Objective: This study aimed to establish and assess the Back Propagation Neural Network (BPNN) prediction model for suicide attempt, so as to improve the individual prediction accuracy. Method: Data was collected from a wide range case-control suicide attempt survey. 659 serious suicide attempters (case group) were randomly recruited through the hospital emergency and patient registration system from 13 rural counties in China. Each case was matched the control by the same community, gender, and similar age (±2 ages). Face to face interviews were conducted for each subject with the structured questionnaire. Logistic regression was applied to preliminarily screen the factors and BPNN was used to establish the prediction model of suicide attempt. Results: Multivariate logistic regression indicated that family history of suicide (OR=4.146), mental problem (OR=3.876) Low education level, poor health, aspiration strain, hopelessness, impulsivity, depression are the risk predictors and social support, coping skills, healthy community are the protect predictors for suicide attempt. Repetitious data simulation process of BPNN indicated that three-layer BPNN with 9 hidden layer neurons is the optimal prediction model. The sensitivity (67.6%), specificity (93.9%), positive predictive value (86.0%), negative predictive value (84.1%), total coincidence rate (84.6%) all manifested that it is excellent to distinguish suicide attempt case. Conclusions: The BPNN method is applicative, feasible, credible and good discriminative effect for suicide attempt. The BPNN established has significant clinical meaning to distinguish suicide attempt for the clinical psychiatrist and lay theoretical foundation for artificial intelligence expert assisted diagnosis system for suicide attempt in the future.

Background There has been much literature on schizophrenia, but very little is known about the characteristics of suicides with schizophrenia in comparison with the suicides with other diagnosed psychiatric disorder and without psychiatric disorders. Methods Thirty-eight suicides with schizophrenia, 150 suicides with other psychiatric disorder, and 204 suicides without a psychiatric disorder were entered in current study. Psychological Autopsy (PA) was applied to collect information of the suicides. Social demographic factors and clinical characteristics of the suicides were measured. The well validated standard scales were applied: Beck Hopelessness Scale (BHS), Landerman’s Social Support Scale (DSSI), Dickman’s Impulsivity Inventory (DII), Spielberger State-Trait Anxiety Inventory (STAI) and Hamilton Depression Scale (HAMD). Suicide intents were appraised by the Beck Suicide Intent Scale (SIS). The SCID based on the Diagnostic and Statistical Manual of Mental Disorders-IV (DSM-IV) was applied to assess the psychiatric status of individuals. Demographic characteristics, clinical characteristics, method of suicide and suicide intents of suicides were compared among the three groups (Schizophrenia group, Other psychiatric disorders group, and None psychiatric disorders group). Results There were 9.7% of suicides who suffered schizophrenia. The current study found being female was the risk factor for suicides with schizophrenia in rural China, which was opposite to the previous studies. The suicides with psychiatric disorder scored higher on hopelessness, anxiety, and depression, but lower on social support and impulsivity than suicides without psychiatric disorder. The suicides with psychiatric disorder were less impulsive than none psychiatric disorders group, too. The schizophrenia group did not show more violence than other psychiatric disorders group. Conclusions This research compared the demographic characteristics, clinical characteristics, method of suicide and suicide intents among the suicides with schizophrenia, with other diagnosed psychiatric disorder and without psychiatric disorders. The result indicated that each groups showed their unique characteristics, which gave us new viewpoints to control and prevent the prevalence of suicides according to their different characteristics.

The detection and analysis of cases of low medication adherence is important for helping to control tuberculosis (TB). The purpose of this study was to detect low adherence in rural TB patients by using the eight-item Morisky Medication Adherence Scale of Chinese version (C-MMAS-8) and to further analyze the adherence-related variables. A total of 358 rural TB patients recruited through multi-stage randomized sampling participated in the survey. Data were collected by the use of interviewer-led questionnaires. First, the reliability and validity of the C-MMAS-8 were determined. Second, the adherence level was assessed, and factors related to low adherence were analyzed by using Pearson’s chi-square test and then in multiple logistic regression model. Finally, the prediction of the logistic model was assessed with Receiver Operating Characteristic (ROC) curves. The C-MMAS-8 could be used to detect low adherence in TB patients with good reliability and validity. By using the referred cutoff points of MMAS-8, it was found that more than one-third of the participants had low medication adherence. Further analysis revealed the variables of being older, a longer treatment time, and being depressive were significantly related to low adherence. The ROC of the model was assessed as good using the cutoff point. We conclude that appropriately tailored strategies are needed for health-care providers to help rural TB patients cope with low medication adherence.

No abstract