How country matters: studying health policy in a comparative perspective

5How Country Matters:

Studying Health Policy in Comparative Perspective

S i r p a W r e d e

INTRODUCTION

The aim of this chapter is to offer an introduction to the use of qualitative methods in comparative health policy. The argument is that when the cases designed for comparative health policy research are composed in a reflexive way, qualitative research provides the most promising means to forward an in-depth, context-sensitive understanding of health policy and of country as a macro-social frame for the social and cultural organization of health care systems.

The use of qualitative methods has been commonplace in health policy analysis from the outset (Mechanic, 1975), and, considering the complexity of the subject of research, this is hardly surprising. Policy making is a complex process politi-cally as well as socially, involving multiple actors and arenas employing diverse material as well as symbolic resources. Qualitative methods offer the means to tackle this complexity of policy making.

Recent literature has, however, critiqued the lack of methodological clarity in comparative health care policy, calling for greater reflexivity in the use of quali-tative methods and particularly, in the use of the case study methods (Walt, et al., 2008, p. 312). In policy analysis, the identification of a ‘case’ is a process where the researcher typically actively selects to examine some instances of the policy process and excludes others, in an effort to be able to identify a meaningful

5437-Bourgeault-Chap05.indd 885437-Bourgeault-Chap05.indd 88 4/15/2010 9:57:26 AM4/15/2010 9:57:26 AM

HOW COUNTRY MATTERS 89

configuration of events, actors and processes. When examining only one case, however, there is a risk that the analysis remains descriptive, without the means to identify more general dynamics of policymaking (Clark, et al., 2002). For researchers interested in explanation and generalization, comparison offers a means to systematize qualitative analysis. To compare is to classify with the aim to explain, and unlike description, comparison therefore presupposes analytical frameworks that surpass the specific case so that the cases to be compared can be interrogated in relation to each other, allowing the search for both similar and different dynamics (Bradshaw and Wallace, 1991). Comparison, however, is a method that needs to be used reflexively. Comparative health policy, to be dis-cussed in this chapter, provides a good example of both the potential and the challenges of comparative qualitative research. Even though comparisons within countries as well as within and between regions and organizations are of interest for policy analysis, the term comparative health policy traditionally refers to comparisons across countries, states or nations. This definition is in the starting point for this chapter that considers how and why country matters in comparative research on health care policy.

In the following, I first discuss qualitative research on health policy, identify-ing the rationale for the reflexive comparative approach suggested in this chapter. The subsequent sections focus first on qualitative case studies as the key meth-odological approach and the principles for how cases are developed. The next section discusses the use of qualitative methods in comparative health policy, presenting first documentary approaches, involving use of materials that have been created by other actors independent from research. Then ethnographic approaches and interviewing are discussed. After the presentation of the key methods, the remainder of the chapter considers key analytical concerns in research design in qualitative comparative studies. The conclusion summarizes the rationale for the main argument of the chapter.

HEALTH POLICY RESEARCH AND THE RENAISSANCE OF QUALITATIVE METHODS

Health policy scholars often emphasize the analytical framework they use at the expense of specifying the methodological design of their research. The neglect to discuss research methods may be associated with the impact of the positivist tradition in social science that favours quantitative research designs. Positivist research tends to use qualitative methods out of necessity, rather than by choice, as the aim is to find universal, culture-free explanations. Approached from this per-spective, qualitative data appears ‘idiosyncratic’ and unavoidably ‘descriptive of particular locations and periods of time’ (Mechanic, 1975, p. 43), thus falling short when compared with the ideal data that would allow for quantitative analysis.

In the same vein, the assumptions underpinning the analytical frameworks employed may also serve as an explanation of the absence of reflexivity concerning


THE SAGE HANDBOOK OF QUALITATIVE METHODS IN HEALTH RESEARCH90

the use of research methods. Early health policy scholars, working in a tradition that assumed universal social laws, commonly took for granted that Western health care systems were following a shared path towards modernization. This convergence hypothesis maintained that Western industrialized countries were separate societies evolving towards ‘increased uniformity of social structures’ (Field, 1989, p. 13). The country as a context for health care delivery systems accounted for particularistic variation, but not in ways interesting to research. The analysis of health policy compared instead the assumed outcomes of health care delivery systems with the aim to compare their effectiveness in advancing medical progress. Sociology, health administration and health economics emerged as distinct approaches to the comparative study of health care delivery systems, reflecting attention to society, organizations, and economics respec-tively, all organized within the country context (Mechanic, 1975). In this vein, much of policy analysis is concentrated on developing tools for the evaluation of policy options. A recent textbook offers an example of the approach, with its focus on themes such as resource allocation, priority setting, or regulation of professionals (Blank and Burau, 2007). The policy development perspective defines the study of health policy as analysis of ‘those courses of action taken by governments that deal with the financing, provision or governance of health services’ (Blank and Burau, 2007, p. 2, emphasis added). Even though qualitative methods are commonly used, their role in research design tends not to be given attention, as the emphasis is on developing typologies that capture the various generalizable models of policy making (see Blank and Burau, 2007).

Another strand in the literature on comparative research is rooted in critical and social constructivist perspectives in the study of health policy. Here a broader understanding of the topic is adopted. Scholars who work with the more open-ended approaches emphasize the view of health policy as a contested area of politics rather than an ultimately strategic process, primarily mastered by the government (Walt, 1994). In this vein, health care policy is viewed as enmeshed in economics, ideologies and culture (Green and Thorogood, 1998) and health policy is most fruitfully conceptualized as processes occurring at multiple levels that can be analytically separated. A leading proponent of this approach, Walt (1994), argues that she recognizes eclectically the relevance of multiple theoreti-cal approaches to capture the complexity of the ‘process and power’ in health care policy. In addition to the international and national level, she emphasizes the face-to-face level of ‘the actual processes of policy and the actors involved at every stage’ (Walt, 1994, p. 6). This is in line with how comparative social scien-tists in many areas of research have since the late 1980s promoted the more reflexive use of qualitative methods, underlining the need to take the complexity of data into account in comparative analysis (Øyen, 1990). Reflecting the renais-sance of qualitative methods, such scholarship on comparative health care policy approaches policy from a social constructivist position, treating policy as inevi-tably shaped by its social context (e.g., Lee, et al., 2002; Freeman, 2006; Gilson and Raphaely, 2008).



