Trait impulsivity in suicide attempters: Preliminary study


Chiaki Kawanishi, MD, PhD, Health Management and Promotion Centre, Yokohama City University Graduate School of Medicine, 22-2 Seto, Kanazawa-ku, Yokohama 236-0027, Japan. Email:


Suicide attempt is a risk factor for suicide. To investigate trait impulsivity among suicide attempters, 93 attempters admitted to an emergency department and 113 healthy controls were evaluated using the Japanese version of the Barratt Impulsiveness Scale (BIS-11J). Impulsivity was analyzed in relation to clinical data in the attempters. Total BIS-11J, attention impulsiveness, and motor impulsiveness scores were significantly higher in the attempters than in the controls. Both total BIS-11J and non-planning impulsiveness scores were significantly higher in attempters with schizophrenia and other psychotic disorders among the diagnostic groups. Control of impulsivity should be considered as one of the targets for suicide prevention.

SUICIDE ATTEMPT IS a potent risk factor for completed suicide.1–3 In approximately 45% of suicide cases a history of unsuccessful attempt was found.4 In Japan, 41% of medically serious suicide attempters have a history of previous suicide attempt.5 An investigation of those who attempt suicide could yield information important for improving prevention and follow-up treatment.

Previous reports outside Asia have shown that impulsivity is associated with suicide attempt.6–8 In the present preliminary study, we investigated trait impulsivity in suicide attempters admitted to an emergency department in Japan.



Participants were suicide attempters admitted to the Critical Care and Emergency Medical Center, Yokohama City University Medical Center between 7 May 2003 and 11 October 2007. This center is one of the four tertiary emergency medical centers in Yokohama, Japan, with a population of 3.5 million people. Of the 544 suicide attempters admitted, 99 agreed to participate and 95 completed the questionnaire. Their intent to die was confirmed. The remaining patients were not asked to participate because of early death, transfer to another unit with prolonged consciousness disturbance, or an extremely short hospital stay. Primary psychiatric diagnosis was made by two trained psychiatrists according to DSM-IV criteria, and patients without axis I or axis II diagnosis or with mental retardation were excluded.

As a result, the data for 93 participants (34 male, 59 female; mean age, 37.7 ± 15.7 years) were used for analysis. Sex and age did not differ significantly from the attempters not participating. The most common axis I diagnosis was mood disorder (n = 27), followed by anxiety or stress-related disorder (n = 24), schizophrenia and other psychotic disorder (n = 17), substance-related disorder (n = 8), eating disorder (n = 2), pathological gambling (n = 1), dementia (n = 1), and gender identity disorder (n = 1); no diagnosis was made in 12 subjects. Twenty-six percent of subjects were diagnosed with personality disorder.

One hundred and thirteen healthy medical personnel (35 male, 78 female; mean age, 33.6 ± 10.1 years) comprised the control group. Mental illness was ruled out among them through interviews and use of the Mini-International Neuropsychiatric Interview.9 Individuals with mental retardation were excluded. None of the controls had a previous history of suicide attempt.

There was no significant difference in sex between the attempters and controls, but age was significantly higher in the former group.


Trait impulsivity was evaluated using the Japanese version of the Barratt Impulsiveness Scale Version 11 (BIS-11J).10 The original BIS, a self-administered test, was designed to measure impulsiveness as a personality trait.11 It consists of 34 items that measure impulsiveness within the three categories of attention impulsiveness, motor impulsiveness, and non-planning impulsiveness. Attention impulsiveness assesses task at hand, thought insertions, and racing thoughts (eight items); motor impulsiveness assesses acting on the spur of the moment and perseverance (11 items); and non-planning impulsiveness assesses planning and thinking carefully, and enjoyment of challenging mental tasks (11 items). The validity and reliability of the BIS-11J has been previously established.10

The BIS-11J was handed out in the emergency department after the patients recovered from serious injury or decreased level of consciousness. The control group also completed the BIS-11J.


The protocol of this study was approved by the Ethics Committee of Yokohama City University School of Medicine, and written informed consent was obtained from all participants.

