【作者】 黎健；

【导师】 钱序； 陈荔丽； 陈英耀； 赵耐青；

【作者基本信息】 复旦大学 ， 流行病与卫生统计学， 2008， 博士

【摘要】 一、研究背景尽管1987年发起了全球性母亲安全行动,但至今仍有一些国家的孕产妇死亡率没有显著下降,目前全球每年约有52.9万孕产妇死亡,几乎99%的孕产妇死亡发生在发展中国家。孕产保健政策的发展和可持续性取决于政策过程是否科学,这个过程在许多发展中国家都比较缺乏。为了降低孕产妇死亡率,广西自1998年开始实施以促进住院分娩为核心的母亲安全政策。科学依据对于决定政策不同阶段的干预重点以及确定政策的未来走向非常重要。政策的利益相关者希望了解政策的发展过程及实施效果,这些信息需要凭借科学的理论方法来获取与分析。目前对于广西母亲安全政策的政策过程及实施效果缺乏系统、全面、深入的评价,未能对政策的利益相关者所关心的问题给予满意的回答,未能提出令人信服的信息为政策的进一步发展提供循证依据。本次研究正是致力于填补这一空白,以促进政策的顺利发展,提高广西的孕产保健水平。二、研究目的本研究从政策学的角度,对广西母亲安全政策的发展过程进行形成性评价,探讨政策的主要决定因素在政策过程中的作用与影响:从孕产保健服务利用方在获取孕产保健服务的改善方面与从孕产妇死亡的特征变化方面来评价政策的实施效果,并评价政策的发展方向;为政策决策者及实施者发现新的需求,促进政策进一步发展提供循证的决策依据和政策建议。三、研究内容与研究方法1.采用方便样本法在广西南丹县、融水县与三江县选择母亲8人、婆婆2人、村妇女主任2人、村妇幼保健员2人及前家庭接生员5人进行个人叙述性访谈;选择政府官员、政策制定者、卫生管理者、医生、市民社会成员、国际组织官员共13人进行关键知情人半结构深入访谈,结合使用文献分析法,采用政策形成性评价模型从政策环境、政策过程、政策内容和政策参与者等影响政策的主要方面对广西母亲安全政策的发展过程进行形成性评价,重点研究卫生系统、人力资源、服务提供与市民社会等政策的主要决定因素对政策的影响。2.以2003年全国第三次卫生服务调查广西数据库为基线资料,采用多阶段分层随机抽样法于2007年在广西邕宁县、阳朔县、融水县、平果县、南丹县、容县等6县各调查最近3年有过活产史的母亲99人共594人,作为政策干预后的结果资料,应用安德森卫生服务利用行为模型,评价广西农村孕产妇对于住院分娩服务利用的改善情况并分析影响因素。3.利用1998年至2006年的广西孕产妇死亡监测点的孕产妇死亡监测资料评价广西孕产妇死亡的特征变化情况,分析影响孕产妇死亡的主要原因及影响孕产妇死亡地点的因素。4.利用1998年至2006年的广西妇幼卫生工作年报表,采用系统决策分析方法评价广西母亲安全政策的综合实施效果,利用时间序列分析方法拟合广西孕产妇死亡率的预测模型并进行近期的外部预测,以评价广西母亲安全政策的发展方向,提出促进政策进一步有效实施的循证政策建议。四、主要研究结果1.广西母亲安全政策自实施以来,在政府重视与社会支持下,在全体政策参与者的共同努力下,通过采取转变家庭接生员职能、建立健全乡镇基本产科服务中心和县级产科急救中心、新农合与降消项目有效结合补助住院分娩、免费接送孕产妇与免费住院待产、母婴安全担架行动、市民社会组织积极促进等富有特色与成效的措施,卫生系统管理、卫生人力资源、服务提供与市民社会等政策主要决定因素积极地相互影响,促进了政策顺利地发展,全区住院分娩率从1998年的45.67%上升到2007年的93.75%,孕产妇死亡率和婴儿死亡率分别从1998年的86/10万和23‰降低到2007年的24.06/10万和12.64‰,并且提前4年达到广西两个《规划》2010年的目标要求。2.分娩地点在基线调查时以在家分娩为主,占58.72%,住院分娩率仅为39.80%:干预后住院分娩率为92.09%,在乡镇卫生院分娩的比例为66.67%。在家分娩的比例逐年下降,在乡镇与县级医院分娩的比例呈逐年上升趋势(P＜0.001),其中尤以在乡镇卫生院分娩的比例上升显著。基线调查时在家分娩的主要原因是经济困难(46.03%)、观念落后(21.34%)、急产(17.57%)和交通困难(9.62%),而干预后调查在家分娩的主要原因是经济困难(37.78%)、急产(35.75%)、交通困难(17.78%)与观念落后(6.67%)。政策干预对于提高少数民族、高龄、农民、非高危、本人及丈夫文化程度低、家庭经济状况较差、居住地离医院远、产前检查次数少和首次产前检查时间晚的孕产妇和经产妇的住院分娩率效果更好。3.经多因素分析调整可能的影响因素后,基线调查时影响住院分娩服务利用的因素有分娩史、民族、孕产妇教育状况,家庭的饮水类型、以最快方式去最近医院的时间、是否被动员过住院分娩与产前检查情况等,按照对于住院分娩的促进作用大小排列,其OR值分别为:5次以上产前检查者25.579、被动员过住院分娩者7.865、高中文化者4.479、家庭饮用自来水者2.735、去医院时间少于10min者1.856、初产好1.71、汉族1.508。干预后调查影响住院分娩服务利用的因素有民族、知道有住院分娩补助、知道产科急救电话、上过孕妇学校、能得到降消或新农合补助、丈夫教育状况、建立广西孕产妇保健手册、产前检查情况与是否认为应主动获取保健服务等,按照对于住院分娩的促进作用大小排列,其OR值分别为:能得到新农合补助者482.758、能得到降消补助者137.912、5次以上产前检查者18.673、知道急救电话者17.880、丈夫高中文化者12.672、知道有住院补助者10.554、汉族7.769、认为应主动获取服务者7.431、建立孕产妇保健手册者4.555、上过孕妇学校者4.049。说明通过政策干预,基本消除了分娩史、经济可及性与交通可及性对于住院分娩服务利用的影响,但民族、文化程度、产前保健服务质量、孕产保健意识与社会动员情况等仍是影响住院分娩的重要因素。4.政策干预提高了产前检查覆盖率(P＜0.05)、平均产前检查次数(P＜0.05)与平均产后访视次数(P＜0.05),并提早了平均首次产前检查时间(P＜0.05)。干预后孕妇学校的入学率为48.31%,应加强产前筛查和新生儿疾病筛查的教育。产前检查质量较高,平均产前检查项目数为8.27项,有72.14%的对象做了全部9项产前检查:平均产后访视项目数为5.65项,接受8项以上产后访视项目者占接受过产后访视者的20.85%,产后访视质量有待进一步提高。政策干预后孕产妇对医院服务的满意度总体水平高,对医院服务无不满意的人由基线时的50%上升到干预后的78.98%。孕产保健意识总体水平较高,获取孕产保健知识的最主要来源是医务人员,占75.87%,其次是宣传标语与孕妇学校,分别占22.30%与22.12%。5.干预后扣除降消项目与新农合的补助,在乡镇卫生院顺产的实际费用中位数比基线调查时低60元,在县级医院顺产的实际费用中位数比基线调查时低150元,在乡镇卫生院和县级医院顺产实际支付的费用大为降低,提高了孕产妇获取住院分娩服务的经济可及性。6.死亡孕产妇中居住在山区、文盲、家庭人均年收入最低、没做过产前检查的孕产妇所占的比例呈逐年下降趋势(P＜0.05):孕产妇死在家中的比例呈逐年下降趋势,而死在县级以上医院的比例呈逐年上升趋势(P＜0.001)。7.孕产妇死亡的主要原因为产科出血(44.25%)、羊水栓塞(9.95%)、妊娠合并心脏病(9.51%)与妊高征(9.29%)。产科出血的死因构成比呈逐年下降趋势,羊水栓塞的死因构成比呈逐年上升趋势(P＜0.05);胎盘滞留(31.75%)和宫缩乏力(27.50%)是产后出血死亡的前两位原因,并且胎盘滞留的死因构成比呈逐年下降趋势(P＜0.001)。8.死亡孕产妇中可避免与创造条件可避免死亡的构成比呈逐年下降趋势,不可避免死亡的构成比呈逐年上升趋势(P＜0.001);可避免和创造条件可避免的死亡以个人家庭与医疗保健系统的知识技能方面的问题为主,个人家庭的知识方面的问题占68.03%,医疗保健系统的知识技能方面的问题占21.58%。