【作者】 张华；

【导师】 姜宝法；

【作者基本信息】 山东大学 ， 流行病与卫生统计学， 2011， 博士

【摘要】 研究背景艾滋病已经成为全球许多国家最严重的公共卫生和社会问题之一。自1981年首例艾滋病病人被发现以来,全球共6000万人感染了艾滋病病毒,将近3000万人死于同艾滋病有关疾病,对社会、文化、经济和政治产生了巨大的影响。HIV主要传播途径是性、血液和母婴传播,全世界约有四分之三的艾滋病传播为性行为传播。我国艾滋病总体呈低流行态势,疫情持续上升,但上升幅度有所减缓,部分地区和高危人群疫情严重,疫情正从高危人群向一般人群从高流行地区向低流行地区扩散,城市疫情增长高于农村。近年来,我国艾滋病的社会传播行为正悄然发生变化,地下不安全采供血行为被彻底清除,异性性传播持续成为主要传播途径,在异性性传播中,约三分之一为配偶之间传播,三分之二为非配偶间传播。在历年报告的病例中,异性性传播从2006年的30%上升到2007年的38.9%、2008年的40.3%和2009年的47.1%。我国艾滋病主要经异性性传播的很大原因是上世纪八十年代以来的性产业的死灰复燃和女性性工作者。我国的女性性工作者人群普遍具有以下特征：接触性生活和从事商业性性行为的年龄偏低、受教育程度较低、流动性强、大部分同时拥有商业性性伴和固定性伴、性病感染率较高、安全套使用率较低、某些女性性工作者还属于药物滥用人群。艾滋病综合监测与文献资料显示,60%的女性性工作者不能坚持每次性行为都使用安全套；与客人发生性交易时坚持每次都使用安全套占13%到54%；约有半数在与固定性伴发生关系时从不使用安全套。女性性工作者在商业性性交易及非商业性性行为中普遍缺乏安全保护措施、性伴更换频繁和自身性病的患病率高等因素,使得她们处在感染艾滋病病毒/性病高风险之中,并在传播艾滋病病毒/性病过程中起重要作用。截止到2005年底,大约12.7万女性性工作者和嫖客感染艾滋病病毒,占评估总数的19.6%。来自国家哨点监测数据表明在女性性工作者人群中艾滋病病毒的平均感染率为1.0%,最高为10.0%。2009年全国艾滋病哨点监测结果显示,10.3%的女性性工作者监测哨点艾滋病病毒抗体阳性检出率在1.0%至5.0%之间。从某种角度看,艾滋病实际上是一种行为性疾病,个人和群体的行为决定其流行态势,至今艾滋病仍无有效治疗方法及安全有效地抗艾滋病病毒的疫苗。因此,艾滋病高危行为的改变是目前最有效地控制艾滋病流行的措施。《2010年全球艾滋病疫情报告》的数据显示艾滋病蔓延势头已获遏制,该报告指出艾滋病疫情地好转得益于包括使用安全套在内的预防措施。艾滋病高危行为转变不仅受到个体因素的影响,还受到社会因素、物质环境因素(自然环境、人工环境)等影响。为此,国外学者在艾滋病高危行为研究中创建并发展了许多高危行为转变理论及模型,主要有知信行理论、健康信念理论、合理行动/计划行为理论、行为转变理论模式和信息-动机-行为技巧模型理论。这些理论从不同侧面揭示了影响艾滋病高危行为改变的主要因素,基于这些理论指导的干预研究也收到了良好的效果。信息-动机-行为技巧模型是在艾滋病高危行为研究中应用最广泛的模型之一,它的适用性已经在多种高危人群中得到了验证,以信息-动机-行为技巧模型理论为指导的艾滋病高危行为干预也收到了良好的效果。本研究拟采用信息-动机-行为技巧模型理论为基础,分析艾滋病高危行为发生的影响因素及影响方式,探讨适合我国女性性工作者人群高危行为的干预措施,为艾滋病高危行为干预提供理论指导。由于受到传统伦理和道德规范的影响,女性性工作者人群常受到主流社会排斥,成为受歧视、边缘化、隐蔽的群体。女性性工作者一般不愿暴露身份,很难区分女性性工作者与正常人群,不能确定女性性工作者人群总体,无抽样框架,对此人群几乎不可能进行概率抽样。因此,传统的概率抽样方法不适合该类人群。目前国内对女性性工作者研究大都采用方便抽样,如目标抽样、以机构为基础的抽样和时间地点定位抽样等。这些抽样方法所抽取的样本不能代表目标人群总体,其研究结果外推时会受到限制。如何有效地招募女性性工作者人群并提高样本的代表性是研究者所面临的主要问题。同伴推动抽样方法,方法源于链式推举,是近年来发展起来的一种近似概率抽样的链式推举抽样。该方法不是由研究者从某抽样框架中抽取研究对象,而是由研究对象从其社会关系网络中限额推举其同伴参加调查,通过研究对象连续不断地推举组成样本。其优点是：1)样本代表性强。2)抽样效率高。同伴推动抽样法是一种针对难以接近人群和隐蔽人群的抽样方法,在国外已经成功地将其运用在男男性接触者、男性双性恋者、注射吸毒和女性性工作者等人群中,在国内同伴推动抽样法方法也成功应用在男男性接触者隐蔽人群中,并取得了一些经验。本研究采用同伴推动抽样法方法调查女性性工作者,以期获得一个能够代表女性性工作者总体的样本。研究目的1.利用重复调查和数量特征敏感问题的随机应答技术等对调查问卷进行信度评价,探讨在女性性工作者人群中应用调查问卷的方法获取资料的可靠性及影响因素；2.应用同伴推动抽样法招募女性性工作者,从理论及实践上探索同伴推动抽样法在女性性工作者人群中的适用性；3.了解女性性工作者人群的社会人口学特征、及其艾滋病/性病感染状况；4.探讨与艾滋病/性病传播相关的女性性工作者高危行为特征；5.选择信息-动机-行为技巧模型作为理论基础,建立适合我国女性性工作者人群的高危行为干预理论模型,分析女性性工作者人群高危行为发生的影响因素及各因素之间相互作用机制,为该人群艾滋病高危行为干预提供理论指导。研究方法1.研究对象从事性服务的女性性工作者,严格定义指与不特定的男人性交以获得报酬的人。满足以下条件：1)调查前3个月生活在济南市；2)在调查前30天内发生过以性换钱的行为；3)年龄在16岁及以上且调查时没喝酒；4)持有一张有效联系卡。2.抽样方法应用同伴推动抽样法方法,招募一定数量的女性性工作者为研究对象。招募过程由最初选择的4个“种子”开始,发给每个种子3张联系卡,由种子介绍其同伴参加研究,被介绍的人得到3张联系卡,又继续介绍别人,以此方式使得推举链不断延长,直到达到预期样本量,招募过程中采用双重激励措施。3.调查内容主要调查研究对象的一般人口学资料、同伴推动抽样法相关内容、信息-动机-行为技巧模型各个组分包含的指标等；采集血液进行艾滋病病毒和梅毒抗体等检测；在自愿的原则上对女性性工作者进行妇科体检,并对宫颈和阴道分泌物进行检测；对调查问卷的重测信度、Chronbach’sα信度系数及和谐信度进行评价,验证调查问卷的可靠性。4.资料分析方法运用Epidata3.1建立数据库,NetDraw软件绘制招募网络分布图,应用RDSAT软件检验样本构成是否趋于平衡、估计总体构成以及通过比较样本构成和估计的总体构成检验样本代表性等；SPSS11.5软件进行资料的整理分析；应用LISERL软件对假设的信息-动机-行为技巧模型进行检验评价等。研究结果1.在重复测量的问题中,绝大部分问题的重测信度较高(r≥0.4)；一敏感问题的和谐信度较高(r=0.42)；Chronbanch’sα信度系数大部分大于0.7。2.共调查427名女性性工作者,招募过程最初选取4个种子,4个种子均将招募链延长下去,招募链每延长一级称为一“波”,最长的招募链为21波。