Statistical analysis

Total and category scores for the BIS-11J and demographic data were analyzed for suicide attempters and controls, and differences between the two groups were determined on t-tests, ANOVAs, and Tukey HSD tests. The significance level was set at P < 0.05 (two-tailed test). Statistical analysis was performed using SPSS 11.0J for Windows (SPSS, Tokyo, Japan).


Table 1 lists the BIS-11J total, attention impulsiveness, and motor impulsiveness scores; these were significantly higher in the attempters than in the controls.

Table 1. BIS-11J score vs presence of suicide attempt
 Attempters (n = 93)Controls (n = 113) t d.f. P
  1. BIS-11J, Japanese version of the Barratt Impulsiveness Scale version 11.

Total BIS-11J score68.8411.3262.278.214.671164<0.001
 Attention impulsiveness16.893.8514.673.004.539172<0.001
 Motor impulsiveness23.995.6420.853.404.707145<0.001
 Non-planning impulsiveness27.964.9526.753.941.9031740.059

Total BIS-11J scores were then analyzed among the attempters in each diagnostic group using anova (Table 2). The attempters with schizophrenia and other psychotic disorder had significantly higher scores for total BIS-11J and non-planning impulsiveness than attempters with anxiety or stress-related disorder (P = 0.030 and P = 0.004, respectively, Tukey tests). Attempters with personality disorder had a marginally higher motor impulsiveness score than those with mood disorder, and anxiety or stress-related disorder (P = 0.051 and P = 0.052, respectively, Tukey tests).

Table 2. BIS-11J score vs diagnostic category in suicide attempters
 Schiz (n = 17)Mood (n = 27)Anx (n = 24)PD (n = 12)F P
  1. Anx, anxiety or stress-related disorder; BIS-11J, Japanese version of the Barratt Impulsiveness Scale version 11; Mood, mood disorder; PD, personality disorder; Schiz, schizophrenia and other psychotic disorder.

Total BIS-11J score74.5310.2567.3012.2964.7111.0872.507.803.2930.025
 Attention impulsiveness18.654.2616.703.8816.084.4417.332.101.4960.222
 Motor impulsiveness24.656.3322.675.8222.584.9027.674.602.8760.042
 Non-planning impulsiveness31.243.4227.934.8726.045.0327.504.746.3320.008

There was no sex difference in regard to impulsivity among either the attempters or controls.


This study has identified trait impulsivity in medically serious suicidal attempters. Impulsivity has been investigated in psychiatric patients according to psychiatric diagnosis,12,13 and a few researchers outside Asia have focused on impulsivity in suicide attempters.6–8 Baca-Garcia et al. reported significantly high impulsivity in Spanish suicide attempters using the BIS.6

In the present study, total BIS-11J and non-planning impulsiveness scores were significantly higher in the suicide attempters with schizophrenia and other psychotic disorder according to ANOVAs. Previous studies have suggested that a lack of hesitation leads to suicide completion in patients with schizophrenia.14,15 The Nakagawa et al. detailed comparative analysis of suicide attempts in patients with schizophrenia spectrum disorder and those with affective disorder found that the former group had frequently used a highly lethal method.16 Impulsivity could be behind those findings.

Baca-Garcia et al. found that male gender and borderline personality disorder were associated with impulsivity on logistic regression analysis.6 In the present study, motor impulsiveness score was marginally higher in attempters with personality disorder than in those with mood disorder or anxiety or stress-related disorder.

This preliminary study has some limitations. First, the attempters were recruited from one urban emergency department in Japan. Second, the sample size was relatively small. Last, this study did not investigate impulsivity in patients without suicidality in comparison with the attempters. Future larger-scale and detailed studies are warranted to verify the present findings.

The present findings, although limited, suggest that impulsivity is an important factor in suicide behavior in suicide attempters, and therefore assessing and controlling impulsivity should be considered in suicide prevention.


This study was supported by a Health and Labour Science Research Grant from the Ministry of Health, Labour, and Welfare of Japan.