangzhou, 305
Tianmushan Road, Hangzhou, 310013, Zhejiang Province, China

RI
b
Clinical Institute of Mental Health in Hangzhou, Anhui Medical University,
Hangzhou, Zhejiang Province, China

SC
c
Mental Health Center, Zhejiang University School of Medicine, Hangzhou, Zhejiang
Province, China

U
d
Institute of Film & Television Animation, China Academy of Art, Hangzhou,
AN
Zhejiang Province, China
e
Department of Health Education, the Center of Disease Control and Prevention of
M

Hangzhou, Hangzhou, Zhejiang Province, China

f
Department of Psychiatry, University of Texas Health Science, San Antonio, USA
D
TE

*Corresponding author: Yong-guang Wang, Department of Brain Functioning

EP

Research, the Seventh Hospital of Hangzhou, Hangzhou, Zhejiang Province, China.

E-mail address: wangyongguang78@163.com
**Co-corresponding author: Jian-fei Shi, E-mail address: shijf659293@126.com
C
AC
 ACCEPTED MANUSCRIPT

Family history of suicide and high motor impulsivity distinguish suicide

attempters from suicide ideators among college students
Abstract
Suicide in college students has become an important public health issue in China.
The aim of this study was to identify the differences between suicide attempters and

PT
suicide ideators based on a cross-sectional survey. Our results indicate that although
female gender, positive screening for psychiatric illness, positive family history of

RI
suicide, elevated overall impulsivity, and elevated motor impulsivity were correlated
with suicidal ideation, only positive family history of suicide and high motor

SC
impulsivity could differentiate suicide attempters from suicidal ideators. Future
research with a longitudinal and prospective study design should be conducted to

U
confirm these findings.
AN
Key words: suicidal ideation; suicide attempt; impulsivity; family history of suicide;
college students
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Family history of suicide and high motor impulsivity distinguish suicide

attempters from suicide ideators among college students

1. Introduction
Since 1999, there has been a sharp increase in the number of college students in
China due to policies to increase enrollment in higher education. As of 2014, there

PT
were more than 25 million college students in China (National Bureau of Statistics of
China, 2014). According to two recent meta-analyses, the overall pooled prevalence

RI
of suicidal ideation and suicide attempt among Chinese college students was 10.72%

SC
(Li et al., 2014a) and 2.8% (Yang et al., 2015), respectively. Although there is no
accurate national suicide rate in this population, suicide in college students has

U
become an important public health issue.
Suicide is a continuous process, starting from suicidal ideation, suicide attempt,
AN
to suicide completion. Previous studies have documented that suicidal ideation and
suicide attempt are strongly predictive of subsequent suicide completion (e.g.,
M

O'Connor et al., 2013). However, although suicidal ideation is an important precursor

D

of suicide attempt, most suicide ideators do not attempt. A prospective study reported
that only 7.4% of individuals with first-onset suicide ideation reported making a
TE

suicide attempt in the following year (ten Have et al., 2009). In addition, key risk
factors seem not to distinguish suicidal attempters from those with suicidal ideation
EP

who do not attempt (Klonsky & May, 2014; May & Klonsky, 2016; May et al., 2012).
As proposed by Klonsky and May (Klonsky & May, 2014; May & Klonsky, 2016;
C

Klonsky et al., 2016), it is important to determine what distinguishes suicide

AC

attempters from suicide ideators who do not attempt.

To date, considerable studies have been conducted to identify the risk factors
associated with suicidal ideation among Chinese college students. Although numerous
potential risk factors have been identified, previous meta-analytic studies have found
that the most reliable and robust risk factors for suicidal ideation were the presence of
psychiatric disorders (Li et al., 2014b) and a family history of suicide (Xia et al.,
2015). These two factors have also been found to predict suicide attempts among
 ACCEPTED MANUSCRIPT

college students. For instance, Cao et al. (2009) reported that the presence of
depression and anxiety were predictive of suicide attempt history. Ai et al. (2008) and
Liu et al. (2010) found that students with a family history of suicide were more
susceptible to suicide attempts. However, it is worth noting that these studies on
suicide attempts did not directly test whether these factors increased risk of attempts

PT
among ideators. Thus, these studies could not determine what differentiates between
suicide attempters and suicidal ideators who do not attempt.

RI
An additional factor that is thought to be critical for the transition from suicidal
thoughts to action is impulsivity (Mann et al., 1999; Bryan & Rudd, 2006). However,

SC
inconsistent results regarding the relationship between impulsivity and suicide attempt
have been reported (e.g., Brezo et al., 2007; Anestis et al., 2014; Klonsky and May,

U
2010). For example, Brezo et al. (2007) conceptualized impulsivity as a unitary
AN
construct, and found that attempters and ideators reported comparable impulsivity on
the Barratt Impulsiveness Scale. On the other hand, Klonsky and May (2010) reported
M

that a specific impulsivity-related trait (i.e., poor premeditation) distinguished suicide

attempters from ideators who do not attempt, whereas a unidimensional measure of
D

impulsivity did not. Indeed, impulsivity is not a unitary construct, but encompasses a
TE

cluster of heterogeneous behavioral tendencies (Stanford et al., 2009; Meule, 2013).

Thus, there is need for research examining the relationship between different
EP

impulsivity traits and suicide attempt.

Accordingly, the present study aimed to investigate whether these potential
variables could distinguish suicide attempters from suicide ideators who do not
C

attempt attempt, based on a cross-sectional survey.

AC

2. Methods
2.1. Sampling and Participants
The current study was approved by the Ethics Committee of the Center of
Disease Control and Prevention of Hangzhou. Data was gathered beginning in 2010
as part of routine web-based psychological assessment for freshmen at three
universities/college in the city of Hangzhou (one provincial university, one municipal
university, and one vocational & technical College). In 2011, the municipal university
 ACCEPTED MANUSCRIPT

withdrew from the study due to computer network problems. In total, out of the
19,376, college freshman who entered our recruitment sites in 2010 and 2011, all but
38 completed the assessment. These 38 individuals either did not complete their
survey or skipped required items; they were defined as non-responders and excluded
from the analysis.

PT
Figure 1 shows the two survey questions that were used to identify suicidal
ideators and suicidal attempters. All participants were first asked, Have you ever

RI
thought about committing suicide?" Those who responded yes were categorized as
suicidal ideators and then asked, Have you ever attempted suicide? Within the full

SC
sample, there were 1,008 suicidal ideators and 18,330 students without suicidal
ideation. Among the 1,008 suicidal ideators, 161 indicated that they had attempted

U
suicide and 847 indicated that they had not attempted.
AN
2.2. Measures
As mentioned above, history of suicidal ideation and suicide attempt were
measured by two survey items. In addition, family history of suicide behavior was
M

measured for all participants with the yes/no item, Do you have a family member or
D

relative who died by suicide? Participants who answered yes were categorized as
having a positive family history of suicide.
TE

The presence of psychiatric disorders among participants was measured using the
12-item General Health Questionnaire (GHQ-12) (Golderberg and Williams, 1988).
EP

The GHQ-12 as a screening instrument for psychiatric disorders has been validated in
China (Yang et al., 2003). The total score for GHQ-12 was 0 to 12 points, with a
C

bi-modal scoring method. Based on previous literature (Yang et al., 2003; Phillips et
AC

al., 2009), participants with a score of 4 or greater were considered positive for the
presence of psychiatric disorder. The Cronbach's alpha was 0.65 for the GHQ-12 in
the present study, indicating acceptable reliability (Nunnally, 1978).
Impulsivity was measured using the Chinese revised version of the Barratt
Impulsiveness Scale (BIS-11) (Li et al., 2011; validated for web-based administration
by Shen et al., 2016), which measures cognitive (inability to focus attention or
concentrate), motor (acting without thinking), and non-planning impulsivity (lack of
 ACCEPTED MANUSCRIPT

future orientation or forethought). The BIS-11 contains 30 items, with each item
scored from 1 to 5. The total score of the BIS-11 and each subscale range from 30 to
150 and 10 to 50, respectively. As mentioned earlier, a unidimensional measure of
impulsivity did not sufficiently differentiate suicide attempters from ideators (Brezo et
al., 2007; Klonsky and May, 2010). The BIS-11 subscales have been shown to yield

PT
meaningful differences in a variety of clinical samples (see Stanford et al., 2009).
Thus, we used a cut-off point of 75th percentile for both total score and each subscale

RI
score to define high impulsivity subjects in each dimension according to previous
literature (Baca-Garcia et al., 2004; Maloney et al., 2009). The Cronbachs alpha for

SC
the BIS-11 in the present study are 0.84 (cognitive), 0.85 (motor), 0.82 (non-planning),
and 0.91 (total score).

U
2.3. Analysis procedure
Statistical analyses were performed using the SPSS statistical package version
AN
17.0. Fisher's exact test and one-way ANOVA was used to compare proportions
between groups for categorical variables and continuous variables, separately. Phi ()
M

and partial eta-squared (partial2) were used to estimate effect sizes, respectively.
D

Following established guidelines (Cohen, 1988), effect sizes were considered very
small if Phi <0.10, small if <0.30, moderate if <0.50 and strong if 0.50. The
TE

significant categorical variables from group comparisons were entered into a logistic
regression analysis. The forward-stepwise and backward-stepwise procedures in the
EP

logistic regression analysis were both used to identify the most stable model. To
reduce the effect of multi-collinearity, the BIS-11 total score positive screening and
C

BIS-11 subscales positive screening were entered into different logistic regression
AC

equations.
3 Results
3.1. Group comparisons
3.1.1. Suicidal ideators vs. students without suicidal ideation
Out of 19,338, 1,008 (5.2%) were confirmed as suicidal ideators. Table 1
summarizes the details of group comparisons between suicidal ideators and students
without suicidal ideation. As Table 1 shows, there were no significant differences
 ACCEPTED MANUSCRIPT

between groups for age (p=0.30, partial 2< 0.01).

Suicidal ideators had a significantly higher percentage of females (60.8%) than
students without suicidal ideation (46.8%), but with a very small effect size (p<0.01,
=0.06). A significantly higher rate of GHQ-12 positive screenings was found in
suicidal ideators (37.0%) than in students without suicidal ideation (14.4%) (p<0.01,

PT
=0.14). Suicidal ideators (8.6%) reported more frequent family history of suicide
than students without suicidal ideation (3.4%), with a very small effect size (p<0.01,

RI
=0.06).
On the BIS-11, suicidal ideators had a higher proportion of positive screenings

SC
than students without suicidal ideation on the total score, cognitive impulsivity
subscale, motor impulsivity subscale, and non-planning impulsivity subscale (all p

U
0.01). The effect sizes were all very small (all 0.10), except for motor impulsivity
AN
(=0.11) (Details in Table 1).
3.1.2. Suicide attempters vs. suicidal ideators without suicide attempt
M

Out of the total sample of 19,338, 161 (0.8%) were categorized as suicide
attempters. Group comparisons between suicide attempters and suicidal ideators
D

without suicide attempt revealed a very different pattern than that seen between
TE

non-ideators and ideators. As Table 2 shows, no significant difference was found

between groups for age (p=0.64, partial 2< 0.01), gender (p=0.48, =0.02), GHQ-12
EP

positive screening (p=0.54, =0.02), or positive screenings on the BIS-11 total score
(p=0.26, =0.04), cognitive impulsivity subscale (p=0.58, =-0.02), or non-planning
impulsivity subscale (p=1.00, <-0.01). The suicide attempters (18.0%) had a
C

significant higher rate of positive family history of suicide than suicidal ideators
AC

without suicide attempt (6.8%) (p<0.01, =0.15). Also, a significantly higher rate of
motor impulsivity was found in suicidal attempters (57.1%) than suicidal ideators
without suicidal attempt (46.0%) (p<0.01), but with a very small effect size (=0.08).
More details are shown in Table 2.
3.2. Logistic regression equations
The results of logistic regression equations for suicidal ideation and suicide
attempt are showed in Table 3. Multi-step logistic regression analyses indicate that
 ACCEPTED MANUSCRIPT

female gender, GHQ-12 positive screening, positive family history of suicide, BIS-11
total score positive screening and high motor impulsivity significantly differentiated
suicidal ideators from all other participants. For suicide attempters, logistic regression
analysis yielded a very different model, with only positive family history of suicide
and high motor impulsivity significantly distinguishing between suicide attempters

PT
and suicidal ideators who had not attempted suicide.
4. Discussion

RI
Suicide in college students is a major public health concern in China. The
primary aim of this study was to explore the potential factors that could distinguish

SC
suicide attempters from suicidal ideators based on a cross-sectional survey. Our
primary findings show that, although many differences were found between students

U
with versus without suicidal ideation, only two factors (i.e., positive family history of
AN
suicide and high motor impulsivity) differentiated between suicide attempters and
those with suicidal ideation who have not attempted. These findings have important
M

implications for future research and for suicide prevention.

Regarding gender, consistent with two recent meta-analyses (Li et al., 2014b;
D

Yang et al., 2015), our results indicate a very weak association between female gender
TE

and suicidal ideation (Li et al., 2014b), and no significant gender difference between
suicide attempters and ideators-only (Yang et al., 2015). These findings are also line
EP

with Chinese national suicide mortality data. Although fifteen years ago it was well
documented that women had more suicide deaths than men in China (Phillips et al.,
2002a; Phillips et al., 2002b), more recent research shows similar rates of suicide
C

mortality for female youth and male youth in China (Sun et al., 2013; Wang et al.,
AC

2014).
With respect to GHQ-12, consistent with Li et al.s work (Li et al., 2014b), our
results indicate the presence of psychiatric disorders independently differentiated
suicidal ideators from all students. However, our findings also suggest that the
presence of psychiatric disorders adds little to distinguish suicide attempters from
suicidal ideators. These findings are consistent with the recent literature with respect
to the association between psychiatric disorders and suicidal behavior. There is
 ACCEPTED MANUSCRIPT

accumulating evidence that although the presence of overall psychiatric disorders

predicts the onset of suicidal ideation, they are less useful in determining which
people with suicide ideation will make a suicide attempt (Nock et al., 2008, 2010;
May and Klonsky, 2016).
It is worthy of note that positive family history of suicide and high motor

PT
impulsivity are the only factors that differentiated suicide attempters from suicidal
ideators in our sample. Familial transmission of suicidal behavior has been well

RI
documented in previous literature (e.g., Roy, 1983; Brent and Melhem, 2008;
Sarchiapone et al., 2009). As a distal risk factor, familial transmission of suicidal

SC
behavior may be attributable to both genetic factors and other factors. Twin (Roy et al.,
1995; Voracek and Loibl, 2007) and adoption studies (Petersen et al., 2013, 2014)

U
have shown that genetic predispositions accounted for a portion of familial
AN
transmission. In addition, suicide contagion (Cheng et al., 2015; Mueller and Abrutyn,
2015) might be transmitted through families, leading to a familial influence on
M

suicidal behavior.
Our results also suggest that, while a positive link was found between BIS-11
D

total score positive screening and suicidal ideators, only high motor impulsivity (i.e.,
TE

acting without thinking) was elevated in participants who have attempted compared to
ideators who have not attempted. These findings are consistent with a study by
EP

Klonsky and May (2010), which indicated that a unidimensional measure of

impulsivity failed to distinguish suicide attempters from ideators. Klonsky and May
(2010) also found that suicide attempters tended to act without thinking through the
C

consequences (i.e., poor premeditation). Indeed, impulsivity is a broad and

AC

multi-faceted construct, and there are different associations between BIS subscales
and different clinical population (see details in Stanford et al., 2009 and Meule, 2013).
Thus, we suggest that researchers should examine the relationship between different
impulsivity traits and suicidal behavior in greater detail in the future.
Several limitations of the present study should be noted. First, we used a
cross-sectional survey rather than prospective research design, and observed overall
small effect sizes. Longitudinal and prospective studies are the best methods to study
 ACCEPTED MANUSCRIPT

differences between suicide attempters and ideators. Future studies should be

conducted to clarify what predicts the transition from suicidal ideation to suicide
attempt within individual people. Second, it is plausible that our use of a web-based
survey increased the risk of participants under-reporting suicidality. Our results
indicate that the lifetime prevalence of the suicidal ideation and suicide attempt was

PT
5.2% and 0.8% in our sample, respectively. These findings are lower than the results
reported by two recent meta-analyses (Li et al., 2014a; Yang et al., 2015). Although