年龄、民族、文化程度、户口所在地与婚姻状况五个人口学特征分别在11～18波后达到平衡。年龄、民族、文化程度、户口所在地与婚姻状况五个特征各组的P值均大于0.05,无统计学意义,样本代表性较强。3.在调查的女性性工作者中年龄最小为17岁,年龄最大为53岁,平均年龄为29.6±7.4,26岁及以上占66.7%,文化程度普遍偏低,外省户籍人口占半数以上,农业户口占59.3%,近半数以上女性性工作者有子女。在所调查的女性性工作者中,40.5%来自于夜总会/卡拉OK/酒吧/歌舞厅,洗脚屋/发廊/路边店/站桩占13.2%,其它(家政服务6人、婚姻介绍所4人、网络4人、兼职3人、朋友介绍3人、在家4人。)占5.6%。中/高层次女性性工作者年龄普遍小于低层次的,而文化程度普遍高于低层次女性性工作者。4.女性性工作者人群艾滋病/性病认知水平较低,23个艾滋病知识平均答对问题个数为12.2±4.5,39.6%知晓13个以上的艾滋病知识,不同文化程度的女性性工作者艾滋病知识得分差异有统计学意义(P<0.05)。女性性工作者获得艾滋病知识的途径主要是大众传媒,75.4%通过电视、41.0%通过报刊/杂志获得艾滋病知识；极少数通过医疗预防服务机构。5.该人群首次性行为年龄13～28岁,平均年龄20.2±2.5岁；首次商业性性行为年龄13～40岁,平均年龄25.6±6.5岁；从事商业性性行为时间1～228月(19年),平均37.0±40.2月,接近半数以上的女性性工作者从事商业性性行为超过2年。最近一周商业性性行为次数为0～70次,平均次数4.5±8.4次；最近一月商业性性行为次数为1-260次,平均次数14.3±25.1次。低层次的女性性工作者与中/高层次的相比,从业时间更长,最近一周和一月商业性性行为次数较多。6.在最近一月商业性性行为中,发生阴道性交时47.1%的女性性工作者能够100%使用安全套,安全套平均使用率为85.1%；发生口交时48.7%能够100%使用安全套；发生肛交时52.2%能够100%使用安全套。不同层次和婚姻状况的女性性工作者在商业性性行为中发生阴道性交时安全套使用率之间的差异有统计学意义(P<0.05)；不同的最近一月商业性性行为次数、从事商业性性行为的时间、首次性行为年龄的女性性工作者在商业性性行为中发生阴道性交时安全套使用率之间差异有统计学意义(P<0.05)。53.4%的女性性工作者有非付费临时性伴,81.5%有固定性伴,48.0%同时拥有非付费临时性伴和固定性伴。性伴构成越复杂的女性性工作者在商业性性行为中阴道性交时安全套的使用率越低,同时拥有非付费临时性伴和固定性伴的女性性工作者安全套使用率最低。有非付费临时性伴的女性性工作者在与商业性性伴、非付费临时性伴发生性行为时,安全套使用率之间的差异有统计学意义(P<0.05)；有固定性伴的女性性工作者在与商业性性伴、固定性伴发生性行为时,安全套使用率之间的差异同样有统计学意义(P<0.05)。7.79.9%的女性性工作者有饮酒行为,12.6%使用过毒品。33.3%有吸毒行为的女性性工作者在最近一个月内发生商业性性行为时能够100%使用安全套,而无吸毒行为的女性性工作者在最近一个月内发生商业性性行为时能够100%使用安全套的占49.9%。35.5%吸毒的女性性工作者年龄在30岁以上,半数户口所在地为外省,88.9%为中/高层次的女性性工作者。8.89.7%的女性性工作者曾怀过孕,近80.0%有流产史,23.1%流产在3次及以上。63.9%在最近一年中出现过一种及以上性病/妇科疾病相关症状,在最近一年做过妇科检查的女性性工作者中77.7%患有妇科疾病。47.8%为了预防性病定期打消炎针或吃消炎药。在患性病后有77.6%能够到正规的医院或者是性病专科门诊就医,但仍有12.6%去药店自己买药。9.在427位女性性工作者中,4.2%乙肝表面抗原呈阳性,梅毒感染率为6.3%。在121位自愿参加妇科体检的女性性工作者中,霉菌感染率为8.3%,衣原体感染率为6.6%。63.0%感染梅毒的女性性工作者年龄在30岁以上,81.5%感染梅毒的女性性工作者为外省人群,51.9%感染梅毒的女性性工作者从事商业性性行为超过2年以上。不同的年龄、文化程度、婚姻状况和户口所在地之间梅毒感染差异均有统计学意义(P<0.05)。10.在传统信息-动机-行为技巧模型的最终模型中,信息和动机能够解释行为技巧22.0%总变异；信息、动机和行为技巧解释预防行为总变异43.0%。信息和动机均可通过行为技巧影响预防行为,信息对预防行为无直接效应,动机对预防行为直接和间接效应均有统计学意义(P<0.05),动机对预防行为的效应稍大于信息的效应。11.发展的信息-动机-行为技巧模型的初始模型模型和最终模型拟合程度评价指标均在可接受范围之内,最终模型能够解释预防行为、健康行为及安全套使用技巧、自我效能总变异的38.0%、57.0%和43.0%。在最终模型中,预防行为的影响因素包括社会参照对象支持、安全套使用经验及态度、健康行为及安全套使用技巧和自我效能。健康行为及安全套使用技巧影响因素包括信息、危险感知和社会参照对象支持；自我效能影响因素包括信息、依从动机、社会参照对象的支持和药物滥用状况；药物滥用状况、信息和社会参照对象支持还可通过健康行为及安全套使用技巧和自我效能间接影响预防行为。结论1.通过该调查问卷的重测信度、Chronbach’sα信度系数和和谐信度的结果可知本次调查可靠性较好,能够在女性性工作者人群中获得大量敏感问题较为真实的数据,但并不排除有些敏感问题存在低报或高报的现象。2.在女性性工作者人群中运用同伴推动抽样法方法是可行和有效的,但是实际操作过程有待丰富和完善,在今后研究中应探索一种适合国情、完善、高效的操作程序,以获得更具代表性的样本。3.女性性工作者人群在地域之间的流动性减弱,工作场所之间流动频繁。女性性工作者的艾滋病/性病知识水平较低,在该人群中存在一些错误艾滋病/性病预防观念及知识和行为相分离的现象。在今后干预中应加大对女性性工作者健康教育工作,改进工作形式以提高服务项目的利用率。4.未婚、低层次、首次性行为小于18岁、从事商业性性行为超过两年、有非付费临时性伴和固定性伴的女性性工作者在最近一个月商业性性行为中安全套使用率较低。女性性工作者与非付费临时性伴和固定性伴发生性行为时安全套使用率较低,女性性工作者的性伴构成越复杂安全套使用率越低。在今后干预工作应让她们充分认识到自身的“桥梁”作用,提高她们维护他人健康的责任感。5.低层次女性性工作者性病症状知晓率较低、从事商业性性行为的时间较长,最近一个月商业性性行为次数多于高层次的女性性工作者,安全套使用率较低。此类人群应为今后干预工作的重点。6.在女性性工作者人群中普遍存在饮酒、吸烟不健康行为,吸毒的女性性工作者安全套使用率明显低于不使用毒品的女性性工作者；且从事商业性性行为时间长于不吸毒女性性工作者。7.传统的信息-动机-行为技巧模型和发展的信息-动机-行为技巧模型均能够有效预测女性性工作者高危行为。信息通过健康行为和安全套使用技巧及自我效能对预防行为的间接效应有统计学意义,这提示在对女性性工作者的艾滋病高危行为干预中,加强艾滋病知识方面教育是必要的,同时应给予行为技巧和预防动机的干预。8.