The three papers in this dissertation apply quasi-experimental and experimental methods for causal inference to the fields of health and development economics. The first paper exploits a plausibly exogenous reduction in the supply of health care in New York City caused by an historic storm to separately identify the impacts of health care access, weather shocks, and their interaction on chronic conditions. The second paper investigates how formal credit, informal risk sharing, and insurance interact. I exploit a natural experiment wherein tens of thousands of microfinance borrowers across rural Haiti received a quasi-random value of insurance benefit in the aftermath of catastrophic hurricanes. I show that subsequent demand for credit is increasing in the value of insurance, that insurance has decreasing marginal effects on the demand for credit, and that formal insurance increases the fungibility of the informal social ties underpinning risk sharing arrangements. Additionally, I present evidence suggestive of collusion in peer-based claims processing as a function of informal financial proximity. Finally, the third paper reports results from a cluster-randomized trial in rural Haiti that I led. It provides the first causal evidence that delivering basic health goods through microfinance institutions, which offer a platform that reaches 200 million rural poor households globally, is clinically effective.

Nurses and midwives are well placed to improve the health outcomes of patients in a number of areas, so the fields of nursing and midwifery must include the ability to conduct rigorous research, synthesize findings into relevant evidence, and use research to inform practice. However, clinical nursing issues in regions such as the Eastern Mediterranean Region (EMR) are often under-researched. The three projects included in this proposal are part of a larger study funded by the Columbia University President’s Global Innovation Fund in collaboration with the Office of Global Initiatives of Columbia University School of Nursing. This dissertation includes three aims: (a) conduct a scoping review of published clinical nursing research to assess the current state of research in the region, (b) conduct a formal program evaluation of a Research Summit that identified clinical nursing and midwifery research priorities and developed an action plan for the EMR, and (c) describe perceptions of barriers and facilitators to research utilization among nurse leaders in the EMR. The results of the scoping review demonstrated that there was limited clinical nursing and midwifery research that has been conducted in the Region. The program evaluation indicated that the Summit was successful and a number of actionable projects have been carried out as a result. Furthermore, the results of the Barriers Scale (Funk et al., 1991) showed that the main barriers to research utilization were lack of funding and resources, lack of support, lack of interest, and lack of training while the main facilitators were placed into the categories of improving support and research culture, resources, and education and training. GDP also significantly impacted the extent to which participants experienced barriers to research utilization. It is important that more Region-specific research be carried out. In order to do this, the research skills of nurse/midwife researchers need to be enhanced and the dissemination of their research supported. Nurse and midwife researchers in the Region should develop consensus regarding specific regional clinical research topics to be given priority and provide support so that nurse and midwifery researchers overcome any barriers they face regarding utilizing clinical research in their practice so that patients can be provided with better and safer care.

Sometimes experts need to provide potentially upsetting advice. For example, physicians may recommend hospice for a terminally ill patient because it best meets their needs, but the patient and their family dislike this advised option. The present research examines whether regulatory non-fit could be used to improve these types of situations. The findings from eight studies in which participants imagined receiving upsetting advice from a physician demonstrate that regulatory non-fit between the form of the physician’s advice (emphasizing gains vs. avoiding losses) and the participants’ motivational orientation (promotion vs. prevention) improves participants’ evaluation of an initially disliked option. Regulatory non-fit de-intensifies participants’ initial attitudes by making them less confident in their initial judgments and motivating them to think more thoroughly about the arguments presented. Furthermore, consistent with previous research on regulatory fit, the studies show that the mechanism of regulatory non-fit differs as a function of participants’ involvement in the evaluation of the option.

