| BackgroundThe psychosocial autopsy is an integrated retrospective research methodology for reconstructing the suicide's state of personal and social circumstance from relative detailed information gained through interviews conducted with informants. Suicidal behaviors are the most complicated ones in human society. Studies on risk factors of suicide have been the interesting topics in the domains of medicine and sociology. And it is an important aspect of suicide prevention. However, these studies have not been well carried out, because it is difficult to gather the real personal information for suicides after committing suicide. Also, the people aged between 15 and 35 years are the peak groups of suicide in our country. The suicide death is the first cause of death among these subgroups. Therefore, using psychosocial autopsy method, this paper presents the synthetic results of risk factors on suicide including the socio-demographic variables, physical diseases and mental disorders, personality characteristics, psychosocial stress, as well as family and community environment.ObjectivesTo describe the characteristics of suicidal behaviors by the people aged 15～35 in rural Hunan, to explore the suicide relative risk in relation to socio-demographic variables, physical diseases and mental disorders, personality characteristics, psychosocial stress, as well as family and community environment, to compare the relative risk of committing suicide between males and females, and to provide scientific bases for suicide prevention and intervention in rural communities.Methods Subject and Method According to the principle of a simple random sampling, we randomly selected five counties from 88 counties in Hunan Province and conducted a psychosocial autopsy study for 121 suicides confirmed by the local hospital or the local public security department from the selected five counties. Subjects of this study included those who aged between 15 and 35 years extracted from the death registration system of the Center for Disease Control and Prevention in these five counties and coded as X68 X84, Y89 in the International Classification of Diseases (10th Edition ). The death date of these suicide cases was from September 1, 2005 to March 31, 2008. We randomly selected 121 live community samples as a control group of similar age (±1 year old) and the same gender. And the control samples resided in the neighboring village where suicide cases lived. Similar to most case-control psychological autopsy method, we respectively interviewed four informants for each paired samples and the semi-structured interviews lasted from 2 hours to 4 hours in order to detailedly gain the information of suicides and control samples. The intimate informants were family members and friends or neighbors of samples and consented to participate in the study. This study was approved by the ethic committee of the Central South University.Contents and InstrumentsWe classified the examined factors into five broad domains: general information, Suicide behavioral characteristics, psychosocial stress, personality characteristics and diagnosis of mental disorders. General information investigated included: personal information including age, gender, occupation, educational status, income, marital status, living arrangement, reproductive status, religious belief, history of suicide attempt, presence of severe or chronic illnesses, history of psychiatric disorders and previous psychiatric treatment; family circumstances including family economic status, family relationships, other family member's physical disease, mental disorders and drug abuse,gambling, alcohol abuse, prostitution and crime behavior problems,family history of suicide and family storage of pesticides; communitycircumstances including receiving education problem, traffic problem,medical care problem, income inequality, drug addiction, alcohol abuse,gambling, superstition, crime, and behavioral problems of abusing aspouse or child.Suicidal behavior characteristics of completed suicides investigatedincluded: the suicide date, time, location, methods of suicide, isolation ofsuicide, communication of suicide ideation before attempt, suicide plans,final arrangements, suicide note, letter, or diary entry, number of suicideattempts in lifetime, methods of suicide attempts, as well as thepre-hospital treatment situation of suicides.Psychosocial stress tested using the Paykel Life Events Scale and theDuke Social Support Index.Personality traits assessed using the Dickman Impulsivity Scale, Moos'sCoping Response Inventory, Beck Hopelessness Scale and theSpielberger's Trait Anxiety Inventory.Psychiatric factors investigated including presence of psychiatricdisorders at the time of death and damage extent of psychological, socialand vocational function using the Structured Clinical Interview forDSM-IV Axis I Disorders, 2001, Research Version and the GlobalAssessment Function Scale for DSM-IV axis V.Statistical AnalysisMultivariate conditional logistic regression model analysis was applied toestimate the relative risk on suicide using SAS software 9.13 and PhRegprocedure, controlling for confounding factors as well as the interactionof age and gender.Results 66.1% (80/121) of the cases who committed suicide by taking pesticide which was the virulent methamidophos most commonly used in the deceased group, 9.9% (12/121) hanging, 7.4%(9/121) drowning and 5.0%(6/121) jumping. In all suicides aged 15～35, no significant difference was found in the suicide patterns between males and females suicides (x2=3.928, P=0.560). Suicide occurred at a peak from May to July, especially with the central tendency of date on the forth of months in female suicides (r=0.305, P<0.01) , while such tendency was not the case in males (r=0.061, P>0.05). The female suicides had a peak suicide time at 10 am (r=0.277, P<0.01) while male at 4 pm (r=0.202, P<0.01). The majority (n = 69, 57.0%) of the subjects died at home, and more than half of them (n =72, 59.5%) committed suicide without somebody nearby or in visual.29.8% (36/121) suicides ever explicitly communicated own suicidal ideation with their relatives or friends prior to the last attempt through conversation or telephone. 32.2% (39/121) of them made definite arrangements before attempt. 57% (69/121) suicides had a suicide plan and thirty-four (28.1%) out of 121 cases left at least one suicide note in the form of message, letter, or diary entry to someone. 