| Objectives:1 To conduct an epidemiological investigation in patients with major depressive disorder(MDD).2 To explore how suicidal ideation(SI)and quality of life(QOL)in patients with MDD were influenced by socio-demographic factors,depressive symptom severity,comorbidity,social support,personality characteristics,and sleep quality.3 To probe into the influence factors of suicidality and predict suicide risk in patients with MDD by constructing the model of decision tree.4 To propound comprehensive prevention and intervention measures which are specific to QOL and SI in patients with MDD.Methods:1,540 respondents were selected randomly from the patients diagnosed depressive disorder or emotional disorders registered in database of psychopath information in Wuhan.After having been tested with Hospital Anxiety and Depression Scale(HADS),patients whose HADS scores were higher than 8 points and were willing to participate in this research were selected.The respondents were required to complete self-made general situation questionnaire,Patient Health Questionnaire depression scale(PHQ-9),Eysenck personality questionnaire(EPQ)adult version,Social Support Rating scale(SSRS),Pittsburgh Sleep Quality Index(PSQI)and World Health Organization Quality of Life scale short version(WHOQOL-BREF).Finally,667 respondents who have both completed all self-reported questionnaires eligibly and were diagnosed as MDD by psychiatric experts with secondary senior positions or above,according to the fourth edition of Diagnostic and Statistical Manual of Mental Disorders in United States(DSM-IV),were included.The data collected from questionnaires was analyzed with SPSS 18.0 statistical software through the chi-square test,t-test,Pearson correlation analysis and multiple linear regression,the non-conditional Logistic regression and Exhaustive Chi-Square Automatic Interaction Detection algorithm,and so on,and used for construction of decision tree modelResults:1.Basic demographic characteristics of patients with MDDThe 667 subjects included in this research consisted of 301 men and 366 women.The average age was 31.0±10.9 years.529 subjects were under 40 years old,accounting for 79.31%.84.01%of the subjects had educational background above technical secondary school or high school.242(36.28%)subjects in other career included students and unemployed,etc.65.17%subjects had monthly family income over 3,000 RMB.2.Comorbidity of MDD with generalized anxiety disorder(GAD)compared with MDD aloneThe former was heavier in depression severity(χ2=5.13,P<0.001),poorer in overall life quality(t=2.50,P=0.013)、general health(t=4.93,P<0.001)、quality of life in physical(t=2.23,P=0.026)and psychological(t=4.04,P<0.001)domains,and higher in SI(χ2=6.04,P=0.014)than the latter.3.MDD patients compared with normal subjectsCompared with Chinese norms,QOL in MDD patients was worse,the total score of social support was lower,neuroticism and psychoticism scale were scored higher,but the score in extraversion scale was lower,the score of seven factors in sleep quality and total scores of PSQI were higher than normal adults.All the difference had statistical significance(P<0.05).SI in MDD patients(76.76%)was much higher than normal adults.4.The influence factors of life quality in patients with MDD(1)Women had significantly higher overall life quality than men(β=0.14,P<0.001).Monthly Family earning was positively associated with overall life quality(β=0.10,P=0.007)and the life quality in environment domain(β=0.12,P=0.002).(2)The extraversion-introversion was positively related with overall life quality(β=0.14,P<0.001),the life quality in physical(β=0.13,P=0.001),psychological(β=0.18,P<0.001),environment(β=0.09,P=0.028)and social(β=0.22,P<0.001)domains.The psychoticism was negatively related with the life quality in physical(β=-0.11,P=0.002),psychological(β=-0.08,P=0.023)and environment(β=-0.14,P=0.001)domains.The Neuroticism was negatively related with the life quality in psychological(β=-0.09,P=0.015)field.(3)The subjective social support was positively related with overall life quality(β=0.15,P<0.001)and the life quality in physical(β=0.10,P=0.021),psychological(β=0.09,P=0.022),and social(β=0.24,P<0.001)domains.The objective social support was positively related with life quality in environment(β=0.09,P=0.031)domain.(4)Depression severity was negatively related with the life quality in all(P<0.001)fields,except for environment field.(5)The comorbidity status was negatively correlated with overall life quality(β=-0.09,P=0.023),the general health(β=-0.19,P<0.001)and the life quality in psychological(β=-0.11,P=0.004)and physical(β=-0.08,P=0.041)domains.(6)The sleep time was negatively correlated with overall life quality(β=-0.11,P=0.036);Sleep disturbance was negatively correlated with overall life quality(β=-0.11,P=0.046),the general health level(β=-0.09,P=0.019)and the life quality in psychological(β=-0.08,P=0.019)domains.Use of hypnotic was negatively correlated with the life quality in physical(β=-0.08,P=0.039)domain.(7)The SI was negatively correlated with the life quality in environment domain(β=-0.11,P=0.004).5.The influence factors of SI in patients with MDDWomen(OR=1.65,95%CI=1.11～2.43),psychoticism(OR=1.08,95%CI=1.00～1.16),severity of depression(OR=1.14,95%CI=1.07～1.21),comorbidity(OR=l.64,95%CI=1.11～2.42),the life quality in environmental fields(OR=0.98,95%CI=0.97～1.00).6.The construction and analysis of decision tree model according to the SI in patients with MDD(1)The construction of decision tree modelWith SI-related factors of MDD patients served as predictive variables according to the results of single factor analysis,and the SI as a target variable,we constructed a model of decision tree for SI.Segmentation validation was adopted,and the ratio of sample size of training sample and test sample was set at 7:3.The tree has three levels,including six terminal nodes,four of which have statistical significance.With a correct predicting percentage of 80.90%in test samples,the model was confirmed effective after the evaluation.(2)The analysis of the decision tree modelSeverity of depression was located in the first level of the decision tree,the sleep quality and life quality of environmental field in the second floor,women and comorbidity in the third.There were interactions between life quality of environmental field and depression severity,comorbidity,and the sleep quality and depression severity,women,and sleep quality and depression severity.There were four paths predicting suicidality in patients with MDD in the model.Conclusions:1.Compared with normal subjectsOur findings demonstrate that MDD patients have poorer QOL,higher probability of suicidal ideation,less social support,and worse sleep quality.They typically have higher psychoticism and neuroticism,and partial introversion.2 The influence factors of life quality in MDD patientsOur findings surgest that women,high family income,the relatively extroverted personality,good subjective and objective social support are protective factors for life quality in MDD patients.Severity of depression,less sleep time,existence of sleep disturbance,use of hypnotic drug,MDD comorbid with anxiety disorder,existing suicidal ideation and higher psychoticism and neuroticism of personality are risk factors for QQL in MDD patients.3 The influence factors of SI in MDD patientsOur findings supporte that Women,exorbitant psychoticism,severity of depression,poor sleep quality and comorbidity of anxiety and depression are risk factors for suicidality in MDD patients,whereas good QOL in environmental domain is a protective factor.4 Results of the decision tree model analysisDepression severity of MDD patients has the greatest influence on SI;sleep quality and environmental life quality take the second place;and the influence of comorbidity and gender are minimal.There is obvious interaction between various factors.The decision paths predicting SI in MDD patients,in descending forecasting order,as follows:severe depression severity and poor life quality in environmental field,moderately severe or more less depression severity,poor sleep quality and comorbidities;severe depression severity but good life quality in environmental field,moderately severe or more less depression severity,normal sleep quality but the female patients.The decision tree model can be used to evaluate the risk of suicide and take pertinent intervention,according to the characteristics of the suicide patients.However,these results warrant futher investigation.Innovation points of paper1.Method innovation(1)We analyzed the data with Logistic regression analysis and decision tree analysis at the same time in this research,which making up for the disadvantage of single methodology analysis.(2)We could not only explore the influencing factors and the interaction among different affecting factors,but also predicted risk by constructing model of decision tree,which provided a new research method for similar studies.2.Content innovation(1)to explore comprehensively the influence factors of life quality and SI in patients with MDD from multiple perspective of psychology,physiology and society,in which the discovery that the life quality in environment domain had a prominent effect on SI and interacted with the severity of depression was new,the research results had important reference value for suicide intervention and prevention.(2)The suicidal characteristics of patients with MDD determined according to the decision paths,had important practical significance for evaluation of MDD suicide and recognition the high-risk of suicide. |
PDF Full Text Request