证实了女性性工作者人群中社会参照对象支持和依从动机在信息-动机-行为技巧模型中的作用,提高社会参照对象对使用安全套的支持程度可能是提高女性性工作者人群安全套使用率的重要策略之一。9.结果表明具有较高危险感知的女性性工作者能更易于进行健康行为,对女性性工作者人群的干预应加强风险意识教育,增强她们对感染艾滋病的危险和不安全性行为的认知。10.自我效能和健康行为与安全套使用技巧对研究人群的安全套使用起重要作用。提示在今后干预中应重视女性性工作者自我效能地提升,增强她们安全套使用的协商技巧和说服能力,树立女性性工作者健康行为的观念,自觉定期进行妇科体检和艾滋病病毒/性病检测。11.研究结果表明药物滥用对安全套使用间接效应有统计学意义,较低药物滥用水平的女性性工作者有着较积极的安全套使用经历和态度。这提示在艾滋病高危行为干预中应提高该人群预防药物滥用的意识。意义和创新1.利用多种方法评价问卷的信度,探讨了相关的影响因素,为以后提高该人群的调查问卷的信度提供科学依据；2.应用同伴推动抽样法招募女性性工作者,样本具有较强代表性。从理论和实践上探讨了该抽样方法在女性性工作者中应用的可行性及影响因素,为此抽样法在女性性工作者或其他隐蔽人群中的应用提供了理论依据和实践基础；3.以信息-动机-行为技巧模型理论为研究基础,扩展了该模型理论框架和应用范围,构建出适合我国女性性工作者的艾滋病高危行为干预模型,揭示了高危行为各影响因素以及它们与预防行为之间的相互作用机制,为今后干预提供了科学依据。更多还原

【Abstract】 BackgroundSince the USA reported the first case in 1981, acquired immune deficiency syndrome/human immunodeficiency virus (AIDS/HIV) has become a worldwide epidemic. AIDS/HIV epidemic has become a major global public health and social problems in many countries and one of the most serious infectious diseases to human life and health. It increasingly contributes to the erosion of civil order and economic growth. Sexual transmission, blood transmission, and mother-to-child transmission are the dominant modes of HIV transmissions, and about three-quarters of AIDS/HIV were through sexual transmission.Although the AIDS epidemic in China is still in the overall low prevalence, there are significant geographic variations in HIV prevalence in difference Province, some specific groups have emerged high HIV prevalence. Illegal blood and plasma donor was eliminated. Heterosexual contact has become the dominant modes of HIV transmission. About one third infected HIV through heterosexual transmission was through their partners. Heterosexual sexual transmission among cumulative HIV infection increased from 10.7% in 2005 to 47.1% in 2009.Since the 1980s the number of female commercial sex workers (FSWs) has markedly increased in China. The number of women engaging in commercial sex is estimated to be 4 to 10 million. Commercial sex plays a critical role in heterosexual transmission of HIV in China. Sex trade is not accepted by Chinese ethical standards and there is a substantial stigma to FSWs. They are discriminated and marginalized in China. However, some women become FSWs possibly due to being attracted by the relatively high income. According to previous reviews, most FSWs in China are young, mobile, less educated, and some FSWs have also reported substance abuse. However, a recent review reported rates of consistent condom use with their clients among FSWs in China of between 13% and 54%. Recent national comprehensive surveillance data showed that 60% of FSWs in China do not use condoms regularly.FSWs are playing more and more important role in the transmission of HIV infection. Since the majority of FSWs not only have commercial sexual partners but also have one or more noncommercial steady or casual sexual partners at the same time, once FSWs get infected with HIV, their sexual partners are easy to get infected also through unprotected sexual behavior. Approximately 127,000 FSWs and their clients were living with HIV/AIDS, accounting for 19.6% of the total number of estimated HIV cases at the end of 2005. Heterosexual transmission of HIV through contact with FSWs is of particular concern. Data from national sentinel surveillance indicated that the prevalence of HIV among FSWs averaged 1%, with the highest rates at 10% in several sites.10.3% national FSWs sentinel surveillance reported the prevalence of HIV from 1.0% to 5.0% in 2009.Because there is no vaccine or cure, HIV prevention depends on one’s ability to modify risky behaviors. The key to containing the spread of HIV is changing individual’s HIV risk behaviors. Consistent and correct condom use seems to be the most effective form of HIV prevention among FSWs. It would be imperative to determine predictors of consistent condom use among FSWs to develop effective interventions for the prevention of HIV spread in China.FSWs’HIV risk behaviors are affected by social, psychological and other complicated factors. Most models of HIV sexual transmission risk behavior in HIV are based upon social/health psychology theories to predict whether or not individuals will use condoms when having sex. These models include the Theory of Reasoned Action, with attitudes and intentions as key factors, the AIDS Risk Reduction Model, emphasizing behavior change stages of labeling, commitment, and enactment, the Information-Motivation-Behavior model, and Social Cognitive Theory, with self-efficacy being a central factor. Each of these models, to greater or lesser extents emphasizes the cognitive pathways involved in behavior change. The ARRM model above, however also, allows for a consideration of affective components (condom enjoyment) and sexual communication skills, in addition to purely cognitive constructs to explicate the pathways to sexual risk.The Information-Motivation-Behavioral Skills (IMB) model developed by Fisher and his colleagues was designed to predict HIV preventive behavior and necessary elements HIV prevention intervention. The conceptualization of the IMB model holds that HIV prevention information, motivation, and behavioral skills are the fundamental determinants of HIV preventive behavior. This study was designed to examine the IMB model and to describe the relationships between IMB model constructs.Nearly all these studies among FSWs were based on institution samples or convenience samples. However, institution samples may not be representative of the general FSW population. Furthermore, FSWs are a heterogeneous group and operate in a multiple-layer hierarchy. FSWs working in the entertainment establishments are in the middle of the spectrum and may be the most accessible group. We know little about the FSWs in either the higher ranks or the lower ranks of the hierarchy.Respondent-Driven Sampling (RDS) was used to recruit participants. RDS is an adaptation of chain-referral sampling and a suitable sampling method for hidden populations, which can provide relatively unbiased and representative population-based estimates. One advantage of RDS over other methods of sampling hidden populations, such as time-location sampling, is that it requires little in-depth formative research among study populations. RDS begins with a set number of non-randomly selected seeds (members of the target population). Seeds recruit their peers (other target population members who make up their social network) who in turn recruit their peers into the study. This occurs through successive waves of recruits which become increasingly more representative of the underlying population as the recruitment progresses. Many researchers have assumed that sampling can proceed without detailed prior knowledge about the local target population. RDS has proved feasible and successful in recruiting hidden populations, such as men who have sex with men, sex workers and drug-injecting sex worker. Objects1. Examine the reliability of the questionnaire using the test-retest reliability, Cronbach’s alpha coefficient, and the harmonious reliability.2. Explore the feasibility of RDS in the recruitment of female sex workers.3. Understand demographics of FSWs and HIV/STDs prevalence.4. Understand HIV-related high risk behaviors among female sex workers.5. Examine the predictors of condom use with clients during vaginal intercourse among FSWs based on the MB model and to describe the relationships between IMB model constructs.Methods1. Study participantsThe selection criteria for eligible FSW seeds and participants were defined as "a female over the age of 16 who has exchanged sex for money and has lived in Jinan in the past month and is not inebriated at the interview." Each seed or participant had a coded coupon.2. Recruitment of seeds and study participantsRDS was used to recruit female sex workers, and four seeds were initially selected. Each seed or participant was asked to recruit no more than three peers. Each seed was given three uniquely coded coupons to refer their peers. The desired sample size will be gotten by this chain-referral sampling method. Seeds and their recruits were given an incentive package including 50 Yuan, HIV prevention pamphlets and four boxes of condoms for successfully participating, plus an additional 20 Yuan for recruiting a FSW.3. MeasuresA structured questionnaire was used to collect data administered by trained female interviewers. Questionnaire-based interviews provided demographic and social information, sexual behaviors and condom use, drug use and health seeking behavioral information, AIDS/STDs knowledge, the constructs of the IMB model, and etc. Blood samples were collected from all participants to test their status of HIV, HBV, HCV, and syphilis prevalence. We also provide gynecological examination to test cervical and vaginal secretions that interviews voluntarily participated.4. Data analysisSurvey data were recorded with EpiData software. NetDraw software was applied for graphing the recruitment chains, and RDSAT software was used to exaime the equilibrium and representation