The social–constructivist view of health care policy recognizes that ‘policy’ can itself be defined in many different ways. From this perspective, the ‘making’ of policy may involve a complex set of actors not limited to the government or the organizations internal to the health system (Walt, et al., 2008, p. 310). In this vein, Buse, et al. (2005, p. 6, emphasis added) suggest that it is useful to think of health care policy as embracing ‘courses of action (and inaction) that affect the set of institutions, organizations, services and funding arrangements of the health system’. The multilevel understanding of health care policy helps to explore the sociocultural embeddedness of health care policy, that is, the way that national health policies and structures reflect deeply rooted values and norms which differ between societies. Saltman’s (1997) analysis of the explanatory power of the convergence thesis versus that of the embeddedness thesis of divergence sug-gests, however, that it is not fruitful to treat sociocultural embeddedness as the new grand theory for comparative research on health policy. Rather, in his view, comparative research should take as a starting point a complex interplay of part-ners between and within countries, dependent upon economic, cultural, historical, and geographical as well as political, elements. This means that while some aspects of health systems between two countries may be converging, at the same time other aspects may be diverging. In Saltman’s view, therefore most suitable to capture such complexity is the notion of a convergence/divergence mix (Saltman, 1997, p. 452). This position is in line with current developments in national health policies that are increasingly influenced and even displaced by global decisions as well as domestic actions. Qualitative methods, and particu-larly comparative case studies, are well suited precisely for capturing such complexity (Bradshaw and Wallace, 1991).

CASE STUDIES ORIENTED TOWARDS INTERNATIONAL COMPARISON

Even though it is common to refer to qualitative comparative research on health care policy as case study research or case-oriented research, no unified, self-conscious methodological approach of conducting case study research oriented towards international comparison exists. Rather, there are several, more or less integrated research traditions that build on the use of historical and qualitative methods that became termed the ‘case-oriented tradition’ post-hoc, to separate them from a quantitative variable-oriented approach (Ragin, 1987, p. ix). In the 1980s, the topic of comparative methodology emerged as a theo-retical concern parallel to the renaissance of qualitative methods, inspiring the more reflexive use of qualitative case studies in comparative research (Bradshaw and Wallace, 1991). To be sure, one strand in comparative literature has engaged with the complexity of case study data by developing an approach that builds on Boolean algebra (Ragin, 1987). This ‘small-N research’ has, however, been shaped by its close association with statistical analyses (see Caramani, 2009).



A methodologically reflexive use of qualitative methods in case-based research entails the comparison of relatively few cases that are carefully chosen for well-specified theoretical and/or substantive reasons (Bradshaw and Wallace, 1991, p. 156). The qualitative case-oriented method to comparison involves a compre-hensive and detailed approach that typically combines several qualitative methods (Lian, 2008; see, for example, Caronna in this volume). Case research has become advocated particularly as the method of choice when the emphasis is on theoriz-ing, but the arguments for reflexivity are both analytical and practical. Case-study research as such is a time and resource intensive process, and comparative case studies introduce further challenges, involving relatively expensive work across multiple languages and cultures (Walt, et al., 2008, p. 313). A thoughtful approach to qualitative methods thus both saves time and money, and secures a higher quality of research. The following sections discuss common challenges, consid-ering the different methods that can be used in case-oriented research.

ASKING COMPARATIVE QUESTIONS ABOUT POLICY: CONSTRUCTING THE ‘CASE’

All kinds of comparative social research face the shared challenge of defining the object of comparison in a consistent way, when shared definitions of basic terms are missing. Health policy terms tend to carry with them the histories of specific policy programs in particular countries, and particular conceptions of, for instance, service organization or health problems are naturalized in that terminology. This is not unique to comparative health care policy: all areas of comparative studies struggle with problem of translating complex concepts between languages and societies (Clark, et al., 2002, p. 277). A political term such as ‘liberalism’ may have very different meanings in different political traditions. Similarly, the orga-nization of political institutions, such as the state, varies greatly and particularly the state has very different powers in health policy in different countries. The consistency of the terms used in different languages becomes an even more com-plicated matter in a historical perspective and when the terms refer to ideas that are influential in political processes (Dutton, 2007, pp. 24–26). The analysis of such terms therefore becomes one of the tasks that researchers have to engage with in order to be able to define the phenomenon under study. It is important to note that even such shared folk concepts as doctor, nurse, or patient mean differ-ent things in different contexts (Mechanic and Rochefort, 1996). Organizational roles in health care are complex products of long-term developments in health policy and therefore, in the words of an early scholar, everywhere situated in ‘a defining and delimiting institutional matrix’ (Davies, 1979, p. 515). She rec-ommends a study of that matrix rather than the variation in occupational roles as a more promising avenue to achieve context-sensitive understanding.

