【作者】 徐广明；

【导师】 肖水源；

【作者基本信息】 中南大学 ， 精神病与精神卫生学， 2010， 博士

【摘要】 研究背景精神分裂症是最重要的公共卫生问题之一。在世界范围内,成年人群中本病的发病率在0.2‰-0.6‰之间,终生患病率接近1%,估计我国有精神分裂症患者780多万。精神分裂症具有病程慢性化、致残率高、预后不良的特点,是我国主要劳动力年龄段十大致残病种之一。自1951年氯丙嗪应用于临床后,抗精神病药物在很大程度上改善了精神分裂症的预后和生活质量,成为主要的治疗手段。在众多的抗精神病药物中,氯氮平是公认治疗阴性症状效果最好的抗精神病药物,对难治性精神分裂症、自杀和精神分裂症患者暴力行为有效,能降低患者住院率,减少治疗费用。但在国外研究发现,氯氮平可导致严重的血液系统不良反应,因而在20世纪80年代开始限制其使用,在美国,目前仅用于难治性精神分裂症。1978年我国开始生产氯氮平,因其疗效好,价格低廉,很快成为常用的抗精神病药物在国内广泛使用。在很长一段时期内,氯氮平是急性精神分裂症首选治疗药物。由于担心其不良反应,加上上个世纪90年代维思通等新型非典型抗精神病药物在中国上市,精神科医生的治疗选择增加,2003年中华医学会颁布的《中国精神分裂症治疗指南》(以下简称《指南》)中明确规定氯氮平不可作为一线抗精神病药物使用。尽管《指南》的颁布必然在很大程度上降低氯氮平的使用率,但该药仍然在我国有较广泛的使用,其主要原因有四个方面：①《中国精神分裂症治疗指南》对临床决策的影响有一定的滞后；②我国目前仍有很大比例的精神分裂症患者难以承受其它非典型抗精神病药物的治疗费用,由于治疗作用好,费用低廉,氯氮平仍然受到相当一部分患者及基层医生的欢迎；③国外研究表明,至少有30%的精神分裂症患者对其它抗精神病药物反应不佳,符合氯氮平治疗难治性精神分裂症的指征；④氯氮平是目前唯一被证实能够有效降低精神分裂症患者自杀率的抗精神病药物。已有的关于氯氮平使用情况的研究均是在《指南》颁布前或紧随其后进行的,可能没有反映出氯氮平使用变化的真实情况。并且,这些研究都是在不同时间、不同地点的横断面调查,数据间缺乏纵向可比性,我国至今没有研究过氯氮平在一个特定的地区使用情况的历史变化,以及接受氯氮平治疗者的临床特征。更为重要的是,国内没有对接受氯氮平治疗的精神分裂症患者的长期结局及其影响因素进行过系统的研究。研究目的本研究调查旨在通过对一个地区精神分裂症患者氯氮平使用情况的纵向研究和评估氯氮平治疗对精神分裂症结局的影响,为《中国精神分裂症治疗指南》的修订和精神分裂症防治政策制定提供科学依据。具体目的如下：1.调查1986-2006年间唐山地区全部住院精神分裂症患者氯氮平使用及其变化情况；2.分析住院精神分裂症患者接受氯氮平治疗的影响因素。3.对出院时氯氮平治疗的精神分裂症患者样本进行出院2年后结局随访,全面评价患者的临床症状、主观生活质量、总体功能状况；4.分析精神分裂症患者出院2年后结局的影响因素。研究方法第一阶段：1986-2006年唐山地区出院精神分裂症患者氯氮平使用率变化调查。调查唐山市全部6所精神病专科医院(3所县级医院,3所市级医院)1986,1996,2001和2006四个年份所有出院诊断是精神分裂症的病历。由经过培训的调查员用专门设计的病历信息调查表搜集病历资料。记录患者出入院的日期、出院诊断、住院号、社会人口学资料、医疗保险状况、病程、住院次数、出院时的药物种类和剂量、精神疾病家族史、既往自杀行为史等信息。氯氮平使用定义为患者出院时服用氯氮平治疗,可以是氯氮平单独治疗或与其它抗精神病药物联合治疗。精神分裂症诊断依据是住院病历上记载的临床医生的出院诊断。该地区精神科医生自1981年中国精神障碍分类第一版(CCMD-Ⅰ)出版发行后一直在使用CCMD系列的诊断标准。第二阶段：氯氮平对出院精神分裂症患者结局影响研究研究对象为2006年1月1日-12月31日在唐山市全部6所精神病院出院,符合DSM-Ⅳ精神分裂症诊断,在唐山本地居住,出院时单独服用氯氮平或单独服用其它抗精神病药物的患者。采用分层整群随机抽样方法获取出院时单独服用氯氮平的患者样本,按住院医院、性别和出院日期(±2天)匹配以出院时单独服用非氯氮平抗精神病药的患者作为对照。随访以面对面访谈方式进行,采用DSM-Ⅳ-TR轴I障碍临床定式检查(研究版)(SCID-Ⅰ)对患者重新诊断；出院病人随访表(自编)收集病人社会人口学资料和出院后康复治疗信息；家庭亲密度与适应性量表第二版中文版(FACESⅡ-CV).社会支持量表评估患者的社会支持和家庭氛围；精神分裂症阳性阴性症状量表(Positive and Negative Syndrome Scale, PANSS)、卡加利抑郁量表(Calgary Depression Scale for Schizophrenia, CDSS)、生命质量量表和总体功能评定量表(Global Assessment Function, GAF)评定患者的临床结局、生活质量和功能状态。结果㈠1986-2006年唐山地区出院精神分裂症患者氯氮平使用率变化四个调查年份6所医院总出院人数为6550人,总计出院7920人次。其中出院精神分裂症患者2717人,占总出院人数41.48%；出院精神分裂症患者人次为3195,占总出院人次的40.34%。1986年出院精神分裂症患者中氯氮平的使用率6.2%(排第七位)；1996年上升至45.2%,成为使用率最高的抗精神病药物；2001年继续上升到59.9%(首位)；2006年下降至35.7%,是继维思通之后第二位最常用的抗精神病药物。除1986年外,氯氮平是联合使用频率最高的抗精神病药物,与氯氮平合并使用频率最高的抗精神病药物是舒必利和维思通。出院时单独服用氯氮平的平均剂量由1986年的(263±106)mg/d上升到2006年(310±139)mg/d,氯氮平与其它药物联合使用的剂量在(173±131)mg/d-(207±120)mg/d之间。四个年份中出院时服用与未服用氯氮平的病例具有不同特征。服氯氮平的患者发病年龄(25±8)岁低于未服氯氮平的患者(27±9)岁,差异有统计学显著性(t=7.076,P<0.001)。服氯氮平患者中家庭经济状况差的比例高于未服氯氮平的患者(22.53%vs 16.51%,x2=18.215,p<0.001)；服氯氮平的患者中精神病家族史阳性比例高于未服氯氮平的患者(32.02%vs 26.43%,x2=11.286,P=0.001)；服氯氮平的患者中有过自杀行为的比例高于未服氯氮平的患者(19%vs 15%,g/=9.049,P=0.003)；服氯氮平的患者中再次住院的比例高于未服氯氮平的患者(39.6%vs 37%,x2=97.88,P<0.001)。县级医院出院患者中服氯氮平的比例高于市级医院(15.71%vs 10.31%,x2=20.48,P<0.001)；提示服用氯氮平的患者中具有不良结局预测因素的比例高,经济条件差。以是否使用氯氮平为因变量,对1996、2001和2006三个年份出院的精神分裂症患者分别进行logistic回归分析(α入=0.05,α出=0.10,P<0.05)结果表明,1996年再次住院(OR=1.100,95%CI：1.016-1.190)和发病年龄(OR=0.945,95%CI：0.926-0.964)两个变量进入方程,决定系数R2=0.09,即与首次住院的患者比较,再次住院的患者服用氯氮平可能性高；发病年龄小的患者服用氯氮平的可能性高。2001年仍然是再次住院(β=0.541,P=0.005,OR=1.717,95%CI：1.177-2.505)和发病年龄(β=-0.031,P=0.008,OR=0.970,95%CI：0.955-0.984)两个变量进入方程,决定系数R2=0.121。2006年有5个变量进入方程,分别是：自杀行为(OR=1.440,95%CI：1.025-2.021)；医院分级(OR=0.291,95%CI：0.201-0.422)；经济状况(OR=1.341,95%CI：1.003-11.972)；家族史(OR=1.476,95%CI：1.104-1.972)；发病年龄(OR=0.968,95%CI：0.953-0.983)；再次住院(OR=2.264,95%CI：1.561-3.283),R2=0.227。有过自杀行为者、在县级精神病院住院、家庭经济状况差的患者、家族史阳性、发病年龄小和再次住院的患者在出院时服用氯氮平的可能性大。结果㈡氯氮平对出院精神分裂症患者结局影响研究95例出院时单独服用氯氮平的精神分裂症患者出院后死于躯体疾病1例,自杀1例,地址和联系电话登记错误2例,拒绝参加随访13例。最终共有78例患者参加并完成随访调查。随访时经过DSM-Ⅳ定式检查符合精神分裂症或分裂情感障碍诊断76例,组成氯氮平组。在127例对照中,出院后自杀死亡1例,通讯地址和联系方式登记错误5例,拒绝参加随访30例,最终有91例患者参加并完成随访调查。经DSM-Ⅳ定式检查符合精神分裂症或分裂情感障碍诊断对照组85例,组成对照组。氯氮平组发病年龄21岁(QL18,Qu26)低于对照组24岁(QL19,QU31),差异有统计学显著性(Z=1.341 P=0.031)；氯氮平组住院次数5次(QL3,QU8)多于对照组3次(QL2,QU5),差异有统计学显著性(Z=2.772P=0.000)；氯氮平组病程13年(QL8,QU20)长于对照组6年(QL4,QUl4),差异有统计学显著性(Z=4.512 P=0.000)；氯氮平组本次住院时间45天(QL26,Qu158)长于对照组35天(QL21,Qu78),差异有统计学显著性(Z=2.262 P=0.021)。氯氮平组精神病家族史和自杀行为史阳性的比例均高于对照组,差异有统计学显著性。据上述资料可以认为氯氮平组病情比对照组严重。随访时氯氮平组53例(74.6%)患者仍服用氯氮平,而对照组仅有33例(38.8%)仍然服用出院时的药物；随访时氯氮平组患者有2例(2.6%)合并其它抗精神病药物,而对照组患者有21例(24.7%)合并其它抗精神病药物。两组随访时用药情况差异有显著性(x2=12.6P<0.001)。氯氮平组患者出院至随访期间有23例(30.3%)再次住院治疗,对照组40例(47.1%)再次住院治疗,差异有统计学显著性(x2=3.27P=0.032)。COX比例风险模型分析患者出院后再次住院的影响因素：出院时是否服用氯氮平(β=-0.74,P=0.01,HR=0.48)、发病年龄(β=-0.21,P<0.001,HR=0.812)、家庭居住地(县城与农村比较β=0.60,P=0.082,HR=0.553；城市与农村比较户-0.78,P=0.022,HR=0.462)和疾病病程(β=0.25,P<0.001,HR=1.281)进入方程。表明氯氮平组患者再住院可能性小、发病年龄小的患者再住院可能性大、农村患者与城市和县城的患者相比再住院可能性大以及病程长的患者再住院可能性大。用多元线性回归模型分析社会支持总分的影响因素：PANSS量表总分(β=-0.324,t=-4.382,P<0.001)和精神分裂症病程(β=-0.177,t=-2.390 P=0.018)进入方程。结果显示病人精神症状的严重性和病程两个变量与病人获得的社会支持呈负相关。两组患者家庭亲密度与适应性量表评分差异无显著性。多元线性回归模型分析家庭亲密度与适应性量表总分的影响因素：PANSS量表总分(β=-0.196,t=-2.522,P=0.013)进入方程。结果表明病人精神症状的严重性与家庭亲密度与适应性量表总分呈负相关。两组患者PANSS量表各项评分差异无统计学显著性。用广义线性模型(Ggeneralized Linear Model, GLM)分析患者随访时PANSS评分的影响因素：发病到首次就诊时间(月)(β=3.044,Z=3.430,P=0.001)、社会支持总分(β=-0.793,Z=-4.150,P=0.000)、家庭年人均收入(β=-0.91,Z=-3.290,P=0.001)、卡加利抑郁量表总分(β=1.180,Z=3.200,P=0.001)进入模型。表明发病到首次就诊时间长度和抑郁的严重程度与精神症状呈正相关、社会支持总分和家庭经济状况与精神病症状的严重程度呈负相关。两组患者随访时生活质量评分差异无统计学显著性。用GLM分析患者随访时生活质量的影响因素：发病到首次就诊时间(月)(β=-0.403,Z=-2.83,P=0.005)、社会支持总分(β=0.192,Z=6.27,P<0.001)、卡加利抑郁量表总分(β=-0.231,Z=-3.92,P<0.001)进入方程。表明发病到首次就诊时间、抑郁的严重程度两个变量与主观生活质量水平负相关、社会支持总分与主观生活质量水平呈正相关。两组患者随访时总体功能评分差异无统计学显著性。用GLM方法分析患者随访时总体功能的影响因素结果：家庭居住地(县城vs城市β=-4.94,Z=-2.29,P=0.021；农村vs城市户-3.70,Z=-1.92,P=0.048)、发病年龄(β=0.19,Z=2.192,P=0.029)、社会支持总分(β=0.37,Z=3.261,P=0.001)、人均年收入(β=0.17,Z=2.311,P=0.022)、服药规律性(规律服药vs不规律服药：户7.35,Z=4.632,P<0.001)进入方程。表明发病年龄和社会支持总分与总体功能呈正相关,城市患者总体功能水平好于县城和农村的患者,规律服药的患者总体功能好。研究的主要结论结论(一)1986-2006年唐山地区出院精神分裂症患者氯氮平使用率变化1.在1986-2006年间,唐山地区出院精神分裂症患者氯氮平使用率经历了一个从明显上升至下降的过程；2.在2006年住院精神分裂症患者中氯氮平仍是使用率较高的抗精神病药物；3.氯氮平使用特点发生变化,表现为单一用药日剂量增加和与其他抗精神病药物联合使用的比重增加；4.氯氮平的临床使用越来越多地受到精神分裂症患者的疾病特征和经济状况影响。结论(二)氯氮平对出院精神分裂症患者结局影响研究1.出院时氯氮平治疗的患者2年后随访时仍维持氯氮平治疗的比例高于其他药物治疗的患者；2.出院时氯氮平治疗的患者出院后2年内再住院率低,出院后至再住院的时间较其它药物治疗的患者长；3.出院时服用氯氮平的患者尽管有更多不利于预后的临床特征,但是随访时两组患者在临床症状、主观生活质量和总体功能方面差异没有显著性。本研究的意义氯氮平是疗效和不良反应都非常突出的抗精神病药物,在我国精神分裂症患者中有很高的使用率。这项研究系国内首次在有700万人口的地级城市水平开展的氯氮平使用模式长期的变化趋势调查并分析其影响因素。与此同时,对出院时服用氯氮平的患者进行2年结局的研究。研究初步证实了在自然状态下氯氮平对患者出院后的结局有促进作用。上述研究结果为进一步修订《精神分裂症治疗指南》中抗精神病药物使用原则,特别是氯氮平的合理使用原则提供了重要的证据。研究结果还表明,应在中国对氯氮平进行系统的成本-效益和风险-效益评估,在此基础上制定适合我国国情的抗精神病药物治疗原则,从而使精神分裂症患者得到合理治疗并有效利用卫生资源。更多还原

【Abstract】 BackgroundSchizophrenia is one of the most important public health issues. The incidence rate of schizophrenia ranges from 0.2%o to 0.6‰among adults, and the lifetime prevalence is approximately 1% all over the world. It was estimated that there are about 7.8 million patients suffering from schizophrenia in China in 1993. The characteristics of schizophrenia are chronic, high disability rates and poor prognosis, and it is one of the top ten diseases which lead to disability in labor age population. Since Chlorpromazine has been applied in the clinical treatment in 1951, antipsychotics have been playing important roles in improving the prognosis and the quality life of schizophrenia patients and becoming a main treatment. Clozapine is accredited as the most efficient antipsychotic to treat negative symptoms. It can be used in treatment refractory schizophrenia, in preventing suicide, in controlling violence behaviors of patients, as well as in decreasing hospitalization rates of patients and reducing costs of treatments. However, there were some reports that clozapine has severe side effects of agranulocytosis. In the United States, the use of clozapine has been restricted in clinical treatments, and it’s use has been limited to the treatment of refractory schizophrenia since 1980s.In China, clozapine was approved by use in clinical treatment in 1978, and has been used as common antipsychotics because of its better effective and lower price. It had been used as first-line antipsychotics for schizophrenia for a long period in China. However, because of its side effects, and because the development of new atypical antipsychotics, such as Risperdal, in the 1990s gave psychiatrists more choices in the treatment of schizophrenia, The Chinese Medical Association published "China schizophrenia treatment guidelines" (the Guidelines) in 2003. The Guidelines state that clozapine can’t be the first-line antipsychotic.Although to a high extend has the Guidelines restricted the usage of clozapine, it is still widely used in the treatment of schizophrenia in China. There are four reasons:(1) It may take some time for the guidelines to really influence the clinical decisions. (2) In current China, Costs of other atypical antipsychotics are too high to be accepted by many schizophrenia patients, while clozapine is preferred by many patients and psychiatrists in lower level hospitals because of its good effectiveness and much lower price. (3) It was reported that other atypical antipsychotics are not efficient for at least 30% of schizophrenia patients. So, it meets one of the criteria that clozapine can be used for the refractory schizophrenia. (4) clozapine is the only antipsychotics which can decrease the suicide rate among schizophrenia patients. Previous studies on clozapine were carried out in China before or just after the release of the Guidelines, and it may not reflect the real situation about the changes of clozapine use pattern. Furthermore, those studies were cross-sectional researches which implemented in different years or different areas, and there is lack of longitudinal comparability among those data. Until now, there has been no study either about the historic