of Sample. SPSS11.0 software was applied for all statistical analyses, such as mean, variance analysis, chi-square test, stratified analysis. The hypothetical IMB was examined by structural equation model (SEM) using the LISREL.Results1. Test-retest reliability was accepted (r≥0.4); most of Chronbanch’s a were over 0.7.2. In total 427 participants, were recruited for the study. The longest recruitment tree had 21 recruitment "waves", the evaluation variables including age, educational level, ethnicity, permanent residence, marital status have reached equilibrium.3. Mean age of the sample was 29.6 years (SD=7.4),66.7% were older than 25 years,99.1% was Han in ethnicity,66.0% received middle school education or less, and 32.6% were divorced or widowed.45.7% had no children. The majority (82.7%) earned more than 2,000 Yuan (or approximately U.S.＄300) per month.40.5% worked in night clubs/dance hall/karaoke bars/bars,13.2% worked in street based entertainment and hair/beauty salons.4. FSWs had low lever in AIDS/STDs knowledge. The AIDS/HIV knowledge score≥13 were 39.6%. Education levels were significant different in AIDS/HIV knowledge score. The major mode for FSWs to access AIDS/HIV knowledge was television, accounting for 75.4%, followed by newspapers/magazine, accounting for 41.0%. A few FSWs access AIDS/HIV knowledge were through health services, and 58.6% have never received free health service.5. The mean age of first lifetime sex was 20.2(SD=2.5), and the age of first lifetime sex was from 13 to 28 years old. The age of first commercial sex was from 13 to 40 years old. Sex work experience was from 1 to 228 months, and the mean of sex work experience was 37.0(SD=40.2) months. More than half had>2 years of sex wok experience. The number of clients was from 0 to 70 in the last week and the number of clients in the last month was 1 to 260. The mean number of clients was 4.5(SD=8.4) in the last week and the mean number of clients in the last month was 14.3(SD=25.1). The FSWs in low-status have and longer sex work experience and more the number of clients compared with the FSWs in middle/high status.6. Using condoms with clients every time during vaginal, oral, and anal intercourse in the last month were 47.1%,48.7%, and 52.2% respectively. Different the number of clients, sex work experience, first sex lifetime and marriage status were have difference condom use rate during vaginal intercourse with clients in the last month(P<0.05).48.0% have commercial sexual partners also have noncommercial steady and causal sexual partners at the same time; 81.5% have noncommercial steady sexual partner(s); 53.4% have noncommercial casual sexual partner(s). FSWs that have noncommercial casual sexual partner(s) were different in condom use when they have sex with clients and noncommercial casual sexual partner(s) (P<0.05).7.79.9% drank alcohol, and 12.6% used drugs.33.3% drug using FSWs reported using condoms with clients every time during vaginal intercourse in the last month. 