Davies’s study is an example of how a shared language does not guarantee the consistency of terminology. The Anglo-American context and the Commonwealth



are examples of how a shared language as well as numerous historical links can facilitate research. By choosing cases with cultural affinity, the researchers hope to avoid the problems of translation and construct their cases with greater preci-sion. With reference to J.S. Mill’s method of difference, such strategy in case selection is often referred to as the choice of ‘most similar’ cases, another pos-sible strategy being a focus on ‘most different’ cases (Ragin, 1987, pp. 44–49). Assumed cultural affinity and a long tradition of comparisons between certain countries may, however, become blinders that are not present in fresher com-parisons. Furthermore, from a social constructivist perspective, a case is never representative of a category of cases but unique. Accordingly, each case needs to be considered in its own terms, rather than assuming that a shared term signifies shared culture. Thus, the issues of translation in a comparative research project are never merely technical, rather translation lies at heart of the development of robust concepts that ‘travel’ between contexts (Wrede, et al., 2006). This means that the analysis of terminology is a key avenue to an understanding about insti-tutional matrices, cultural ideas, political traditions, policy concerns et cetera. Therefore, a shared language does not remove the challenges related to language in the construction of concepts.

Compared with the complex variation of the ways similar ideas are formulated in different health policy contexts, physiological phenomena like pregnancy, birth, or death appear to provide conceptual stability in the area of health care around which research concepts can be built. While such phenomena offer fruit-ful foci for comparative analyses, such seemingly definite notions need to be approached with caution. Even when the physiological processes in themselves may be very similar, the social and cultural meanings of these phenomena vary (De Vries, et al., 2001). This is not an argument against comparative analysis. Rather it is a methodological concern to be accounted for (Clark, et al., 2002). One means of analyzing such variation is by examining the complex processes of classification that organize social phenomena (Bowker and Star, 1999).

Taking into consideration the challenges involved in conceptualizing the object of comparisons, it becomes evident that the construction of cases to examine involves the active hand of the researcher who chooses how to define and delimit the phenomenon under study, the level of abstraction on which the phenomenon is identified and the context against which it is considered. With the various con-figurations of health care policy processes worldwide, what is defined as macro, meso, or micro level is unavoidably a matter of empirical consideration, not one of employing fixed previously existing definitions. Thus, cases are to be per-ceived as interpretive constructs rather than objective facts to be identified in an empirical context (Carmel, 1999).

Policy scholarship offers numerous frameworks to employ for a systematic discussion of policy making processes (Blank and Burau, 2007; Walt, et al., 2008). Rather than looking for the ultimate frameworks, a reflexive approach to the construction of cases recognizes both the research interests that drove their development and the intellectual and political context the frameworks were



developed in. Issues in health policy are political everywhere, but the history of how an issue has been politicized varies between policy contexts. Readymade frameworks are particularly problematic when the comparative study involves countries that previously have rarely been included in comparative studies. A grounded approach involves mapping of policy systems and actors as well as tracing of policy processes. The collection of materials in health policy analysis often takes places through a snowball method of combining different approaches to map the complex scene where the phenomenon under study takes place, in order to be able to delimit the core case to be subjected to more detailed study.

Shiffman’s (2007) study of the comparative level of political priority given to maternal mortality in five low-income countries provides an example of a study that builds on pre-existing wider case studies to be able to carry out a more focused analysis. Based on a secondary analysis of five single-country case studies by the same author, the study employs a process-tracing methodology, involving multiple sources of data about agenda setting on maternal mortality in the countries under consideration. In a similar vein, Lian (2008) compared primary health care reforms of the 1990s to examine claims on convergence. The researcher constructed his cases following a predefined list of organizational dimensions that was used as a tool for selecting data and organizing analysis. Aiming to achieve depth rather than breadth in his analysis of ‘similarities and differences in patterns of structural change’ Lian (2008, p. 30) like Shiffman used a combination of different qualitative methods, including documentary analysis and fieldwork, to build his cases. In this vein, the combination of several different methods often found in case research form an asset for the ‘tracing of policy processes’ and identification of ‘patterns of change’ in organizational structures. Diversity of data also allows the triangulation of the types of bias or, more aptly put, perspective in data. Policy data is always situated somehow in relation to the process examined. The rhetoric in political talk and texts is a particular consid-eration to be accounted for in the analysis, especially when tracing policy processes. Not surprisingly, scholars struggle to avoid confusing ‘words with actions’ (Lian, 2008, p. 30). At the same time, it is evident that the expressions of political culture may be particularly interesting to capture in a comparative study.

Health policy analysts are further constrained by the way they themselves are viewed or ‘situated’ as researchers, most markedly, whether they are viewed (by themselves or others) as insiders or outsiders (Walt, et al., 2008, p. 314). Such limiting conditions shape the research, but the researcher can also actively try to influence the conditions of positionality, within certain limits. In comparative research, while it is not possible to become native of some other country context, it is possible to increase familiarity of that context with numerous strategies, and researchers can also choose to collaborate in teams that involve both insiders and outsiders. Reflexivity of the role of health policy research for the policy processes examined is also of importance.

A recent reformulation of case study methods advocates qualitative approaches precisely for their particular potential to grasp the rich ambiguity of politics and



planning in a modern democracy. In his critique of the ‘misunderstandings con-cerning case-study research’, Flyvbjerg (2006) argues that the case study in itself is a result. In his view, the story should be told ‘in its diversity’, allowing it to unfold from the many-sided, complex and sometimes conflicting stories told by the actors in the case (Flyvbjerg, 2006, p. 238). Echoing Walt’s (1994) eclectic approach, Flyvbjerg (2006) explicitly warns against linking the ‘telling of the story’ to the theories of any one academic specialty. In this vein, and supporting both the notion of convergence/divergence mix and the related aim of capturing the story in its diversity, Lian (2008, p. 38) concludes on the basis of his com-parison of primary health care reforms in three countries that each country has its own unique story and even while influenced by global ideas, the present developments represent a continuation of that story.

METHODS OF CREATING DATA FOR CASE STUDIES

The following focuses in greater detail on how researchers gather or, from a social constructivist perspective more aptly, create data on their health policy cases.