In American society, where the biomedical profession is institutionalized and its therapies dominate health care, a decision to deviate from prevailing norms can be considered remarkable. Yet research done by Eisenberg et al. (1993) found that unconventional or alternative medicine had an "enormous presence" in U.S. health care and that one in three persons had utilized an unconventional therapy in 1990. The objective of this thesis was to explore this phenomenon. Individuals who had used alternative and biomedical intervention to treat a disorder were recruited through practitioners of alternative therapies. Through semi-structured interviews, case histories focusing on the illness trajectory and treatment choices were obtained. These were analyzed following a method outlined by Glaser and Strauss (1967) for generating grounded theory. The specific purpose of this study was to examine the case histories for a basic social process and to identify key variables which influenced the utilization of both a biomedical and alternative intervention. It was the position of this thesis that treatment responses involved more than the decision to accept or decline a specific therapy. Rather, choices were made by social actors embedded within unique social worlds which were differentiated by social variables (e.g., culture, ethnicity, social class) particular to the individual (see, among others, Brown 1989, Fitzpatrick 1989, Freund and McGuire 1991, Gerhardt 1989, Kleinman 1988, Mumford 1983). Treatment decisions were made within a framework of internalized patterns of interpretation and meaning, dynamically influenced by ongoing interactional and relational patterns; an individual's treatment response could be conceptualized as an emergent, multi-phased process (Pescosolido 1992). Analysis of respondents' case histories supported the notion of a dynamic response process. Individuals described an evolving illness trajectory directed at achieving a desired state of wellness. Disparities between treatment outcomes and the desired goal prompted an evaluation and choice-making process. All respondents had developed a health care conceptualization in which biomedical and alternative therapies were either categorized as effective for specific conditions or occupied a particular place on a health care continuum, thereby perceiving a scope of treatment options beyond those available within biomedicine.

Advance directives (ADs) convey consumers' wishes about accepting or refusing future treatment if they become incompetent. The issue of making a particular AD, more commonly referred to as a living will, is the focus of this thesis. The typical direction of the living will is that life-sustaining activities such as the provision of mechanical ventilation should be withheld so that a person may die what is hoped to be a 'peaceful death'. Clearly the whole basis of the thinking behind the recognition of ADs is that patients' wishes should prevail. ADs have been championed by some as a means of preserving both dignity and autonomy at the end of life in the face of increasing medical advances in technology to preserve life indefinitely. ADs are seen as a means of promoting peace of mind in will-makers, of allowing carers and family to honour the person's wishes and of stimulating communication between all involved parties.While at present there is no statutory right for people in New Zealand to make ADs, it is considered they have rights to do so in common law as consumers of health and disability services in this country. Little is known about the views of New Zealanders making ADs or their justification for doing so. This small qualitative study, using a discourse analysis approach after Potter and Wetherell (1987), aimed to investigate how the participants justified making ADs. Six people were interviewed and the transcriptions were analysed identifying three dominant interpretative repertoires and three corresponding subject positions.The findings indicated that the participants positioned themselves: as independent self-determining individuals who knew when they were ready to make ADs after witnessing undesirable deaths of close family members; as judges of knowing when inappropriate treatments lead to undignified deaths; and as concerned parents who want to relieve their families of uncertainty in the future when making surrogate end-of-life decisions for them. Witnessing a prolonged family member's death in the past was a contributing factor to the participants making ADs. The participants' recall of these events led them to make their wishes known in advance so that their families, in turn, would not have to go through a similar experience at the terminal stages of their lives. The overarching motivations for formalising ADs was to avoid having life artificially prolonged by receiving life-sustaining treatments as well as the desire to die a dignified death.This study highlights the need for healthcare professionals to value the importance of advance planning with well adults before they lose the capacity to give informed choices at the end of life. The taking of a values history as part of this advance planning may inform family and healthcare professionals about peoples' general values and at the same time confirm and record end-of-life choices for future reference.

Doctor of Philosophy(PhD) / This thesis consists of a series of studies of new models of multidisciplinary community health care in four compartments. These compartments are acute, subacute, outpatient and maintenance care. The purpose of the individual studies is to demonstrate the benifits of munltidisciplinary community health care in delivering alternatives to current practice by replacing hospital care or improving traditional community care.