35.5% age of drug using FSWs was>30, and 88.9% drug using FSWs were middle/high status.8.89.7% have ever been pregnant, and 80.0% had abortion(s).23.3% had over three abortions.77.7% of FSWs who had gynecological disease in the last fell gynecological disease.44.7% have used antibiotics for preventing HIV.63.9% reported ever appeared STD-related/ gynecological disease symptoms. If FSWs get STDs,77.6% FSW will choose hospital to treatment, and 12.6% will choose to treat by medicine from shopping. Only 19.4% of the women were tested for HIV in the past year.9. Syphilis infection and HBV rate were 6.3% and 4.2%. Mould and Chlamydia infection rate among FSWs who voluntarily received gynecological examination in this study were 8.3% and 6.6%.51.9% infected syphilis FSWs have more than two years sex work experience.63.0% age of infected syphilis FSWs were>30. Different age, education level, martial status, and registered permanent residence have significant different syphilis rates(P<0.05).10. The final traditional model was confirmed as a good fit (x2=143.76,73 df, CFI=0.98, RMSEA=0.08). Condom use among FSWs was predicted by motivation and behavioral skills. Information and motivation had the indirect effects on condom use mediated by behavioral skills.11. The final developed model was accepted x2=405.50,206 df, CFI=0.97, RMSEA=0.05, NNFI=0.95. The explained variances for condom use, health behaviors and condom use skills, and self-efficacy were 38%,57%, and 43%, respectively. Significant predictors of condom use were social referents support, experiences with and attitudes toward condoms, self-efficacy, and health behaviors and condom use skills. Health behaviors and condom use skills was significantly predicted by HIV knowledge, perceived risk, and social referents support. Significant predictors of self-efficacy included HIV knowledge, motivation to comply, social referents support, and substance use.The indirect effects on condom use as reported in the trimmed model were also examined. Variables that indirectly affected condom use through health behaviors and condom use skills or self-efficacy variables included HIV knowledge (p<0.001), social referents support (p≤0.001), and substance use (p≤0.05).Conclusions1. The reliability of the questionnaire was acceptable. However, we recognize that self-reporting bias may exist. The reliability of study participant responses may be questionable due the sensitive nature of responses about sexual practices.2. RDS methods were feasible and effective to attain sufficient samples of FSW. Further research is needed to assess how the practicalities of implementing RDS to abtain more representative FSWs population.3. FSWs reduced mobility in regions and strength mobility in workplaces. FSWs have low AIDS/HIV knowledge level. These misconceptions may result in less compliance in using a condom. These results suggest that more attention is needed toward strengthening education about specific AIDS/HIV knowledge where it is lacking.4. FSWs who were unmarried, low-status, the fist sex