Documentary methods

Documentary methods involve the use of already existing materials. Documents are bound to a specific point in time and therefore they can be used to examine, for instance, how a specific actor argued in a specific stage of a policy process. The data content in documents is fixed and therefore the researcher is limited to formulating research questions that the available documents can respond to. Accordingly, it is common for researchers to combine documentary materials with other means of acquiring data, such as interviewing. For some studies the fact that documents are not influenced by the researcher may, however, be a par-ticular advantage, when the researcher is interested in, for instance, spontaneous expressions of policy rhetoric. The researcher may also want to map debates and examine exchanges between policy actors et cetera.

Even though the data exists independently from research and the researcher does not shape the individual documents, documentary research does involve the active role of the researcher who chooses which documents to include in the study, often from a great number of possible materials. These particular condi-tions do not, however, exclude the multiple uses of documents if such conditions are taken into consideration (Prior, 2003).

Documentary research also requires language expertise as well as in-depth knowledge of the society under scrutiny (Dutton, 2007, p. 2). A researcher work-ing alone is often constrained in his or her work by the shortage of context-specific expertise (Wrede, et al., 2006). Close collaboration between scholars may help to overcome this challenge. Altenstetter and Björkman (1981), for instance,



broadened the scope of their separate case studies, using an analytic strategy that differs from those implemented in the earlier studies. Instead of focusing on the detailed level, Altenstetter and Björkman (1981) applied a documentary approach to comparatively map the health planning institutions in the five countries included in their respective studies. They adopted a broad time frame to capture the processes they were contrasting and comparing across countries. Their sec-ondary analysis of their work allowed Altenstetter and Björkman to triangulate data, methods, and investigator.

Documentary research is a basic source of data for health care policy study in general and comparative analysis in particular. When discussing their study on the role of multinational health care companies, Waitzkin and his colleagues (2005; Jasso-Aguilar, 2004), for instance, reviewed a wide range of professional and business publications as well as policy documents from multiple different sources and used a selection of these texts as their primary data. One key aim was to allow the identification of the central actors or stakeholders, map their ties with each other as well as the arenas where they act and the strategies they use. The mapping of key actors served the planning of interviews that comple-mented the documentary research. A second key aim achieved with the analysis of documents was to map the historical unfolding of the phenomenon under study. Both aims presuppose the collection of documents for a long research period, in this example 25 years (Waitzkin, et al., 2005). van Herk and his col-leagues (2001) whose comparative study of the politics of medical audit in the United Kingdom and the Netherlands covered a 30-year period also chose a long research period but focused in the final analysis on the early stages of the process in question, arguing that this made the identification of stakeholders easier, when they were able to use their data to observe the gradual appearance of new stakeholders.

While policy documents are commonly used to map policy processes and contexts and to identify and trace policy actors and stake holders, social con-structivist approaches that consider the role of language, rhetorical argument and stories in framing policy debate remain rare in the study of health care policy (Gilson and Raphaely, 2008, p. 303). A few examples of the potential of such analyses of policy discourse or policy rhetoric, however, do exist. Working in this vein, Freeman (2006, p. 66) holds that documents and reports are principal vehi-cles of health policymaking, the analysis of which allows an examination of the belief systems of governments with focus on how normative ideas such as equity are converted into health policy. Arguing for the importance of taking account of cross-national learning, Freeman’s (2006) study focused on the travel of ideas as well as ‘translation processes’ between different cultural contexts, providing an example of how this type of approach can be developed into an international comparison across countries in a way that that avoids treating countries as iso-lated cases. Freeman employs the concept of ‘document architecture’ to examine the production and organizations of documents as well as the ways documents are located in relationships with other documents.



The documentary materials used in health care policy research are usually tex-tual documents. Documentary analysis as a methodological approach can, however, study any cultural product, such as architectural designs, a building or an object (Prior, 2003). Health care policy could thus be examined comparatively by compar-ing hospital design or other examples of material culture. The field of Science and Technology Studies suggests novel research strategies pointing in such new direc-tions. For instance, Bowker and Star (1999) present a historical analysis of the formation of the international classification of diseases (ICD), a transnational stand-ardization devise with a complex history of ‘tangled and crisscrossing classification schemes’ (Bowker and Star, p. 21). In a similar vein, Timmermans and Berg (2003, p. 22) examine how universal medical standards are made, revealing situated knowledge, blurred agency and emergent politics through which ‘the world of med-icine is “remade and molded” through standardization’. Berg (2004) provides numerous examples of how documentary research can be used in country-framed comparative case studies to trace historical processes as well as identify key actors in different contexts. Such analyses reveal how seemingly universal standards are the outcome of interactions between top-down design and bottom-up adoption.

Ethnographic approaches

Though health policy research is seldom based only on ethnographic research, there are examples of both early qualitative studies of health policy and more recent work that in the manner of ethnographic research identifies fields and sites to be examined and includes observation. This means that research links the pro-cess of health policy making with practices in concrete social milieus and actors that the researcher can observe and study. For instance, Björkman (1985, p. 401) describes his in-depth studies of health policymaking in three countries as ‘based on thirty months of field work in Sweden, Britain and the United States (…). The materials are derived from extensive interviews with personnel in the health sector – both governmental and private – as well as from participant-observation in health agencies’. Björkman (1985) did not, however, limit himself to field methods rather his research also included substantial investigation of govern-ment documents and budgets.

In a similar fashion, many other researchers have combined observation in research fields to other methods. Shiffman (2007), for instance, used interviews and observation in combination with documentary research to consider the dif-ferent interpretations of the events and circumstances offered by the different forms of data. Many scholars refer to ethnographic field work as a method of getting first-hand knowledge about society and culture (Lian, 2008). De Vries (2004) has demonstrated how a researcher who is an outsider to a country can find insights about its culture not only in the formal field work undertaken but in all aspects of living in a country. In-depth ethnographies framed by country tend to be limited to a handful of cases, and due to the character of the method, collaborative approaches are challenging.



Recent alternatives to framing the case in field work through country have potential to enrich comparative health policy research. In the emerging literature, research design aims at grasping the different dimensions of ‘global cultural flows’ with studies that involve field work in multiple sites (Clarke, 2005, p. 165). A recent research project that focuses on ‘emerging infections in the global city’ offers an example of what could be characterized as complex com-parison. Working in this vein, researchers who follow a network approach bring insights from ethnographic field work as well as from other forms of qualitative research from different locations to a shared framework that assumes flows and ‘traffic’ between the networked ‘global cities’ (Ali and Keil, 2008). A similar turn to emphasizing connections and interrelatedness across national boundaries has occurred in the field of health geography that links health, culture, and place (Gesler and Kearns, 2002). When examining the meanings of place, field work can be employed as a key research strategy. Comparative health care policy that instead of a country treats a global city, or indeed ‘global ruralities’, as its ‘spatial unit of analysis’ is able to construct new, more focused comparative cases. Such research typically flexibly combines field work with other qualitative and even qualitative methods (e.g., Rodwin, 2008).

Interviewing

Interviewing can be conducted as a part of an ethnographic study or separately. Policy researchers commonly use interviewing to explore policy processes and related action, and to systematize information about policy making and about the views of specific policy actors vis-à-vis the issues in question. This type of interview is often identified as a key informant or expert interview (Flick, 2009). The analysis of expert interviews typically proceeds on a meso or macro level.

Waitzkin and his colleagues (2005; Jasso-Aguilar, et al., 2004) describe a mul-ticountry study examining how multinational corporations in the area of health care have expanded their influence from the United States to Latin America. The interviewing of experts was conducted for a particular purpose, to investigate specific practices of health care delivery (Jasso-Aguilar, et al., 2004, p. 138). The researchers used a semistructured interview protocol with close-ended and open-ended items, a strategy that allows consistency in a situation where the conditions for organizing interviews vary. This type of interview protocol also guides the interviewee to respond to the questions with reference to his or her expertise, rather than with reference to his or her personal experience (Flick, 2009, p. 167).

The choice of respondents on the basis of their specific insights about the topic examined presupposes relatively systematic prior knowledge about the context to be studied so that researchers can identify suitable interviewees (e.g., van Herk, et al., 2001). Close consideration of the strategy for selecting informants is particularly important, when the interviews are used to acquire documentary data where the factual content of the data is important. The materials need to be



treated critically, as any respondent is limited by a partial perspective, and it is the responsibility of the researcher to make sure that the pool of interviewees includes the perspectives identified as relevant for the study.

In drawing upon expert interview data, it is important to take into considera-tion that the interviews often provide post-hoc accounts, where the respondents are able to take into consideration the end result of the process. Particularly in such situations, expert interviews are best combined with documentary data. When both post-hoc and data produced during the process are available, the researcher has possibilities to examine how the interpretations of different dimen-sions of the policy process vary and differ, depending on the contextual aspects framing the ways actors think. Furthermore, as experts in the field, the interview-ees commonly have a view about the role of the researcher and their interpretation of this role may impact how they view their own roles as respondents and experts (Walt, et al., 2008).

When expert interviews are conducted in different countries terminology may vary greatly, reflecting the various traditions of organizing health care (e.g., Dent, 2003, pp. viii–xiv). Documentary research conducted prior to interviews serves as a method of identifying key ‘vocabularies’ and ‘dialectal uses’ of the shared policy terms. Furthermore, when research interviews are conducted in multiple languages, interviewing, as well as transcribing, coding and analyzing the mate-rials, is particularly demanding. Translation to one shared language is expensive, and may introduce new problems when some data is translated and other not, for instance. The presence of several languages in the materials may limit the choice of analytical strategies, particularly those focused on language. An advis-able strategy supported by both practical and analytical concerns is to conduct qualitative analysis of interviews on the original materials and only translate quotations to be used in the reporting to another language. Sometimes research-ers choose to use the original words in their reporting, explaining their meaning, to emphasize the different conceptualization of the phenomenon considered (e.g., De Vries, 2004).

It is helpful to identify expert interviews as distinct from experiential interviews. Experiential interviews focus on the individual experiences of the respondent. This type of interviews is less commonly used in the comparative study of health policy. They typically serve the micro level of analysis, but may be combined with analysis conducted on other levels. The work of Bourgeault and her col-leagues (2001) on everyday experiences of implicit rationing provides a rare example of such a study. While the researchers analyzed the micro experiences of health care providers, they perceived such analysis also as means of accessing the meso context of managerial strategies.

The interview strategies used in expert interviews and in experiential inter-views differ. When semi-structured interviewing is common in expert interviews, experiential interviews typically follow a more loose structure, as researchers aim at obtaining in-depth, experiential accounts, for instance by focusing on epi-sodes rather than themes (Flick, 2009, pp. 185–186). For example, in order to



secure data where the respondents rather than the researchers formulated the issue Bourgeault and her colleagues (2001, pp. 636–637) conducted conversa-tional interviews, posing loose questions and requesting the respondents to discuss their experiences and the issue of rationing that was the topic of the study emerged from these discussions. In the analysis, the researchers organized the accounts of the interviewees around the main claims of managed care. Even though the focus in the data collection was on the micro level, the researchers’ use of strategic sampling allowed them to create robust data. They selected inter-viewees from different types of workplaces and with different work roles, to secure materials with a variety of experiences of the phenomenon under study.

Focus groups rather than individual interviews can be used both with experts and with other types of respondents, but in all kinds of research, this strategy dif-fers from individual interviews as the element of group dynamics needs to be taken into account. For comparative health policy research, the method may offer a means of acquiring rich data in a rapid way, allowing the exploration of a new field or diversity of views among the participants. Focus groups do, however, pose both practical and analytical demands on the researcher (see chapter by Barbour in this volume).

ACHIEVING SENSITIVITY TO CONTEXT

Perhaps most importantly, sensitivity to context in comparative health care policy research means accounting for the institutional arrangements that structure health care policy. Countries have divergent paths (Lian, 2008). In this vein, Immergut (1990) provides an example of how to compare the lobbying efforts of national medical associations in relation to one specific type of policy initiative, govern-ment efforts to enact national health insurance. She examines three cases providing examples of different institutional paths. Immergut (1990, pp. 395–398) concep-tualizes institutional context by envisioning the health care system as a political system with sets of interconnected arenas involving mechanisms that allow a core of political representatives to veto legislative proposals. She further exam-ines the rules of representation within each context in order to ‘predict where such ‘veto points’ are likely to arise’ (p. 396). The framework allows Immergut to pursue the analysis of her country cases paying attention to the necessarily par-ticularistic explanations to why specific veto points occur and how they influence the particular developments in the different countries.

Currently, health care systems all over the world are undergoing rapid and profound transformations in ways that not only fundamentally reorder relations between the different actors but redefine the basic logics of health care delivery and reframe policy making. The new developments cannot be captured by approaches that take the country as an unquestionable context or ‘frame’ for health care policy (Lee, et al., 2002). While the country-context and governmen-tal policies remain important for comparative health care policy, globalization as



well the complex constellations of nonstate actors need also to be taken into consideration. Attention is turning to ‘complex cross-border, inter-organizational and network relationships, with policies influenced by global decisions as well as by domestic actions’, meaning that ‘while government and its hierarchical institutions remain important, all policy analysis must also take into account a range of open-ended, more ad hoc arrangements which increasingly affect decision making’ (Walt, et al., 2008, p. 309).

Recent internationally framed scholarship on health care policy is beginning to tackle the emerging ‘crossborder’ influences on health care policy. Often qualitative case-based methods are used. Cartwright’s (2000) case study of tele-medicine traces with diverse documents how primarily United States-based corporations develop new technologies that transgress national boundaries. These technologies, in turn, transform the construction of social space as well as social identities, as the corporations target profitable regions globally. By tracing the making of new, networked ‘geographies of care’ Cartwright (2000) demonstrates how telemedicine is made into a technique of globalizing populations that are ordered hierarchically by the inherent centre–periphery relations on which the idea of telemedicine builds. Cartwright traces her documentary materials from diverse sources, but qualitative documentary data can also be collected in more focused way. Ranson and his colleagues (2002) examine the public health impli-cations of multilateral trade agreements that standardize policy arrangements across countries. Their documentary analysis centre on how supranational agen-cies apply the trade agreements to concrete disputes. Their focused analyses are combined with more general mapping of the roles that the different supranational agencies play allowing them to examine the broader implications of how institu-tional developments in the global trade environment restructure and reorder health care policy.

A similar agent-centred starting point for the construction of the case for doc-umentary research can also be used in combination with other qualitative methods such as discourse analysis. Freeman (1998) examines how supranational agen-cies set ideas in motion, identifying the Organisation for Economic Cooperation and Development (OECD) as a major player in the politics of health care reform in Europe. Atkinson (2002), in turn, used ethnographic fieldwork in combination with a quantitative survey to examine the implications of the World Development Report of 1993 in the context of a low-income country. In a similar vein, but with an entirely qualitative research strategy, Mills (2006), in her analysis of the maternal health policy developments in Mexico, combined traditional ethno-graphic methods of observation and interviews with analysis of documents. Her ethnography of policy making and implementation provides a rich understanding of the local context, but she also manages to reveal how the implementation of the United Nation’s Millennium Development Goals (MDGs) displaces the authority of the national state. Together, these studies provide examples of how qualitative case-based research can study the impact of the global flows of resources and ideas as well as the role of global policy networks.



TRACING POLICY TRANSFERS AND TRAVEL OF IDEAS

It has already been suggested that treating countries as isolated cases and separate societies may miss the interconnected character of many social develop-ments. It is this interconnectedness that lies underneath the dynamics of the convergence/divergence mix in a country. Interconnectedness is not unique for present developments that commonly are linked to economic, political and cul-tural globalization. Comparative health policy history is rare, but the few examples demonstrate the interplay of ‘common ideals, divergent nations’ (Dutton, 2007). Dutton’s comparison of the development of health care financing and organiza-tion in France and the United States provides also an example of how reflexively chosen shared notions and critical turning-points can offer a starting-point for contrasting the divergent ways the history has unfolded in two distinct but not unrelated sociocultural settings.

Historical analyses, including the analyses of technological and scientific developments in health care (Bowker and Star, 1999) demonstrate that voluntary and involuntary policy transfers and travel of ideas are not unique to the present time, even though the constitution and roles of the agents that drive policy across borders have changed. In the same vein, globalization is not a unitary engine of social change, but a short word for numerous trends. As a result of the new tran-snational features in policy processes, the number of actors and arenas increase in number and reconfigure. The policy environment is becoming populated by complex cross-border, inter-organizational and network relationships (Walt, et al., 2008, p. 309). Qualitative case-based methods are well suited to capture these complexities and when reflexively designed, comparison across cases is possible at the same time as the interconnectedness of cases is recognized. Recent scholarship argues that the study of the rise of new health policy actors and proc-esses and the changes in the content of health policy requires new, network-oriented research approaches (Lee, et al., 2002). In a slightly different vein, Timmermans and Berg (2003), through their study that identifies a ‘standardization movement in health care’, offer an analysis of the role of science and technology in the con-vergence of medical practice. The authors show that this movement has global implications but divergent local adaptations. Their empirical case examples are taken primarily from two countries, the United States and the Netherlands, but without employing an explicit comparative research design. Instead, the research-ers apply different research techniques and combine several case studies to trace the travel of the different ideas that they link to their broader discussion of the ‘dynamic, emergent politics of standardization’. Similarly, efforts to map the impact of economic globalization on health policy lead researchers to trace the stakeholders of transnational policy processes (Waitzkin, et al., 2005).

The scholarship that draws attention to the different networks that shape health care and health care policy is related to a wider theoretical perspective of the rise of network society in a globalizing world. The network society is underpinned by the worldwide spread of ideas, made increasingly easy and rapid by new



communication technologies (Lee, et al., 2002, pp. 13–15). The view of network society does not, however, profess a convergence approach as the network approach is commonly rooted in the recent theoretical turn in social theory to complexity (e.g., Ali and Keil, 2008). The network approaches share with the most recent developments in case study research the aim of rejecting simplifica-tion as a necessary research strategy. Instead, scholars seek to accommodate an analysis of more complex objects of study.

CONCLUSION

As policy processes are changing everywhere, health policy scholars tackle a much larger array of policy actors than was the case when health care policies were less multidimensional. Health policy processes are increasingly complex and new arenas emerge. Traditional comparative health care policy relied on the country as a self-evident way to frame cases, even when the meaning of this macro-social frame was not necessarily analyzed. Now, when alternative ‘spatial units’ as well as networks offer alternative ways to frame comparative health research, researchers are considering more closely how the country matters for health care policy. Thoughtful approaches to conceptualizing the country context demonstrate that the country still does matter, as the world continues to be spa-tially ordered as countries with different political cultures. Case studies of specific health policy processes need to take into account the divergent historical paths of different countries and their relevance for future developments, parallel to that of the emergent global connections. For many, the solution is to turn to qualitative case-based methods that offer, as shown in the foregoing section, many imagina-tive approaches to capture the complexity of health care policy and examine it comparatively. At the very basic level, the demand for methodological reflexivity means that researchers in the field of comparative health policy need to identify their research design and the way they use qualitative methods more clearly than often is the case at present. Future research benefits from approaches that employ the ‘cross-fertilization’ potential of the available interdisciplinary scholarship that is increasingly collaborative and international in its outlook.

REFERENCES

Ali, Harris S. and Keil, Roger (eds) (2008) Networked Disease: Emerging Infections in the Global City. Malden, MA: Wiley-Blackwell.

Altenstetter, Christa and Björkman, James W. (1981) ‘Planning and Implementation. A Comparative Perspective on Health Policy’, International Political Science Review, 2(1): 11–42.

Atkinson, Sarah (2002), ‘Political Cultures, Health Systems and Health Policy’, Social Science and Medicine, 55(1): 113–124.

Berg, Marc (ed.) (2004) Health Information Management: Integrating Information Technology in Health Care Work. London: Routledge.



Björkman, James W. (1985) ‘Who Governs the Health Sector? Comparative European and American Experiences with Representation, Participation, and Decentralization’, Comparative Politics, 17(4): 399–420.

Blank, Robert H. and Burau, Viola (2007) Comparative Health Policy. Revised and updated 2nd edn. Houndsmills: Palgrave Macmillan (1st edn, 2004).

Bourgeault, Ivy Lynn, Armstrong, Pat, Armstrong, Hugh, Choiniere, Jacqueline, Lexchin, Joel, Mykhalovskiy, Eric, Peters, Suzanne, and White, Jerry (2001) ‘The Everyday Experiences of Implicit Rationing: Comparing the Voices of Nurses in California and British Columbia’, Sociology of Health and Illness, 23(5): 633–653.

Bowker, Geoffrey and Star, Susan Leigh (1999) Sorting Things Out: Classification and Its Consequences. Cambridge, MA: The MIT Press.

Bradshaw, York, and Wallace, Michael (1991) ‘Informing Generality and Explaining Uniqueness: The Place of Case Studies in Comparative Research’, International Journal of Comparative Sociology, 32(1–2): 154–171.

Buse, Kent, Mays, Nicholas and Walt, Gill (2005) Making Health Policy. Milton Keynes: Open University Press.

Caramani, Daniele (2009) Introduction to the Comparative Method with Boolean Algebra (Quantitative Applications in the Social Sciences). Thousand Oaks: Sage Publications.

Carmel, Emma (1999) ‘Concepts, Context and Discourse in a Comparative Case Study’, International Journal of Social Research Methodology, 2(2): 141–150.

Cartwright, Lisa (2000) ‘Reach Out and Heal Someone: Telemedicine and the Globalization of Health Care’, Health, 3: 347–377.

Clark, Gordon L., Tracey, Paul, and Smith, Helen Lawton (2002) ‘Rethinking Comparative Studies: An Agent-Centred Perspective’, Global Networks, 2(4): 263–284.

Clarke, Adele (2005) Situational Analysis. Grounded Theory after the Postmodern Turn. Thousand Oaks: Sage Publications.

Davies, Celia (1979) ‘Comparative Occupational Roles in Health Care’, Social Science and Medicine, 13A: 515–521.

Dent, Mike (2003) Remodelling Hospitals and Health Professions in Europe – Medicine, Nursing and the State. London: Palgrave/Macmillan.

De Vries, Raymond (2004) A Pleasing Birth. Midwives and Maternity Care in the Netherlands. Philadelphia: Temple University Press.

De Vries, Raymond, Benoit, Cecilia, van Teijlingen, Edwin, and Wrede, Sirpa (eds.) (2001) Birth by Design: Pregnancy, Maternity Care, and Midwifery in North America and Europe. New York: London.

Dutton, Paul V. (2007) Differential Diagnosis: A Comparative History of Health Care Problems and Solutions in the United States and France. Ithaca: Cornell University Press.