changes of the clozapine use in a specific area, or about the clinical features of patients who accept the clozapine treatment in China. Moreover, there is no systematic research on long-term outcomes or its associated factors among the schizophrenia patients who receive the clozapine treatment.ObjectivesThe study is to conduct a longitudinal observation about the situation of clozapine use among schizophrenia patients in Tangshan region, to evaluate the affects of clozapine use on the outcomes of schizophrenia, and then to provide scientific evidences to modify the Guidelines and to develop new policies for the prevention and treatment of schizophrenia.The specific objectives are as follow:(1) To describe the status of clozapine use and its changes among all schizophrenia inpatients in Tangshan from 1986 to 2006;(2) To explore the related factors of clozapine treatment among schizophrenia inpatients;(3) To follow up schizophrenia inpatients who used clozapine at the time of being discharged in 2 years, and to evaluate their clinical symptoms, subjective quality of life and general functions;(4) To explore related factors to the outcomes of schizophrenia inpatients in 2 years after being discharged.MethodsPart One:a survey on the changes of clozapine use rate among discharged schizophrenia patients in Tangshan from 1986 to 2006Medical records from six psychiatric hospitals (three county-level and three city-level hospitals) in 1986,1996,2001 and 2006 were all reviewed by trained interviewers using self-designed Medical Information Questionnaire. The information included the admission date, discharged date, reference number of hospitalization, socio-demographic data, health insurance status, duration of the disease, admission times, antipsychotics prescription at the time of discharge, family history of mental disorders, and the history of suicide behaviors. The definition of clozapine treatment is that patients used clozapine when being discharged, either only with clozapine, or combining with other antipsychotics. The evidences of schizophrenia diagnoses were from their medical records directly. Psychiatrists in this region have been using the Chinese Classification of Mental Disorders (CCMD) since CCMD-Ⅰwas published in 1981. Part Two:a study about the effectiveness of clozapine use during hospitalization on the outcomes of schizophrenia patients two years after dischargeSchizophrenia patients who met the criteria of DSM-IV, and were residents in Tangshan, as well as being discharged from January 1st to December 31st 2006 from the six psychiatric hospitals with only clozapine or one of any other antipsychotics treatments were recruited to the study. Patients in the clozapine group were selected from those with clozapine treatments by stratified cluster random sampling. Patients with one of any other antipsychotics treatments were consisted of the control groups matched with the patients on clozapine by the same hospital, same gender and similar admission dates (±2 days). Information was collected by face to face interviews using certain questionnaires, which included DSM-Ⅳ-TR disorders diagnosis by Structured Clinical Interview for-axisⅠ-research version (SCID-Ⅰ), social-demographic data and rehabilitation information after being discharged from hospitals by self-designed follow-up questionnaire, social supports and family environments of the patients by Family Adaptability and Cohesion Scale, Second Edition- Chinese Version (FACESⅡ-CV) and Social Support Scale, as well as clinical symptoms, quality of life and general functions by Positive and Negative Syndrome Scale (PANSS), Calgary Depression Scale for Schizophrenia (CDSS), Quality of Life Scale and Global Assessment Function (GAF).Results:Part One:Changes of clozapine use rates in discharged patients with Schizophrenia in Tangshan from 1986 to 2006There were 6550 discharged patients and 7920 men-times from the six psychiatric hospitals in the four index years in total. The number of discharged schizophrenia patients was 2717 which was 41.48% to all the discharged patients, and the discharged schizophrenia men-times were 3195 which was 40.34% to all the men-times.Clozapine use rate in schizophrenia patients at discharge was 6.2%, which is the seventh drug use rate of schizophrenia in 1986. The use rate increased to 45.2% in 1996 and kept increasing to 59.9% till 2001. It was the most frequently used antipsychotics in the two years. However, the rate fell down to 35.7% in 2006, and clozapine was the second most frequently used antipsychotics following risperdal.Clozapine was the most frequently antipsychotic drug used in poly-antipsychotic drug therapy except 1986, combining with sulpiride and risperidal.Average prescription doses of single clozapine use at time of discharge rose from (263±106) mg/d to (310±139) mg/d from 1986 to 2006. The doses of combined clozapine use fluctuated between (173±131) mg/d and (207±120) mg/d.There are different characteristics between patients who used clozapine and other antipscychotics when they were being discharged in the four years. Onset age of patients treated by clozapine was significantly lower than the patients treated by other antipsychotics (25±8 vs 27±9 years,t=7.076, P<0.001). The percentage of low economic level among patients treated by clozapine was significantly higher than that among patients treated by other antipsychotics (22.53% vs 16.51%, x2=18.215, P<0.001). The percentage of patients with family history of mental disorders in clozapine use group was significantly higher than that in other drugs group (32.02% vs 26.43%,c2=11.286, P<0.001). Rate of past suicidal behavior in clozapine use patients group was significantly higher than that in other antipsychotics use group (19% vs 15%, x2=9.049, P=0.003). The rehospitalization rate in patients treated by clozapine was significantly higher than that in other antipsychotics use group (39.6% vs 37%,x2=97.88, P<0.001). The rate of patients with clozapine treatment in county level hospitals was significantly higher than the rate in city level hospitals (15.71% vs 10.31%,x2=20.48, P<0.001). Based on the results above, we could find that patients with clozaphine treatment surrounded by more predictors of adverse outcome and lived in poorer economical condition compared with patients treated by other antipsychotics. Logistic regression analyses were