This research was designed to see whether the clinical method espoused by Moira Stewart et al in the book "Patient-Centered: Transforming The Clinical Method" is appropriate for Pakeha general practitioners to use in clinical consultations with Maori patients. This thesis uses qualitative methodology. One of my supervisors and I selected from the kuia (old women) and kaumatua (old men) of Hauraki those whom I would approach to be involved. Nearly all responded in the affirmative. The kuia and kaumatua talked about their tikanga, about the basis of tikanga, about the spirituality of their Maori worldview. They talked about the need to maintain their tikanga, about qualities that they respect. They described different roles within Maoridom, especially those of the kuia, whaea (mothers) and Tohunga (experts). They refer to a GP as a Tohunga because of the GP�s special expertise. The GP is able to use his or her special expertise to heal Maori patients, but needs to be able to get through barriers to do so. They are also clear that Maori and Pakeha live in two different worlds which can merge in some circumstances. I came to two conclusions. The first is that the elements of Patient-Centered Medicine are relevant to the consultation of a Pakeha GP and Maori patient, and provides a framework that is productive. The second conclusion is that there is a better framework for working with Maori patients, within which Patient-Centered Medicine can be practiced more effectively. Maori already have a framework (tikanga) in which they function, and if in their settings, especially the marae, he or she is welcomed and has a place in their world; tikanga accommodates the GP as a Tohunga and Maori respond to him or her as such. In summary, a Pakeha GP who has some knowledge of tikanga or Maori culture and who has a basic knowledge of the Maori language of tikanga of Maori culture and who has a basic knowledge of the Maori language can work very well for his or her Maori patients by working within the framework of Tikanga Maori and by being patient-centered in consultation.

Thesis (M. P. A.)--University of Hong Kong, 2007. / Title proper from title frame. Also available in printed format.

Degree awarded (2003): DPhPH, Health Services Management and Leadership, George Washington University / Abstract Community partnerships are an increasingly popular strategy for improving community health. This popularity is based less on evidence than on rhetoric. This research developed and tested a systems model of partnership to improve the practice of collaboration in public health. Basing the need for partnerships on the multi-sectoral nature of health, the model used open systems theory to set out requirements for partnership. Institutional theory suggested that problems faced by partnerships may result from partners meeting requirements for legitimacy. Change is, therefore, required, both in organizations and in their institutional environment. Using exploratory case studies, the study design involved site visits to two community health partnerships (West Virginia Community Voices and Healthy New Orleans). Mixed qualitative methods included semi-structured interviews, focus groups, and document review. Analysis involved interpreting informants responses in terms of evidence representing the model and for new elements. Evidence from practice suggested several revisions to the model. One involved applying a typology of organizational affiliation, with partnership toward one end of the continuum. Use of this typology permitted an extension of the model to understand the form of affiliation practiced by Community Voices and of Healthy New Orleans. Multiple opportunities to network and build coalitions in Community Voices led to increased chances of success in achieving health improvement goals. Networking opportunities for individual volunteers led to an informal Healthy New Orleans organization. Results of this research led to an analytic fit between the two sites and the community health partnership model. Recommendations are offered for practice, research, and for funding agencies. With further research, the model can be used to develop practical tools to guide and assess partnerships as a strategy to improve health, as well as to identify environmental barriers to partnership and strategies for change. / Advisory Committee: Kathleen Maloy JD PhD (Chair), Vincent Lafronza ScD, Chris Johnson EdD

This study aims to identify conceptions of the body, as well as "rational" treatment among primary care practitioners (PCPs) and emergency medicine practitioners (EMPs) dealing with managed care plans. I conducted ethnographic study throughout 1996, and interviewed six PCPs and four EMPs. In the course of my interviews, I discovered that EMPs' and PCPs' perceived meaning and power in regards to treatment of their patients has been increasingly resituated by the power of scientific discourse as utilized by technocrats (such as consumer, provider, and insurance organizations who formulate managed care plans). Technocrats have developed scientific measurements to monitor physicians' performances, quantifiably or scientifically. Technocrats conceptualize ideal treatment as a cost effective care. They guide PCPs to supervise ideal treatment of the managed care delivery networks, providing care without referring patients to specialists and administrating to hospitals. Consequently, PCPs' power to pursue the newly conceptualized ideal treatment influences other arenas of specialty, such as EMPs. On the other hand, EMPs develop preferred treatment which is formulated through their medical school education and clinical experiences. The preferred treatment is associated with time because EMPs have to treat acute conditions of their patients within a limited amount of time. However, EMPs sometimes cannot complete their preferred treatment because they have to send their patient back to the patients' contracted PCPs in order to save health care expenditure. Technocrats implemented the engineering concept of quality control, and the concept is incorporated into the principle of managed care plans, and preventive medicine. As a result, the managed care plan networks become like production lines of large manufacturing factories, and PCPs work as laborers to maintain the bodies of enrollees in healthy conditions. / Graduation date: 1998