lifetime<18, more than two years sex work experience, have steady sexual partner had low condom use rates with clients during vaginal intercourse in the last month. FSWs have low condom use with their stable partners and noncommercial casual sexual partners. The AIDS/HIV intervention should increase their responsibility for safe sex and the awareness of FSWs about AIDS/HIV.5. FSWs in low-status have low STD-related symptom knowledge level, and longer sex work experience, more the number client in the last month, and low condom use rate. In the future intervention should focus on FSWs in low-status.6. Most of FSWs drank alcohol and smoked. Condom use rate of drug using FSWs was lower than no-drug using FSWs; sex worker experience was longer than no-drug using FSWs.7. The specific elements of the IMB model that are critical for condom use among FSWs were identified and confirmed the central proposition that information and motivation works mainly through behavioral skills to influence HIV prevention behaviors. The relationships between constructs of the IMB model were also examined. Our findings have implications for the development of HIV risk reduction interventions for FSWs. Further study should explore effective interventions based on the current study’s findings.8. The current study confirmed the role of social referents support and motivation to comply. HIV intervention for FSWs needs to be multilevel, integrating social factors into the intervention. Enhancing social referents support for condom use may be an important strategy to increase condom use.9. AIDS/HIV interventions should increase the awareness of FSWs about HIV and should address assertiveness training to refuse unsafe sex.10. Self-efficacy and health behaviors and condom use skills play an important role in condom use in FSWs. An awareness of the importance of engaging in health-promoting behaviors should be developed in FSWs. Also AIDS interventions should address enhancement of special communication skills so as to successfully negotiate safe sex.11. This study revealed that an indirect effect of substance use on condom use was significant and that lower levels of substance use accompanied positive experiences with and attitudes toward condom.Significance and innovation1. Using three methods examined the reliability of questionnaire. These results may contribute to improve the reliability of questionnaire in this population.2. Our sample is an approximate representative of FSWs using RDS recruit participants. We explored the feasibility of RDS and identified some barriers to successful recruitment among FSWs, which will provide reference for future research in this population and other hidden populations.3. The study based on the IMB model and developed the theory. The developed IMB model is effective to predict the determinant factors of condom use with client among FSWS in China. Further research is warranted to develop preventive interventions on the basis of the IMB model to promote condom use among FSWs in China.更多还原