conducted to analyze the associated factors to clozapine use in discharged schizophrenia patients in the 3 investigated years,1996,2001, and 2006 (αenter=0.05,αexcluded=0.10, P<0.05). In 1996,2 predictors, rehospitalization (OR=1.100,95%CI:1.016～1.190) and onset age (OR=0.945,95%CI:0.926～0.964) entered the model (R2=0.09). It indicates readmitted and earlier onset patients were more likely to be treated by clozapine. In 2001, the same 2 predictors consisted of the model as in the 1996’s model (R2=0.121, first rehospitalizationβ=0.541,P=0.005, OR=1.717,OR 95%CI:1.177～2.505 and onset ageβ=-0.031,P=0.008,OR=0.970, OR 95%CI:0.955～0.984). However, a different model was found in 2006, and five predictors consisted of the model (R2=0.227). They were suicidal behavior (OR=1.440, 95%CI:1.025～2.021), classification of hospital (OR=0.291,95%CI: 0.201～0.422), economic status (OR=1.341,95%CI:1.003～11.972), positive family history with mental disorders (OR=1.476,95%CI: 1.104～1.972), onset age (OR=0.968,95%CI:0.953～0.983), and first rehospitalization (OR=2.264,95%CI:1.561～3.283). Patients who had previously attempted suicide, were admitted in county level hospital, had worse economic condition, had positive family history, had younger onset age, and were rehospitalized were more likely to be treated by clozapine.Part Two:The affection of clozapine use on outcomes among discharged schizophrenia patientsAmong the 95 schizophrenic patients who were discharged with mono-clozapine treatment, one patient died of physical disease,1 died by suicide,5 were failed to be contacted because of incorrect address or phone number, and 13 refused to participate. Therefore, there were 78 patients were followed up. According to the examination with SCID for DSM-IV-TR at follow-up,76 patients met the inclusion criteria and consisted of clozapine group.Among the 127 control patients,1 patient died by suicide,5 were failed to be contacted because of incorrect address or phone number and 30 refused to participate. So, ninety-one patients were followed up. According to the examination with SCID for DSM-IV-TR at follow-up, 85 patients met the inclusion criteria and consisted of the control group.Onset age in the study group (M21, QL18, QU26) was significantly lower than that in the control group (M24, QL19, QU31), (Z=1.341 P =0.031). The number of admissions in the study group (M5, QL3, QU8) was significantly higher than that in the control group(M3, QL2, QU5), (Z =2.772 P=0.000). Course of disease in the study group(M13, QL8, QU20) was significantly longer than that in the control group(M6, QL4, QU14), (Z=4.512 P=0.000). Moreover, length of hospitalization when investigated in the study group (M45, QL26, QU158) was significantly longer than in the control group (M35,QL21, QU78), (Z=2.262 P=0.021). Additionally, rates of positive mental disorders family history and suicidal behavior in the study group were significantly higher than in the control group. At the follow-up, there were 53 patients (74.6%) in the study group still used clozapine as what they used when being discharged, but only 33 patients (38.8%) in the control group still had the same treatment as what they had when being discharged. There were 2 patients in the study group treated by other antipsychotic drug combining with clozapine, and 21 patients (24.7%) in the control group treated by other combined antipsychotic drugs. The differences of drug use status between the two groups had statistic significance during the follow-up (x2=12.6, P<0.001).During the time after discharge and follow-up,23 patients (30.3%) in the study group had been readmitted, which was significantly lower than the control group with 40 patients (47.1%) were readmitted (x2=3.27 P=0.032).Cox Proportional-Hazards Regression was conducted to analyze the related factors of rehospitalization:whether treated with clozapine when discharged or not(β=-0.74, P=0.01, HR=0.48), onset age (β=-0.21, P<0.001, HR=0.812), location of inhabitation (county vs villageβ=-0.60, P=0.082, HR=0.553, city vs villageβ=-0.78, P=0.022, HR=0.462), and course of disease (β=0.25, P<0.001, HR=1.281) composed of model. Patients who were treated by clozapine, had lower onset age, lived in village, and had longer course of disease were more likely to be readmitted.Multiple regression was conducted to analyze the related factors to social support:total score of PANSS (β=-0.324, t=-4.382, P<0.001) and course of schizophrenia (β=-0.177, t=-2.390 P=0.018) consisted into the model. It referred that the severe level of schizophrenia symptoms and course of disease negatively related to social support.There was no significant difference of FACESⅡ-CV (Chinese version of the Family Adaptability and Cohesion Scale) scores between the clozapine group and the control group. Multiple linear regression model was conducted to analyze the related factors to FACESII-CV scores. It stated that severe level of schizophrenia symptoms and course of disease negatively related to family adaptability and cohesion.There was no significant difference of PANSS scores between the study group and the control group. Generalized Linear Model was conducted to analyze the related factors to PANSS score:duration of onset to the first treatment(month,β=3.044, Z=3.430, P=0.001), total scores of social support scale (β=-0.793, Z=-4.150, P=0.000), the family per-capital income (β=-0.91, Z=-3.290, P=0.001), and total scores of CDSS (Calgary Depression Scale for Schizophrenia) composed of the model. It indicated that duration of onset to the first treatment and severity of depression were positively related to schizophrenia symptoms; on the contrary, total scores of social support scale and family economical status were negatively related to schizophrenia symptoms.There was no significant difference in scores of life quality between clozapine group and the control group. Ggeneralized Linear Model was conducted to analyze the related factors to life quality: duration of onset to the first treatment (month,β=-0.403, Z=-2.83 P=0.005), total scores of social support scale (β=0.192, Z=6.27, P<0.001), total scores of CDSS (β=-0.231, Z=-3.92, P<0.001) entered the model. It meant the duration of untreated psychosis and the severity of depression were negatively related to the level of subjective life quality, and the scores of social support was positively related to the level