15.0% (18/121) of these cases ever attempted suicide, 4 of whom attempted more than twice, and the female suicides had higher frequency of suicide attempt than males(x2=4.998, P=0.0025).Sixty-two (51.2%) out of 121 cases were not sent to hospitals for being dead when found, and another 59 cases were transferred for first aid while 10 of them died on the way to hospitals. The median time of getting first aid for the fourty-nine alive cases when found was 30 minutes. Psychiatric illnesses were more frequent among the suicides than in the controls. 63 individuals who committed suicide met the DSM-IV diagnostic criteria. The overall prevalence of mental disorders was 52.1% (63/121). There was no significant difference for prevalence rate between male and female suicides (x2 = 1.357, P = 0.244). In the suicides group, the most common diagnoses were mood disorders (27.3%), schizophrenia (19.8%), and substance use (5.8%). In the control group, only 4 controls had a diagnosed psychiatric disorder. The overall prevalence of mental disorders was 3.3% (4/121). And there was no significant difference for prevalence rate between male and female controls (P = 1.000).85.7% of the rural suicides suffered one or more negative life events, and the average frequency of each person who committed suicide was 2.6. The three most frequent negative life events among the 121 people who killed themselves were serious or fatal illness (28.1%), serious disruptive arguments with his/her spouse (21.5%) and economic difficulties (17.4%). 32.2% (39/121) of suicides experienced acute negative life events before committing suicide, and mainly concerning to love affairs or marriage(12.4%). And 54.5% (66/121) of suicides experienced chronic negative life events before committing suicide, and mainly concerning to health issues. The total score for negative life events stress among the suicides was 16.9±3.0, 16.9±3.2 for males and 17.0±3.5 for females, and no significant difference was found between males and females (Z=-1.257, P=0.209). 20.7% (40/121) in control group suffered negative life events and the three most frequent negative life events among the 121 controls were family members having got serious or fatal disease (5.8%), severe loss of property and house (3.3%) and having serious or chronic diseases themselves (3.3%). 4.1% (5/121) of controls experienced acute negative life events before interview, and 16.0% (20/121) had had main health-related chronic negative life events. The total score for negative stress of life events was 1.4±3.0, and there was no significant difference between males and females (Z=0.577, P=0.676).Controlling for social demography variables, health-related factors,psychosocial stress, personality characteristics, family factors,community factors, as well as the interaction of age and gender, themultivariate conditional logistic regression analysis revealed somesignificant suicide risk factors as below:The education status was significantly related to suicide. Compared topeople with Senior high school education and above, the relative risk ofsuicide was 3.89 (95%CIs: 1.61-7.81) among those who received primaryeducation.A low family annual income increased suicide risk. Compared to a familyannual income lower than 5,000 yuan, a family annual income more than20,000 yuan markedly reduced suicide risk (OR=0.49, 95%CIs:0.19-0.95).This study found that a serious or chronic illness had a significant effecton increasing suicide risk (OR=2.52, 95%CIs: 1.50-40.12). And, a mentaldisorder had a strong suicide risk (OR= 5.02, 95%CIs: 1.52-48.16).Compared to the group scoring less than 5 for negative stress, the groupscoring 5～9 increased suicide risk for the stress of marriage, family,health, work or study, law-related and other negative life events(OR=9.01,95%CIs:2.68-34.22; OR=5.99, 95%CIs:2.68-34.22; OR=11.33, 95%CIs:2.91-34.35; OR=2.33, 95%CIs : 1.98-16.75; OR=2.09, 95%CIs :2.67-34.11, respectively) . Yet, there was the case in the group scoringover 15. In addition, those who ever experienced 1 negative life event hada significantly higher suicide risk than those who never did (OR=17.90,95%CIs: 3.32-96.40), however, the odds ratio dropped for those whoexperienced 3 or more (OR=3.75, 95%CIs: 1.98-35.13).Suicide risk was lower among those who scored 40 or more in social support assessment than those who scored less than 30 (OR=0.02, 95%CIs: 0.01-0.22).The positive coping skills on positive judgment, seeking support and guidance and problem solving had the strong protective effect on suicide. Compared with those who scored less than 5 for assessment of coping skills, those who got 10 or more markedly reduced suicide risk for positive judgment, seeking support and guidance and problem solving (OR=0.10, 95%CIs: 0.01-0.68; OR=0.03, 95%CIs: 0.01-0.25; OR=0.02, 95%CIs: 0.01-0.45, respectively).The dysfunctional impulsivity personality was significantly related to suicide behavior. Those who scored 6 or more had higher risk for suicide compared to those scored under 4 (OR=14.87, 95%CIs: 3.11-71.13), and the dysfunctional impulsivity personality had a stronger impact on suicide risk in females than in males (P<0.01).A high level of hopelessness was the risk factor associated with the highest relative risk and the highest attributable risk for suicide among the people aged 15～35. Compared to the group scoring less than 6 for the level of hopelessness, the group scoring more than 11 had a particularly high risk of suicide (OR=26.01, 95%CIs: 4.35-155.37 ), but no interaction of gender was statistically concluded (P>0.05).Conclusions:This study suggests that at least four evidence-based interventions, which are strengthening the management of pesticides, improving the identification and discovery of suicide signals among the public in rural areas, enhancing the pre-hospital treatment capacity and strengthening identification and management of the patient with mental disorders in the communities of rural Hunan. These seem to be appropriate strategies to reduce suicides aged 15～35 in rural Hunan. Our results lean strong support to a multi-factorial approach to the understanding and prevention of suicide for the youth by way of a range of multipronged and multi-levelled initiatives that aim at individual, family, and societal levels.
|
PDF Full Text Request