9.经多因素分析调整了可能的影响因素后,民族、教育状况、分娩史、计划生育情况、居住地、家庭年人均收入与产前检查情况是影响孕产妇死亡地点的因素。与在家里死亡的孕产妇相比,高中文化者(OR=11.78)、居住在平原者(OR=2.726)、初产妇(OR=2.036)、人均年收入4000-8000元者(OR=1.917)、汉族孕产妇(OR=1.904)在县区级以上医院死亡的可能性更大:与在家里死亡的孕产妇相比,高中文化者(OR=3.594)、产前检查多于5次者(OR=2.879)、汉族孕产妇(OR=2.383)、计划内生育者(OR=1.647)、初产妇(OR=1.396)在乡镇卫生院死亡的可能性大,研究结果说明上述孕产妇获取住院分娩服务或医疗急救服务的可及性较好,也提示了少数民族、计划外生育、文化程度低、家庭经济状况差、居住在山区、产前检查次数较少的孕产妇和经产妇获取住院分娩服务或者在发生致命产科风险时获取医疗急救服务的可及性差,在家中死亡的风险大,应是下一步政策干预的重点人群。10.经TOPSIS法评价,广西母亲安全政策实施9年来综合效果呈逐年上升趋势:采用ARIMA时间序列模型预测2008年广西孕产妇死亡率为17.695/10万。五、政策建议1.维护良好的卫生系统外部环境,强化政府在广西母亲安全政策发展中的主导地位和公共责任;2.进一步加强与市民社会的合作,建立广西母亲安全政策与其他社会政策有效的协调与配套机制:3.继续增加孕产保健事业的经费投入,建立制度化的孕产保健经费保障机制和投入增长机制;4.建立起覆盖全人群、功能完善、分工合理的孕产保健服务体系:(1).加强农村基层产科建设,提升服务能力与服务质量,建立并完善各个层级服务网络之间相互转诊的良性循环机制:(2).加强产科人力资源队伍建设,制定广西孕产保健人力资源的长期发展战略:(3).建立妇幼卫生系统与计划生育系统的协调机制,将计划外妊娠的孕产妇全面纳入孕产保健系统管理;5.设立孕产妇待产所进一步提高偏远农村孕产妇对住院分娩服务的地理可及性;6.因地因人制宜采取健康促进策略,重点关注少数民族和低文化妇女,将孕产保健知识纳入义务教育,提高住院分娩的可持续性。更多还原

【Abstract】 Ⅰ.BackgroundAlthouth the global Safe Motherhood Initiative was launched in 1987,the maternal mortality ratio（MMR） in some countries does not decrease significantly till now.An estimated 529,000 women die each year worldwide during pregnancy,childbirth or immediate postpartum,in which almost 99 percent maternal deaths take place in developing countries.The development and sustainability of maternal health care policy depends on whether the policy process is scientific,which is comparatively deficient in many developing countries.In order to reduce MMR,Guangxi Zhuang Automous Region launched the safe motherhood policy with the focus of improving rate of institutional delivery since 1998.The scientific evidence is very important to determine the priority of intervention at different policy stages and the future pathway of safe motherhood policy.The policy stakeholders wish to understand the development process and implementation effectiveness of policy and these informations need to be obtained and analysed through scientific theoretics and methodology.Currently,there lack of systematic,comprehensive and in-depth evaluation of policy process and implementation effectiveness for Guangxi safe motherhood policy,lack of satisfactory answers for the questions concerned by policy stakeholders,lack of convincing information to provide evidence-based suggestion for further policy development.This study is intended to narrow those gaps to some extent,and is served for facilitating the policy development and maternal health care improvement in Guagnxi.Ⅱ.Study goalThis study is to conduct a formative evaluation for development process of Guangxi safe motherhood policy and focus on discussing the impacts of key policy determinants on policy process,to evaluate the implematation effectiveness of policy from the aspect of improvement of maternal health service utilization and the aspect of characteristic changes of maternal death,together with evaluating the macroscopic development direction of policy,and to provide evidence-based policy suggestions in order to promote policy makers and health managers find new demands and facilitate the further development of policy.