RI
our participants had been clearly informed that the survey is for research purposes
only, it is reasonable to believe that some of the students may have had concerns over

SC
confidentiality. Third, as our survey collected data retrospectively, our findings may
have been influenced by recall bias. Again, longitudinal and prospective studies

U
would decrease this risk. Fourth, the fact that we used self-report data may have
AN
increased the risk of misreporting. The presence of psychiatric disorders was
determined with the GHQ-12. Although this instrument has been validated in previous
M

literature as a screening tool for psychological disorders, clinical interview for

psychiatric disorders would be a stronger method in future research. Finally, the
D

sample size of suicide attempters in our study was small, which limited our power to
TE

detect differences.
Despite of these drawbacks, we believe that our findings will be helpful in
EP

providing a better understanding of suicidal behavior among Chinese college students.

Whereas some of the most-reported risk factors for suicidal ideation had limited
utility in distinguishing suicide attempters from suicidal ideators, family history of
C

suicide and high motor impulsivity appeared to increase the risk of suicide attempt in
AC

our sample. Future research with longitudinal and prospective study designs should be
conducted to further evaluate our findings.
Acknowledgments
This work was supported by the National Natural Science Foundation of China
(Grant no. 81471366), the Science and Technology Department of Zhejiang province
(Grant no. 2015C03G2010109), and the Science and Technology Bureau of Hangzhou
(Grant nos. 20150733Q33 and 20140633B18).
 ACCEPTED MANUSCRIPT

References:
Ai et al., 2008 M. Ai, L. Kuang, M.J. Wang, Y. Shen, D.Q. Li, X.X. Gao, J.M. Chen,
X.M. Li, Y.J. Niu, M.R. Phillips. A study of suicide attempts and related factors of
college students in Chongqing. Chin J Nerv Ment Dis, 34(10), (2008), 594-597. (In
Chinese)

PT
Anestis et al., 2014 M.D. Anestis, K.A. Soberay, P.M. Gutierrez. Reconsidering the
link between impulsivity and suicidal behavior. Pers Soc Psychol Rev., 18(4), (2014),

RI
366-386.
Baca-Garcia et al., 2004 E. Baca-Garcia, C. Vaquero, C. Diaz-Sastre, E. Garca-Resa,

SC
J. Saiz-Ruiz, J. Fernndez-Piqueras, J. de Leon. Lack of association between the
serotonin transporter promoter gene polymorphism and impulsivity or aggressive

U
behavior among suicide attempters and healthy volunteers. Psychiatry Res, 126(2),
AN
(2004), 99-106.
Brent and Melhem, 2008 D.A. Brent, N. Melhem. Familial Transmission of Suicidal
M

Behavior. Psychiat Clin N AM, 31(2), (2008), 157177.

Brezo et al., 2007 J. Brezo, J. Paris, R. Tremblay, F. VitaroM. Hbert. G. Turecki.
D

Identifying correlates of suicide attempts in suicidal ideators: a population-based

TE

study. Psychol Med, 37(11), (2007), 1551-62.

Bryan & Rudd, 2006 C.J. Bryan, M.D. Rudd. Advances in assessment of suicide risk.
EP

J Clin Psychol, 62(2), (2006), 185-200.

Cao et al., 2009 H.Y. Cao, Y.H. Sun, Y.S. Yao, C.X. Zhou, L.S. Yang. Suicide
behavior and its risk factors among medical undergraduates in Anhui province. Chin J
C

Sch Health, 30(1), (2009), 38-39. (In Chinese)

AC

Cheng et al., 2015 Q. Cheng, H. Li, V. Silenzio, E.D. Caine. Suicide contagion: a
systematic review of definitions and research utility. PLoS One, 9(9), (2015),
e108724-e108724.
Cohen, 1988 J. Cohen. Statistical power analysis for the behavioural sciences.
Hillsdale, NJ: Erlbaum, 1988.
Golderberg and Williams, 1988 D. Golderberg, P. Williams. A users guide to the
general health questionnaire. Windsor: NFER-Nelson, 1988.
 ACCEPTED MANUSCRIPT

Klonsky and May, 2010 E.D. Klonsky, A. May. Rethinking impulsivity in suicide.
Suicide Life Threat Behav, 40(6), (2010), 612-619.
Klonsky and May, 2014 E.D. Klonsky, A.M. May. Differentiating suicide attempters
from suicide ideators: a critical frontier for suicidology research. Suicide Life Threat
Behav, 44(1), (2014), 1-5.