subjective life quality.There was no significant difference of global assessment function (GAF) score between the two groups. GLM was applied to analyze the related factors to the GAF score:living area (county vs urbanβ=-4.94, Z=-2.29, P=0.021; village vs urbanβ=-3.70, Z=-1.92, P=0.048), onset age (β=0.19, Z=2.192, P=0.029), total score of social support (β=0.37, Z=3.261, P=0.001), family per capital income (β=0.17, Z=2.311, P=0.022), regularity of medication(regular vs irregular:β=7.35, Z =4.632, P<0.001) enter the model. It implicated that onset age and social support score were positively related to the GAF, the GAF level was better among city patients than county and rural patients, and the patients who had medicine regularly were better.ConclusionsFirst series/cluster of the conclusions(1) The use rate of clozapine among discharged patients in Tangshan increased until 2002, and then decreased in 2006.(2) Clozapine was one of the antipsychotics with higher use rate among schizophrenia inpatients in 2006.(3) There were several changes of clozapine use from 1986 to 2006. The daily dose and the weight of combination use with other antipsychotics has been increased.(4) Clozapine use was affected by the different clinical characteristics of the disease and their economic status. Second series/cluster of the conclusions(1) At the follow-up, the proportion of patients still on the antipsychotic drug prescribed at the time of discharge in clozapine group was higher than that in the control group.(2) During the two-year follow-up, the rate of re-hospitalization among patients with clozapine treatment when being discharged was lower, and the time between discharged and the next admission was longer than those with other treatments. (3) Although the patients with clozapine treatment when being discharged had more adverse clinical features (previous) to the prognosis than those with other treatments, the two groups had no statistical difference in their clinical symptoms, subjective quality of life and general functions by the end of follow up.SignificanceClozapine has been a widely used antipsychotic agent in China since 1970s which is prominent both in efficacy and in side effects. This has been the first investigation on changing use pattern of clozapine and on factors that influnce this course in a municipal region with population of 7 million over the period 1986～2006 in China. In addition, patients taking clozapine at the time of discharge were followed up two years after their discharge. This preliminary naturalistic observation indicates that clozapine use can promote positive outcomes of schizophrenia patients after their discharge in community. The study provided the new evidences for further revising China Schizophrenia Treatment Guidelines, especially for the principles of rational clozapine use. The results also suggested that the systematically cost-effectiveness and risk-benefits analysis of clozapine should be carried out in Chinese context and antipsychotics use principles suited to the present conditions of our country should be formulated on the basis of such researches, thereby to utilize the mental health resources properly and effectively.更多还原