Ⅲ.Study contents and methodologies1.Using non-random,convenience sampling to select 8 mothers,2 mother-in-laws,2 village woman directors,2 village maternal health workers and 5 former traditional birth attendants in 3 counties to conduct narrative interviews,to sample a total of 13 key policy informants including polictician,policy maker,health manager,doctor, civil society organization,international NGO officer to conduct semi-structured in-depth interviews,combined with using of literature analysis method.The policy formative evaluation model was used to conduct the formative evaluation for Guangxi safe motherhood policy from aspects of policy context,policy process,policy contents and policy actors with the focus on the impacts of key determinants of health system, human resources,service deliver),and civil society on policy process.2.With Guangxi databank of the 3^rd National Heatlh Service Survey as baseline data, the post-intervention data of policy intervention derived from the investigation of 594 women in 6 counties who had live delivery history in the last 3 years using multi-stage stratified randomized sampling.Using Anderson health service utilization behavioral model as analytical framework to evaluate the improvement of institutional delivery service utilization of Guagnxi rural pregnant women and analyse its determinants.3.Using the surveillance data of Guangxi pregnant women death between 1998 to 2006 to evaluate the characteristic changes of maternal death,analyse the causes of death of pregnant women and the determinants of the death place.4.Using Guangxi MCH routine annual reporting data to evaluate the comprehensive implementation effectiveness of Guangxi safe motherhood policy with technique for order preference by similarity to ideal solution（TOPSIS method）.To set up a mathematic predict model of MMR of Guangxi using autoregressive integrated moving average model（ARIMA model） and conduct a extrapolated prediction in the near future.To propose evidence-based policy suggestions to facilitate further effective implementation of policy.Ⅳ.Main study results1.Since Guangxi safe motherhood policy was launched in 1998,all levels of governments attach high importance to it and all society participate in it positively. Through some effective strategies including transforming the function of traditional birth attendants,setting up and perfecting towmship essential obstetric centers and county emergency obstetric centers,subsidy of institutional delivery by New Rural Cooperative Medical Scheme（NCMS） combined with the Project of Lowering MMR and Eliminating Newborn Tetanus（PLMENT）,free picking up pregnant woment to hospital and free waiting for delivery in hospital,safe motherhood stretcher action in mountainous areas and positive supporting by civil society orgamzations,four key policy determinants,namely health system management,human resources,service delivery and civil society,interact positively to facilitated successfully the policy development.The rate of institutional delivery of Guangxi increased from 45.67%in 1998 to 93.75%in 2007,the maternal mortality ratio and infant mortality ratio decreased respectively from 86/100,000 and 23‰to 24.06/100,000 and 12.64‰, which all achieve the targets of Guangxi Outline for Women’s Development in 2010 in advance of 4 years.2.The rate of home delivery was 58.72%and the rate of institutional delivery was only 39.80%at baseline survey.The rate of institutional delivery increased to 92.09% at post-intervention survey and the rate of delivery in township health center was 66.67%.The rate of home delivery decreased gradually and the rate of delivery in township health center and county level hospital showed the increased tendency gradually（P＜0.001）,especially for the rate of delivery in township health center.The main causes of home delivery at baseline survey was economic difficulty.