Skilled Birth Attendance in the Tanzanian Lake Region. A study on women s preferences for obstetric care facilities

Skilled Birth Attendance in the Tanzanian Lake Region A study on women s preferences for obstetric care facilities Bart van Rijsbergen Skilled birth attendance in the Tanzanian Lake Region: A study
of 144
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Skilled Birth Attendance in the Tanzanian Lake Region A study on women s preferences for obstetric care facilities Bart van Rijsbergen 2011 Skilled birth attendance in the Tanzanian Lake Region: A study on women s preferences for obstetric care facilities, the influence of emergency and poverty conditions on these preferences, and the existence of preference heterogeneity. Master Thesis Bart van Rijsbergen (s ) Nijmegen, May 2011 Research Master Social and Cultural Sciences Radboud University Nijmegen Supervisor: Dr. Ben D Exelle Table of Contents LIST OF FIGURES AND TABLES LIST OF ABBREVIATIONS ABSTRACT PREFACE AND ACKNOWLEDGEMENTS VII VIII IX XI 1. MATERNAL HEALTH, AN INTRODUCTION TO A PERSISTENT CHALLENGE REDUCING MATERNAL MORTALITY THE DEMAND FOR MATERNAL HEALTHCARE TANZANIA IN CONTEXT GEOGRAPHY DEMOGRAPHY POLITICS AND ECONOMICS TANZANIAN HEALTH CARE MATERNAL HEALTH CARE THEORY AND HYPOTHESES HEALTH SEEKING BEHAVIOUR THE BEHAVIOURAL MODEL OF ANDERSEN THE BEHAVIOURAL MODEL APPLIED HYPOTHESES DATA AND MEASUREMENTS SURVEY SUBSAMPLE A SUBSAMPLE B SOCIO-ECONOMIC CHARACTERISTICS CONJOINT ANALYSIS DESIGN SELECTION OF ATTRIBUTES AND THE CORRESPONDING LEVELS v 4.7 FRAMING CONJOINT ANALYSIS FIELDING THE IMPORTANCE OF EMERGENCY AND POVERTY CONDITIONS INTRODUCTION HOME DELIVERY: ECONOMETRIC SPECIFICATION HOME DELIVERY: RESULTS FACILITY DELIVERY: ECONOMETRIC SPECIFICATION FACILITY DELIVERY: PREFERENCES WHEN EMERGENCY OBSTETRIC CARE IS NEEDED FACILITY DELIVERY: THE INFLUENCE OF INDIVIDUAL SOCIO-ECONOMIC CHARACTERISTICS PREFERENCES HETEROGENEITY INTRODUCTION ECONOMETRIC SPECIFICATION: MIXED LOGIT RESULTS OF THE MODELS WITHOUT INDIVIDUAL CHARACTERISTICS RESULTS OF THE MODELS WITH INDIVIDUAL CHARACTERISTICS RESULTS OF THE WILLINGNESS TO PAY DISCUSSION AND CONCLUDING REMARKS DISCUSSION LIMITATIONS AREAS FOR FUTURE RESEARCH CONCLUSION REFERENCES 57 APPENDICES 61 APPENDIX A - VIGNETTES APPENDIX B - HOUSEHOLD QUESTIONNAIRE ENGLISH APPENDIX C - WOMEN S QUESTIONNAIRE ENGLISH APPENDIX D - MAPS vi List of Figures and Tables Figures Figure 1 The behavioural model of Andersen 14 Figure 2 Example of vignette 29 Figure 3 Maps of research location 131 Tables Table 1 Variables used in the analyses 16 Table 2 Sample characteristics 22 Table 3 Characteristics subsample A 23 Table 4 Characteristics subsample B 24 Table 5 Attributes and levels used in the ranking exercise 27 Table 6 Scenarios used in the ranking exercise 28 Table 7 Probit regression on home delivery 34 Table 8 The effect of emergency conditions on preferences for facility attributes 37 Table 9 The effect of socio-economic characteristics and emergency 40 Table 10 Estimation results for rank-ordered logit and mixed-logit models 46 Table 11 Willingness to pay estimates in Tanzanian Shilling 50 vii List of Abbreviations CIDIN DHS GDP IIA MDG MMR MoHSW MUHAS NBS POPDEV SE SD TQ TZS UNDP WHO WTP Centre for International Development Issues Nijmegen Demographic and Health Survey Gross Domestic Procuct Independence of Irrelevant Alternatives Millennium Development Goal Maternal Mortality Rate Ministery of Health and Social Welfare Muhimbili University of Health and Allied Sciences National Bureau of Statistics Population and Development Programme Standard Error Standard Deviation Technical Quality Tanzanian Shilling United Nations Development Programme World Health Organization Willingness To Pay viii Abstract Maternal mortality rates in developing countries are high because of high rates of home delivery and hence low use of high-quality obstetric care. The Tanzanian government has implemented an exemption scheme for delivery care assuming poverty is the most important barrier to obstetric care. However, little is known about women s preferences for delivery care. To investigate how preferences correlate with women s socioeconomic characteristics and the presence of emergency conditions, we organize a survey and a conjoint ranking experiment with a sample of more than 800 women in Tanzania s Lake Region. The analysis of the data, uses probit-, rank-ordered logit as well as mixed logit models. The availability of drugs and equipment, the attitude of the provider and geographic distance all exert a significant influence on individual preferences. We find strong evidence of preference heterogeneity among respondents. The weights women attach to each of these attributes of obstetric care providers, depend on their individual socio-economic characteristics, such as wealth, experience with complications during previous delivery and economic activity. We also find that poverty increases the likelihood of delivering at home and that this effect is stronger in case of emergency conditions. Analyses of the preference data