Field, Mark (1989) ‘Introduction’. In Mark Field (ed.), Success and Crisis in National Health Systems: A Comparative Approach. New York: Routledge, pp. 1–22.

Flick, Uwe (2009) An Introduction to Qualitative Research. 4th Edition. London: Sage Publications.Flyfvjerg, Bent (2006), ‘Five Misunderstandings about Case-Study Research’, Qualitative Inquiry, 12(2):

219–245.Freeman, Richard (1998) ‘Competition in Context: The Politics of Health Care Reform in Europe’,

International Journal for Quality in Health Care, 10(5): 395–401.Freeman, Richard (2006) ‘The Work the Document Does: Research, Policy, and Equity in Health’, Journal

of Health Politics, Policy and Law, 31(1): 51–70.Gesler, Wilbert M. and Kearns, Robin A. (2002) Culture/Place/Health. London: Routledge.Gilson, Lucy and Raphaely, Nika (2008) ‘The Terrain of Health Policy Analysis in Low and Middle

Income Countries: A Review of Published Literature 1994–2007’, Health Policy and Planning, 23: 294–307.

Green, Judith and Thorogood, Nicki (1998) Analysing Health Policy: A Sociological Approach. London: Longman.



Immergut, Ellen (1990), ‘Institutions, Veto Points, and Policy Results: A Comparative Analysis of Health Care’, Journal of Public Policy, 10(4): 391–416.

Jasso-Aguilar, Rebeca, Waitzkin, Howard, and Landwehr, Angela (2004) ‘Multinational Corporations and Health Care in the United States and Latin America: Strategies, Actions and Effects’, Journal of Health and Social Behavior, 45: 136–157.

Lee, Kelley, Buse, Kent, and Fustukian, Suzanne (2002) ‘An Introduction to Global Health Policy’. In Kelley Lee, Kent Buse and Suzanne Fustukian (eds.), Health Policy in a Globalising World (pp. 3–17). Cambridge: Cambridge University Press.

Lian, Olaug S. (2008) ‘Global Challenges, Global Solutions? A Cross-National Comparison of Primary Health Care in Britain, Norway and the Czech Republic’, Health Sociology Review, 17(1): 27–40.

Mechanic, David (1975) ‘The Comparative Study of Health Care Delivery Systems’, Annual Review of Sociology, 1: 43–65.

Mechanic, David and Rochefort, David A. (1996) ‘Comparative Medical Systems’, Annual Review of Sociology, 22: 239–270.

Mills, Lisa (2006) ‘Maternal Health Policy and the Politics of Scale in Mexico’, Social Politics, 13(4): 487–521.

Øyen, Elsa (1990) ‘The Imperfection of Comparisons’, in Elsa Øyen (ed.) Comparative Methodology: Theory and Practice in International Social Research (pp. 1–18). London: Sage and International Sociological Association.

Prior, Lindsay (2003) Using Documents in Social Research. London: Sage Publications.Ragin, Charles C. (1987) The Comparative Method. Moving Beyond Qualitative and Quantitative

Strategies. Berkeley: University of California Press.Ranson, Kent M., Beaglehole, Robert; Correa, Carlos M., Mirza, Zafar, Buse, Kent, and Drager, Nick

(2002) ‘The Public Health Implications of Multilateral Trade Agreements’. In Kelley Lee, Kent Buse, and Suzanne Fustukian (eds.), Health Policy in a Globalising World (pp. 18–40). Cambridge: Cambridge University Press.

Rodwin, Victor G. (2008) ‘Health and Disease in Global Cities: A Neglected Dimension of National Health Policy’. In S. Harris Ali, and Roger Keil (eds), Networked Disease. Emerging Infections in the Global City (pp. 27–48). Malden, MA: Wiley-Blackwell.

Saltman, Richard B. (1997) ‘Convergence versus Social Embeddedness. Debating the Future Directions of Health Care Systems’, European Journal of Public Health, 7(4): 449–453.

Shiffman, Jeremy (2007) ‘Generating Political Priority for Maternal Mortality Reduction in 5 Developing Countries’, American Journal of Public Health, 97(5): 796–803.

Timmermans, Stefan and Berg, Marc (2003) The Gold Standard: The Challenge of Evidence-Based Medicine and Standardization in Health Care. Philadelphia: Temple University Press.

Waitzkin, Howard, Jasso-Aguilar, Rebeca, Landwehr, Angela and Mountain, Carolyn (2005) ‘Global Trade, Public Health, and Health Services: Stakeholders’ Construction of Key Issues’, Social Science and Medicine, 61: 893–906.

Walt, Gill (1994) Health Policy. An Introduction to Process and Power. London and New Jersey: Zed Books.

Walt, Gill, Shiffman, Jeremy, Schneider, Helen, Murray, Susan F., Brugha Ruairi, and Gilson, Lucy (2008). ‘“Doing” Health Policy Analysis: Methodological and Conceptual Reflections and Challenges’, Health Policy and Planning, 23: 308–317.

Wrede, Sirpa, Benoit, Cecilia, Bourgeault, Ivy Lynn, van Teijlingen, Edwin R., Sandall Jane, and De Vries Raymond G. (2006) ‘Decentred Comparative Research: Context Sensitive Analysis of Maternal Health Care’, Social Science & Medicine, 63: 2986–2997.

van Herk, R., Klazinga, N.S., Schepers, R.M.J., and Casparie, A.F. (2001) ‘Medical Audit: Threat of Opportunity for the Medical Profession: A Comparative Study of Medical Audit among Medical Specialists in General Hospitals in the Netherlands and England, 1970–1999’, Social Science and Medicine, 53: 1721–1732.


chandana.k

Text Box

AQ: Please confirm 'van Herk' placement is ok?