（46.03%）, backward consciousness（21.34%）,precipitate labor（17.57%） and traffic difficulty （9.62%） by order,and the main causes of home delivery at post-intervention survey was economic difficulty（37.78%）,precipitate labor（35.75%）,traffic difficulty （17.78%） and backward consciousness（6.67%） by order.The policy intervention showed better result in terms of promoting institutional delivery for those pregnant women who were the minority,the elder,the peasant,the non-high risky,with low education background,the poor,living in distant from health center,with few times of prenatal checkup,late for first prenatal checkup and multiparas.3.Multivariate logistic regression analysis showed the determinants of institutional delivery service utilization at baseline survey were delivery history,nationality, education background of pregnant women,the type of family drinking water,time needed to get to the nearest hospital by the most convenient traffic,whether or not be advocated to institutional delivery and the frequency of prenatal checkup.The OR value were 25.579 for those with over 5 times of prenatal checkup,7.865 for those being advocated to institutional delivery,4.479 for those with education of senior middle school,2.735 for those drinking tap water,1.856 for those getting to hospital in less than 10 min,1.71 for primiparas and 1.508 for Han nationality respectively. After policy intervention,the determinants of institutional delivery utilization were nationality,knowing the subsidy of institutional delivey,knowing the phone of obstetric emergency aid,accepting the education of pregnant women school,being able to get the subsidy from NCMS or PLMENT,the education background of husband,getting the pamphlet of maternal health care,the frequency of prenatal checkup and whether or not considering it should utilize maternal health service positively.The OR value were 482.758 for those being able to get subsidy from NCMS,137.912 for those being able to get subsidy from PLMENT,18.673 for those with over 5 times of prenatal checkup,17.88 for those knowing the phone of obstetric emergency aid,12.672 for those whose husband with education of senior middle school,10.554 for those knowing the subsidy of institutional delivery,7.769 for Han nationality,7.431 for those thinking it should utilize maternal health service positively, 4.555 for those getting the pamphlet of maternal health care and 4.049 for those accepting the education of pregnant women school respectively.The policy intervention mainly eliminated the impacts of delivery history,economic accessibility and traffic accessibility on institutional delivery utilization,however,nationality, education background,quality of prenatal health care service,consciousness of maternal health care and social mobilization are still the important determinants of institutional delivery service utilization.4.The policy intervention increased the coverage