from the conjoint ranking exercise, however, indicate that costs tend to be less important in emergency conditions than in normal conditions. Moreover, we do not find any correlation between wealth and women s preferences that might support the lower likelihood of wealthier women to deliver at home. Apparently, poor women face just too many constraints when trying to access emergency obstetric care. Moreover, we find that women who are not economically active attach a much higher weight to distance, which is likely the result of their lower mobility. This result supports the idea that economically inactive women might prefer home delivery above delivery at health facilities because of its lower distance to travel. The implications of the findings for the targeteffectiveness of Tanzanian health policies are discussed. ix Preface and Acknowledgements On the cover of this master thesis, a single name is written, but many people have put effort in it. There is too little space to thank everybody individually, but I would like to express my gratitude to the people we visited in the Tanzanian Lake Region. They gave us a warm welcome and taught us the true meaning of the Swahili word Karibu. More than once, we were invited to join for lunch or we went homewards with all kinds of presents such as dried cassava or sweet potatoes. A special thanks to all the women whom we interviewed, for sharing their personal and often very intimate stories with us. Their accounts form the backbone of this thesis. To be able to collect our own data in Tanzania was a great opportunity and an exciting adventure for me. It was possible only because of many Tanzanian people who assisted us in many ways. I want to express my gratitude to Professor Kamazima for his support in many stages of the research process, to Mr Chitama and Ms Mosha for their assistance and troubleshooting on multiple issues while working in Tanzania. I would also like to thank Mr Kakoko and Ms Mpembeni for sharing their knowledge and Mr Samki for making the nice drawings, used in my experiments. We could not have collected the data without the knowledge, skills and dedication of the enumerators. Thanks to Jacinta, Marygoreth, Asinta, Modester, Concilia, Florida, Mwashi, Rosemary, Philomena, Janeth, Leah, Bahati, Rosemary and Loyce. I would like to thank Ben D Exelle for being my supervisor, co-author, and for spending so much time on valuable discussions, for his critical comments and for always being in reach, despite the distance separating us. Without him, this thesis would not exist at all. Special thanks to Judith Westeneng with whom I worked very closely during the fieldwork, never met such a dedicated and hard working person. Thanks to Ruerd Ruben for his support and encouragements during the whole research master. Evert Ketting and Janine Huisman for their support during the preparation of the training sessions in Mwanza. From the Radboud University Nijmegen I would also like to express my gratitude to Thomas de Hoop, Rob Baltussen and Theo van der Weegen, who provided feedback on various aspects of my thesis. I would also like to thank the people xi working at the computer department of our faculty, especially Igor Moonen. Without him, we would still be engaged with data entry. On the more personal level, I would like to thank my fellow students from the Research Master Social and Cultural Science for their friendship, during the last years. Henny, Dim and Femke for their generously support and love, which has always been there. Without them, I would have never ended up here. Neither last nor least I want to thank Stan and Roos. They brightening up my life and give me strength like no one else can. The last, very, very special thanks go to Marjanke, for her never-ending support and above all for loving me. Nijmegen, May 2011 Bart van Rijsbergen xii 1 Maternal Health, an Introduction to a Persistent Challenge Every minute of every day somewhere in the world a woman dies as a result of complications arising during pregnancy and childbirth The majority of these death are avoidable (WHO, 1999: p.4) At some point in their lives, most women aspire to become pregnant, which in essence, is a normal and life-affirming process. Pregnancy is often a defining phase in a woman s life, and pregnancy can be a joyful and fulfilling period, for her both as an individual and as a member of society. However, it can also be one of misery and suffering, when the pregnancy is unwanted, or when complications or adverse circumstances compromise the pregnancy, cause ill health or even maternal death. Pregnancy may be natural, but that does not mean it is problem-free (WHO, 2005). It carries with it serious risks of death and disability. Each year, an estimated 210 million women become pregnant worldwide, of which some eight million suffer life-threatening complications related to pregnancy and childbirth, and many experience long-term morbidities and disabilities. In 2000, an estimated 529 thousand women died during pregnancy and childbirth from largely preventable causes (WHO, 2004b). The