PT
Klonsky et al., 2016 E.D. Klonsky, A.M. May, B.Y. Saffer. Suicide, Suicide Attempts,
and Suicidal Ideation. Annu Rev Clin Psycho, 12(1), (2016), annurevclinpsy

RI
021815093204.
Li et al., 2011 X.Y. Li, M.R. Phillips, D. Xu, Y.L. Zhang, S.J. Yang, Y.S. Tong.

SC
Reliability and validity of an adapted Chinese version of the Barratt Impulsiveness
Scale. Chin Ment Health J, 25(8), (2011), 610-615. (In Chinese)

U
Li et al., 2014a Z.Z. Li, Y.M. Li, X.Y. Lei, D. Zhang, L. Liu, S.Y. Tang, L. Chen.
AN
Prevalence of suicidal ideation in Chinese college students: a meta-analysis. PLoS
One, 9(10), (2014), e104368.
M

Li et al., 2014b Y. Li, X. Lei, D. Zhang, L. Liu, S. Tang. Determinant factors of

suicidal ideation among college students: A meta-analysis. Chin J Clin Psychol, 22(4),
D

(2014), 638-640.(In Chinese)

TE

Liu et al., 2010 W.T. Liu, L. Kuang, M. Ai, X.M. Li, M. Chen, D.D. Lou, Y.J. Niu,
M.R. Phillips. Study on the related factors to suicidal behavior between two groups in
EP

university students. Chin J Epidemiol, 31(5), (2010), 517-520. (In Chinese)

Voracekand Loibl, 2007 M. Voracek, L.M. Loibl. Genetics of suicide: a systematic
review of twin studies. Wien Klin Wochenschr, 119(15-16), (2007), 463-475.
C

Maloney et al., 2009 E. Maloney, L. Degenhardt, S. Darke, E.C. Nelson. Impulsivity

AC

and borderline personality as risk factors for suicide attempts among

opioid-dependent individuals. Psychiatry Res, 169(1), (2009), 16-21.
Mann et al., 1999 J.J. Mann, C. Waternaux, G.L. Haas, K.M. Malone. Toward a
clinical model of suicidal behavior in psychiatric patients. Am J Psychiatry, 156(2),
(1999),181-189.
May and Klonsky, 2016 A.M. May, E.D., Klonsky. What distinguishes suicide
 ACCEPTED MANUSCRIPT

attempters from suicide ideators? A meta-analysis of potential factors. Clin Psychol:

Sci Pract, 23, (2016), 5-20.
May et al., 2012 A.M. May, E.D. Klonsky, D.N. Klein. Predicting future suicide
attempts among depressed suicide ideators: A 10-year longitudinal study. J Psychiatr
Res, 46(7), (2012), 946-952.

PT
Meule, 2013 A. Meule. Impulsivity and overeating: a closer look at the subscales of
the Barratt Impulsiveness Scale. Front Psychol, 4, (2013), 177.

RI
Mueller and Abrutyn, 2015 A.S. Mueller, S. Abrutyn. Suicidal disclosures among
friends: using social network data to understand suicide contagion. J Health SocBehav,

SC
56(1), (2015), 131-148.
National Bureau of Statistics of China, 2014 National Bureau of Statistics of

U
China.2014. http://data.stats.gov.cn/easyquery.htm?cn=C01. (Accessed 2016 Aug 7)
AN
Nock et al., 2008 M.K. Nock, G. Borges, E.J. Bromet, J. Alonso, M. Angermeyer, A.
Beautrais, R. Bruffaerts, W.T. Chiu, G. de Girolamo, S. Gluzman, R. de Graaf, O.
M

Gureje, J.M. Haro, Y. Huang, E. Karam, R.C. Kessler, J.P. Lepine, D. Levinson, M.E.
Medina-Mora, Y. Ono, J. Posada-Villa, D. Williams. Cross-national prevalence and
D

risk factors for suicidal ideation, plans and attempts. Br J Psychiatry, 192(2), (2008),
TE

98-105.
Nock et al., 2010 M.K. Nock, I. Hwang, N.A. Sampson, R.C. Kessler. Mental
EP

Disorders, Comorbidity and Suicidal Behavior: Results from the National

Comorbidity Survey Replication. Mol Psychiatry, 15(8), (2010), 868876.
Nunnally 1978 J.C. Nunnally. Psychometric theory. New York: McGraw-Hill, 1978.
C

O'Connor et al., 2013 R.C. O'Connor, R. Smyth, E. Ferguson, C. Ryan, J.M. Williams.
AC

Psychological Processes and Repeat Suicidal Behavior: A Four-Year Prospective

Study. J Consult Clin Psychol, 81(6), (2013), 1137-1143.
Petersen et al., 2013 L. Petersen, T.I. Srensen, P.K. Andersen, P.B. Mortensen, K.
Hawton. Genetic and familial environmental effects on suicide--an adoption study of
siblings. PLoS One, 8(10), (2013), e77973.
Petersen et al., 2014 L. Petersen, T.I. Srensen, P.K. Andersen, P.B. Mortensen, K.
Hawton. Genetic and familial environmental effects on suicide attempts: a study of
 ACCEPTED MANUSCRIPT

Danish adoptees and their biological and adoptive siblings. J Affect Disord, 155,
(2014), 273-277.
Phillips et al., 2002a M.R. Phillips, X. Li, Y. Zhang. Suicide rates in China, 199599.
Lancet, 359(9309), (2002a), 835-840.
Phillips et al., 2002b M.R. Phillips, G. Yang, Y. Zhang, L. Wang, H. Ji, M. Zhou. Risk

PT
factors for suicide in China: a national case-control psychological autopsy study.
Lancet, 360(9347), (2002b), 1728-1736.

RI
Phillips et al., 2009 M.R. Phillips, J. Zhang, Q. Shi, Z. Song, Z. Ding, S. Pang, X. Li,
Y. Zhang, Z. Wang. Prevalence, treatment, and associated disability of mental

SC
disorders in four provinces in China during 2001-05: an epidemiological survey.
Lancet, 373(9680), (2009), 2041-2053.

U
Roy et al., 1995 A. Roy, N.L. Segal, M. Sarchiapone. Attempted suicide among living
AN
co-twins of twin suicide victims. Am J Psychiatry, 152(7), (1995), 1075-1076.
Roy, 1983 A. Roy. Family history of suicide, Arch Gen Psychiatry., 40(9), (1983),
M

971-974.
Sarchiapone et al., 2009 M. Sarchiapone, V. Carli, L. Janiri, M. Marchetti, C. Cesaro,
D

A. Roy. Family history of suicide and personality. Arch Suicide Res, 13(2), (2009),
TE

178-184.
Shen et al., 2016 Z. Shen, Y. Wang, Y. Wang. A study on the reliability and validity of
EP

the Chinese revised version of Barratt Impulsiveness Scale in Chinese college student
with a web-based survey. Zhejiang Pre Med, 28(4), (2016), 368-370. (In Chinese)
Stanford et al., 2009 M.S. Stanford, C.W. Mathias, D.M. Dougherty, S.L. Lake, N.E.
C

Anderson, J.H. Patton. Fifty years of the Barratt Impulsiveness Scale: An update and
AC

review. Pers Indiv Differ, 47(5), (2009), 385-395.

Sun et al., 2013 J. Sun, X. Guo, J. Zhang, C. Jia, A. Xu. Suicide rates in Shandong,
China, 1991-2010: rapid decrease in rural rates and steady increase in male-female
ratio. J Affect Disord, 146(3), (2013), 361-368.
ten Have et al., 2009 M. ten Have, R. de Graaf, S. van Dorsselaer, J. Verdurmen, H.
van 't Land, W. Vollebergh, A. Beekman. Incidence and course of suicidal ideation
and suicide attempts in the general population. Can J Psychiatry, (12), (2009), 824-33.
 ACCEPTED MANUSCRIPT

Wang et al., 2014 C.W. Wang, C.L. Chan, P.S. Yip. Suicide rates in China from 2002
to 2011: an update. Soc Psychiatry Psychiatr Epidemiol, 49(6), (2014), 929-941.
Xia et al., 2015 Y. Xia, P. Wang, Y. Mei, Z.Y. Yang. A meta-analysis of risk factors for
suicidal ideation among Chinese college students. J of Pub Health and Prev Med,
26(4), (2015), 74-77. (In Chinese)

PT
Yang et al., 2003 T.Z. Yang, L. Huang, Z.Y. Wu. Study on the appropriateness of the
Chinese version of the General Health Questionnaire as a screening instrument for

RI
psychological disorders in mainland China. Chin J Epid, 4, (2003), 769773. (In
Chinese)

SC
Yang et al., 2015 L.S. Yang, Z.H. Zhang, L. Sun, Y.H. Sun, D.Q. Ye. Prevalence of
suicide attempts among college students in China: a meta-analysis. PLoS One, 10(2),

U
(2015), e0116303.
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Table 1. Comparisons between suicidal ideators and students without suicidal ideation
Suicidal ideators Students without suicidal

PT
ideation F/OR (95% CI) P Estimates of effect size
(n=1,008) (n=18,330)

RI
Age (years) 18.700.85 18.730.83 1.06 =0.30 partial 2< 0.01
Gender (Female) 60.8% (613) 46.8% (8577) 1.77 (1.55-2.01) <0.01 =0.06

SC
GHQ-12 positive screening 37.0% (373) 14.4% (2633) 3.50 (3.06-4.01) <0.01 =0.14
Family history of suicide (yes) 8.6% (87) 3.4% (623) 2.69 (2.13-3.39) <0.01 =0.06

U
BIS-11
Total score positive screening 40.7% (410) 26.6% (4884) 1.89 (1.66-2.15) <0.01 =0.07

AN
High cognitive impulsivity 33.1% (334) 29.5% (5406) 1.19 (1.04-1.36) =0.01 =0.02
High motor impulsivity 47.8% (482) 26.2% (4809) 2.58 (2.27-2.93) <0.01 =0.11

M
High non-planning impulsivity 32.6% (329) 28.4% (5208) 1.22 (1.07-1.40) <0.01 =0.02

D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Table 2. Comparisons between suicide attempters and suicidal ideators without suicide attempt

PT
Suicide attempters Suicidal ideators without
suicide attempt F/OR (95% CI) P Estimates of effect size

RI
(n=161) (n=847)

Age (years) 18.730.89 18.700.85 0.22 =0.64 partial 2< 0.01

SC
Gender (Female) 64.3% (102) 60.3% (511) 1.14 (0.80-1.61) =0.48 =0.02
GHQ-12 positive screening 39.1% (63) 36.6% (310) 1.11 (0.79-1.57) =0.54 =0.02
Family history of suicide (yes) 18.0% (29) 6.8% (58) 2.99 (1.85-4.84) <0.01 =0.15

U
BIS-11

AN
Total score positive screening 44.7% (72) 39.9% (338) 1.22 (0.87-1.71) =0.26 =0.04
High cognitive impulsivity 31.1% (50) 33.5% (284) 0.89 (0.62-1.28) =0.58 =-0.02
High motor impulsivity 57.1% (92) 46.0% (390) 1.56 (1.11-2.20) =0.01

M
=0.08
High non-planning impulsivity 32.3% (52) 32.7% (277) 0.98 (0.69-1.41) =1.00 <-0.01

D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Table 3. Logistic regression equations for suicidal ideation and suicide attempt
Risk factors for suicidal ideation Risk factors for suicidal attempt

PT
(among suicidal ideators)

RI
OR (95% CI) P OR (95% CI) P
Gender (Female) 1.58 (1.38-1.80) <0.01 NA

SC
GHQ-12 positive screening 3.06 (2.67-3.51) <0.01 NA
Family history of suicide (yes) 2.40 (1.88-3.05) <0.01 2.99 (1.84-4.86) <0.01
BIS-11

U
Total score positive screening 1.54 (1.35-1.76) <0.01 NA

AN
High cognitive impulsivity NA NA
High motor impulsivity 2.58 (2.27-2.93) <0.01 1.57 (1.11-2.21) =0.01
High non-planning impulsivity NA NA

M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC

Figure 1.Sampling profile.