rate of prenatal checkup（P＜0.05）, the average frequency of prenatal checkup（P＜0.05）,the average frequency of postpartum visiting（P＜0.05） and shifted the average time of first prenatal checkup to an earlier time（P＜0.05）.The rate of accepting education of pregnant women school was 48.31%at the post-intervention survey and the education of antepartum screening and screening against neonatal disease should be strengthened.Quality of antenatal checkup was high.The average items of prenatal checkup was 8.27 and 72.14%of women accepted the wholly 9 items of antenatal checkup.The average items of postpartum visiting was 5.65 and only 20.85%of women accepted over 8 items of postpartum visiting.It should improve the quality of postpartum visiting further.The satisfactory degree of hospital service was high on the whole after policy intervention and the proportion of women who thought the hospital service were satisfactory increased from 50%at baseline survey to 78.98%at post-intervention survey.The consciousness of maternal health care was high on the whole after policy intervention. The most primary source of acquiring maternal health care knowledge was to come from doctor with the rate of 75.87%,the second source was through propaganda slogan with the rate of 22.30%and the third source was through pregnant women school with the rate of 22.12%.5.Through the deduction of subsidy from NCMS and PLMENT,the median of actual expense of normal delivery in township health center and in county level hospital at post-intervention survey were 60 and 150 Yuan respectively lower than at baseline survey.Policy intervention improved the economic accessibility of aquiring institutional delivery.6.Within the dead pregnant women,the proportion of those who lived in mountainous areas,who were illiterates,who were the poorest,who never accepted prenatal checkup showed the declined tendency gradually（P＜0.05）.The proportion of women who died at home showed the declined tendency gradually and women who died at county and above level hospitals showed the raising tendency gradually（P＜0.0501）.7.The main cause of maternal death was obstetric hemorrhage（44.25%）,amnionic fluid embolism（9.95%）,pregnancy combined with heart disease（9.51%） and gestational hypertension（9.29%）.The proportional mortality rate of hemorrhage showed the declined tendency gradually and the proportional mortality rate of the amnionic fluid embolism showed the raising tendency gradually（P＜0.05）.The first two causes of postpartum hemorrhage was placental retention（31.75%） and uterine inertia（27.50%）,besides,the proportional mortality rate of placental retention showed the declined tendency gradually（P＜0.001）.8.The proportion of avoidable maternal death and avoidable maternal death by creating condition showed the declined tendency gradually,the proportion of unavoidable maternal death showed the raising tendency gradually（P＜0.001）.The primary problems within avoidable maternal death and avoidable maternal death by creating condition were personal,family and medical facilities’s knowledge and skill problems,within which the proportion of personal and family knowledge problems was 68.03%and the proportion of medical facilities’s knowledge and skill problems was 21.58%.9.Multinomial logistic regression analysis showed that the nationality,education