past decade has brought increasing recognition that health and development are intertwined. Nevertheless, reproductive health still is a marginalised field of study within development studies (Dejong, 2006), despite its cross-disciplinary nature which makes it a very suitable subject for the area of development studies. This chapter starts with the issue of maternal mortality reduction and in the second paragraph, an overview is given of previous literature on the demand for maternal health care. 1 Chapter Reducing Maternal Mortality Despite the recognition of the importance of combating the high burden of pregnancyrelated death and suffering and the launch of many initiatives, the progress required to meet the Millennium Development Goal to reduce maternal mortality rates by 75 percent between 1990 and 2015 (MDG5) lacks far behind. Although globally a decline of 34 percent was reported from the levels of 1990, developing countries continued to account for 99 percent of the maternal deaths. Absolute numbers of mortality are high in south-central Asia, while maternal mortality rates remain especially high in sub- Saharan Africa (Lawn, Cousens, & Zupan, 2005). To reach MDG5, an average annual decline of 5.5 percent in the maternal mortality ratio (MMR) from 1990 to 2015 is needed. The estimated mean annual decline of 0.6 percent in the maternal mortality ratio of Tanzania since 1990, is far too little to meet the target of reducing the maternal mortality ratio by three quarters, between 1990 and 2015 (WHO, 2010). There is global consensus on what must be done to improve maternal health. The importance of skilled care during pregnancy, childbirth and the immediate postnatal period has been recognized as crucial for the birth outcome (WHO, 1999). A skilled attendant is defined as an accredited health professional - such as a midwife, doctor or nurse - who has been educated and trained in the skills needed to manage normal pregnancies, childbirth and the immediate postnatal period, and in the identification, management and referral of complications in women and newborns (WHO, 2004a). Birth can take place in a range of appropriate places, from home to a range of health facilities, depending on availability and need. The World Health Organization does not recommend any particular setting and state that home delivery may be appropriate for normal delivery, provided that the woman is assisted by a person who is suitably trained and equipped and that referral to a higher level of care is an option (WHO, 1999). The choice between home or facility confinement is a much-discussed topic in many developed countries. It is seen by some as a women s right to choose, but others express their concerns regarding safety. Home deliveries in developed countries are assisted by a skilled professional most of the time, and referral to a higher level of care is a realistic option in most cases. The majority of deliveries in developing nations however, are conducted at home without the assistance of a physician, nurse, or a midwife (Worell, 2 Maternal Health, an Introduction to a Persistent Challenge 2001), and referral to higher level of care is often difficult or impossible. Undesirable outcomes of home delivery are documented in developing countries by multiple authors (Koblinsky, Campbell, & Heichelheim, 1999; Wagle, Sabroe, & Nielsen, 2004). In a study conducted in Tanzania, Walraven, Mkanje, Roosmalen, van Dongen & Dolmans (1995) found that home births resulted in an almost three times higher perinatal mortality compared to facility confinement. 1.2 The Demand for Maternal Healthcare The success of efforts to improve the use of reproductive health care depends on an understanding of the importance of factors affecting the demand for services. There is a lot of literature about factors that may impede or promote facility delivery in sub- Saharan Africa (Allen, Kayaleh, Pryor, & Zoerhoff, 2008; Appiah-Kubi, 2004; Gage, 2007; Mrisho et al., 2007; Olsen, 2009; Parkhurst & Ssengooba, 2009). A lot of research has already devoted attention to investigate the effect of demand-side characteristics on health service use (Appiah-Kubi, 2004; Arends-Kuenning & Kessy, 2007; Sahn, Younger, & Genicot, 2003). It is described as a complex behavioural phenomenon with a true plethora of factors influencing the outcome. The use of maternal health care was shown to be correlated to socioeconomic status (Feinstein, 1993; Gage, 2007; Goldenberg & Jobe, 2001; Graham, Fitzmaurice, Bell, & Cairns, 2004; Mrisho, et al., 2007), demographic factors including age, gender and ethnicity (Alexandre, Saint-Jean, Crandall, & Fevrin, 2005; Andersen, 1995; Zabin & Kiragu, 1998) as well as mother s education which is positively related to the use of safe motherhood services (Addai, 2000; Alexandre, et al., 2005; Appiah-Kubi, 2004; Elo, 1992). Furthermore it was demonstrated that it correlates with economic activity (Addai, 2000; Henderson et al., 1994), household composition (Glick, Razafindravonona, & Randretsa, 2000; Sahn, et al., 2003), current pregnancy (Becker, Peters, Gray, Gultiano, & Black, 1993) and total number of deliveries (Alexandre, et al., 2005; Elo, 1992). A lot of research focuses on multiple barriers to maternal health care and the large variation in utilization for different maternal health services. This type of research however, is difficult to translate into coherent health policies (Kruk, Paczkowski, Mbaruku, Pinho, & Galea, 2009). Especially given that limited resources coupled with an 3 Chapter 1 unlimited demand for health care, means that policymakers have to make decisions regarding the allocation of scarce resources across competing health-system interventions. The extent of consumer involvement in this decision making process has traditionally been limited both at the micro- (doctor-patient) as well as the macro level of planning and developing healthcare services (Shackley & Ryan, 1995). But only when healthcare-planners know what factors are important to consumers of health care, will they be able to plan a service that meets the demands of users. Conjoint analysis, which has its origin in mathematical psychology, is used to elicit consumer s views on health care and has the potential to provide evidence-based information that is relevant to policymakers. Research on preferences of women in developing countries for maternal healthcare is limited, and the importance of the attributes of delivery care facilities for women s preferences has been amply studied. There is evidence that women s preferences for delivery care are influenced by costs (Gage, 2007; Graham, et al., 2004), distance to the health centre (Dor, Gertler, & van der Gaag, 1987; Frederickx, 1998; Thaddeus & Maine, 1994) and quality of care (Kruk, Paczkowski, Mbaruku, et al., 2009; Leonard, Mliga, & Haile Mariam, 2002). An even less-documented area is related to women s variation in preferences for obstetric care facilities, and the influence of emergency conditions on these preferences. This is an important gap in the literature, as most cases of maternal death occur in case of complications. It is assumed that at least 15% of all pregnant women develop serious obstetric complications (WHO, 1994). Moreover, existing evidence is mixed on the influence of emergency conditions. On one hand, there is evidence that suggests that women do not change their behaviour when facing emergency conditions. According to Jahn, Kowalewski, & Kimatta (1998), despite frequently occurring complications, there is a reluctance to change the pre-selected delivery setting in case of severe complications. Others, however, assume it more likely that emergency conditions influence healthseeking behaviour (Thaddeus & Maine, 1994). This study analyses the preferences for obstetric care facilities of women in the Tanzanian Lake region and how these correlate with important individual characteristics, such as number of births, education, wealth and previous complications with delivery. 4 Maternal Health, an Introduction to a Persistent Challenge Furthermore we assess the influence of emergency conditions on preferences for obstetric care facilities. The assessment of these preferences and the variance within these preferences, could be helpful to improve targeting effectiveness of reproductive health initiatives. In particular, this research is expected to provide new insights in how specific reproductive health programmes can increase the use of skilled assistance during delivery. In the subsequent chapter 2 we will sketch the maternal health situation specific to Tanzania. In chapter 3 we will outline the conceptual framework used in our analysis, based on the behavioural model of Andersen (1968, 1995) and elaborate on the hypotheses used. The data we use is collected within the framework of the population, reproductive health and economic development project named POPDEV. This survey is conducted by the Centre for International Development Issues Nijmegen (CIDIN) and Muhimbili University of Health and Allied Sciences (MUHAS) and aims to collect data on a wide range of topics, including demographics, fertility preferences, birth history, family planning, general health, intra-household relations and women s economic activities. The data and the various variables used in the analyses are discussed in detail in chapter 4. In chapter 5 we discuss the influence of emergency and poverty conditions on women s preferences. In chapter 6 we elaborate on heterogeneity in preferences, along with a discussion of the results and concluding remarks in chapter 7. 5 2 Tanzania in Context This chapter provides some background information on Tanzania, its demographical, political and socio-economic situation. Special attention is paid to the Tanzanian health care system in paragraph four and in the last paragraph an overview is given of the maternal health services in Tanzania. 2.1 Geography Jamhuri ya Muungano wa Tanzania, also called The United Republic of Tanzania, is the largest country in East Africa, covering more than 945 thousand square kilometres. It is known b
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks