Citation

Lee JI, Lee MB, Liao SC, Chang CM, Sung SC, Chiang HC, Tai CW. J. Formos. Med. Assoc. 2010; 109(2): 138-147.

Affiliation

School of Medicine, Kaohsiung Medical University College of Medicine, Kaohsiung, Taiwan.

Copyright

(Copyright © 2010, Scientific Communications International)

DOI

10.1016/S0929-6646(10)60034-4

PMID

20206838

Abstract

BACKGROUND/PURPOSE: Suicide is an important public health problem and one of the leading causes of death worldwide. The present study investigated the prevalence of suicidal ideation (SI) and its associated risk factors in the general population. METHODS: A nationwide community survey was conducted using a computer-aided telephone interview system with residents aged >/= 15 years, who were selected by a stratified, proportional randomization method. The questionnaire comprised demographic variables, five items of psychopathology selected from the Brief Symptom Rating Scale (BSRS-5) and questions about personal experience with suicide. In total, 2054 respondents, 1002 male (48.8%), and 1052 female (51.2%), completed the survey. RESULTS: The weighted prevalence of SI was 2.84% in the past week, 5.50% in the past year, and 18.49% during a lifetime. Significant risk factors for SI in the last week included presence of SI over the past year [odds ratio (OR) =1763.6], SI during the lifetime (OR =267.6), psychiatric morbidity (OR = 30.3), depression (OR =26.1), inferiority (OR =11.2), hostility (OR = 10.9), anxiety (OR = 10.5), insomnia (OR =6.7), history of seeking help for psychological distress (OR = 7.9), divorce (OR =6.4), unemployment (OR = 5.0) and having suicidal behavior in relatives or friends (OR =3.8). Stepwise multiple regression analysis demonstrated that the five symptom items of BSRS-5 and unemployment significantly predicted 25.3% of the variance of SI. Using the BSRS-5 score 3 or 4 as a cut-off to predict SI, the rate of accurate classification was 85.88%, with sensitivity of 0.83 and specificity of 0.86. CONCLUSION: A telephone interview survey containing the BSRS-5 items is an efficient way to identify determinants of SI in the general population.

Language: en