background,delivery history,situation of family planning,dwelling space,family per capita yearly income and frequency of prenatal checkup were determinants of maternal death place.Compared with dying at home,those with education of senior middle school（OR=11.78）,those living in plain areas（OR=2.726）,primiparas （OR=2.036）,those with 4000-8000 family per capita yearly income（OR=1.917） and those with Han nationality（OR=1.904） had the higher probability of dying at county level hospitals.And compare with dying at home,those with education of senior middle school（OR=3.594）,those with over 5 times of prenatal checkup（OR=2.879）, those with Han nationality（OR=2.383）,those within the family planning（OR=1.647） and primiparas（OR=1.396） had the higher possibility of dying at towhship health centers.The research results suggested the pregnant women with minority,those outside of the family planning,those with low education background,those with poor family economy,those dwelling in mountainous areas,those with low frequency of prenatal checkup and multiparas had the low accessibility of obtaining institutional delivery or obtaining emergency aid when subjected with fatal obstetric risk,hence, they had more possibility of dying at home.Those pregnant women should be the key population of policy intervention.10.The comprehensive implementation effectiveness of Guangxi safe motherhood policy showed the increasing tendency as a whole using evaluation of TOPSIS method.The MMR of Guangxi in 2008 is predicted as 17.695/100,000 using ARIMA time series model. Ⅴ.Policy recommendation1.To maintain favorable external environments beyond the heatlh sectors and clarify the dominant position and public responsibility of all levels of government in the development of Guangxi safe motherhood policy.2.To further strengthen the cooperation with civil society and establish the efficient coordinated and integrated mechanism between Guangxi safe motherhood policy and other social policies.3.To increase the funding input of maternal health care enterprise continuously and formulate the institutional funding guaranteeing mechanism and input increasing mechanism of maternal health care.4.To establish the universal crowd-coveraged,function-perfected and rational work-divided maternal health care service system.（1）.To strengthen the construction of rural essential obstetric service centers,increase the service capacity and quality,establish and perfect the efficient mutual referral system between various levels of maternal health service networks.（2）.To strengthen the construction of obstetric human resources,formulate the long-term development strategy of maternal health human resources of Guangxi.（3）.To establish the coordinated mechanism between MCH system and family planning system and bring the pregnant women outside of the family planning into systematic management of maternal health care wholly.5.To establish the maternity waiting home to further increase the geographic accessibility of remote rural pregnant women to institutional delivery service.6.To adopt the health promotion strategy appropriately with the highly emphasis of minority and low education background pregnant women,integrate the knowledge of maternal health care into compulsory education to increase the sustainability of institutional delivery.更多还原