An overview of ethnography in healthcare and medical education research

Research in healthcare settings and medical education has relied heavily on quantitative methods. However, there are research questions within these academic domains that may be more adequately addressed by qualitative inquiry. While there are many qualitative approaches, ethnography is one method that allows the researcher to take advantage of relative immersion in order to obtain thick description. The purpose of this article is to introduce ethnography, to describe how ethnographic methods may be utilized, to provide an overview of ethnography's use in healthcare and medical education, and to summarize some key limitations with the method.


INTRODUCTION
Researchers from the social science disciplines are able to take advantage of a wide array of research methodologies. Such methodologies range from the traditional quantitative appro aches within the positivist tradition to the qualitative appro aches premised upon the ideologies of constructivism. A third alternative, mixed methods designs, have also gained popular ity in recent years. Inherent to each of these approaches are different, and sometimes opposing, philosophical and episte mological views. These views dictate the nature of the research design. On one hand, positivists focus on studying that which can be directly observed and confirmed by the senses. Such research seeks to test theorybased, testable hypotheses while remaining objective and value neutral. On the other hand, con structivists believe that knowledge is socially constructed and situated within a particular context. Since each context is uni que and will have different perspectives, the world has many different meanings -none of which may be more valid than another. Hence, research designs are primarily aimed at de scribing the context or group of interest.
Creswell [1] cites these three types of designs: qualitative, quantitative, and mixed methods. A mixed method is a com bination of qualitative and quantitative. These two methods are at opposite ends of the continuum and the continuum is the level of measurement: nominal, ordinal, interval, and ra tio. The type of research method used is referring to the meth od used to collect the data. Researchers using quantitative me thods are testing theories through examining statistical rela tionships between variables. The data is measured numerically with an ordinal, interval, or ratio scale. Qualitative methods use a nominal scale. That is, they may use numbers to label categories but it is only for the purpose of sorting information. The number is simply a label. Qualitative methods aim to ex plore a culture or group of individuals to understand more about the social or human problem this group experiences. Qualitative methods are often open ended or participatory in nature. Qualitative research, whether standing alone or in a mixed method, adds rich information to any investigation oth erwise not discoverable. With quantitative methods, investiga tors rely on literature and past surveys to target the proper in formation. Qualitative research allows for variable discovery. That is, the results may glean something never before addres sed in similar research. The use of qualitative methods is es sential to get closer to exhaustive information on any given topic or population. Oftentimes, the qualitative research leads the researcher to further quantitative investigation. These are not competing but rather complimentary methods.
Qualitative research is often overlooked as an option when considering the methodological approach to a research ques tion. This is especially true in academic domains such as med icine where evidencebased practice has emerged as a popular treatment philosophy based largely on the quantitative resear ch tradition. However, qualitative research can provide rich information otherwise not discovered through quantitative approaches. Ethnography is one qualitative approach that in volves relative submersion into the setting to be studied, and is an appropriate methodology for a wide variety of research topics within healthcare and medical education. While, to some extent, ethnography has been applied in healthcare settings and in the medical education environment, we feel that there is a general lack of research employing this methodology. This opinion has also been expressed by Leung [2]. At times, defi cits in particular research methodologies within various aca demic domains may be attributed to a general lack of knowl edge regarding the methodology itself or ways in which the method may be applied. Therefore, the purpose of this intro ductory paper is to explain ethnographic methodology, dis cuss how the method may be used, provide a discussion of ethnography's use in healthcare and medical education, and briefly summarize some key limitations with the method. Since ethnography is a method not easily summarized in a single paper, we are writing a series of articles to follow which will address specific aspects related to conducting ethnographic research.
We begin by describing ethnography, synthesizing the works of Leung [2], Savage [3], LeCompte and Schensul [4], Pope [5], and Atkinson and Pugsley [6]. Leung [2], for example, dis cusses ethnography as a social research method occurring in natural settings characterized by learning the culture of the group under study and experiencing their way of life before attempting to derive explanations of their attitudes or behav ior. The culturally based approach can be related to ethnicity, nationality, gender, regional origin, occupation, generation, or in healthcare the focus might be a particular pathology such as cancer, HIV, heart disease, or diabetes. Ethnographies are normally conducted in a single setting, and data collection is largely dependent upon participant observation and interviews. Savage [3] notes that ethnography may also require historical research prior to beginning actual field work. In terms of time considerations, ethnography is a research method character ized by longterm fieldwork since thick description of the par ticipants and setting may only be acquired from sufficient ex posure to them. Savage [3] further notes that ethnography is not used for developing generalized conclusions but rather studying a specific group of people regarding a specific topic and drawing conclusions only about what was studied.
According to LeCompte and Schensul [4], there are seven defining characteristics of ethnography. These include: 1) be ing carried out in a natural setting, not in a laboratory; 2) in volving intimate, facetoface interaction with participants; 3) presenting an accurate reflection of participant perspectives and behaviors; 4) utilizing inductive, interactive, and recursive data collection to build local cultural theories; 5) using both qualitative and quantitative data; 6) framing all human behav ior within a sociopolitical and historical context; and 7) using the concept of culture as a lens through which to interpret study results.
Finally, in order to fully understand ethnographic method ology, it is necessary to briefly describe the fundamental ideas and guiding principles of ethnography which have been de rived across the many academic disciplines that make use of its application. Atkinson and Pugsley [6] nicely summarize these ideas which have been succinctly stated in Table 1.

USES OF ETHNOGRAPHY
Ethnography is a useful qualitative approach to address a particular type of research question. LeCompte and Schensul [4] suggest that ethnography should be used to: · Define a problem when the problem is not yet clear. · Define a problem when it is complex and embedded in mul tiple systems or sectors. · Indentify participants when the participants, population sectors, stakeholders, or the boundaries of the study popu lation are not yet known or identified. · Clarify the range of settings where a problem or situation currently occurs when not all of the possible settings are fully identified, known, or understood. · Explore the factors associated with a problem in order to indentify, understand, and address them either though re search or intervention studies, when they are not known. · Document a process. · Identify and describe unexpected or unanticipated outcomes. · Design measures that match the characteristics of the tar get population, clients, or community participants when existing measures are not a good fit or need to be adapted. · Answer questions that cannot be addressed with other me thods or approaches. · Ease the access of clients to the research process and its pro ducts.
Consider the following example by Perry et al. [7] on the use of ethnography to address a research question.
Summary: Many individuals must deal with loved ones who are too disabled to care for themselves. As pressure con tinues to keep healthcare costs down, many patients stay fe wer days in the hospital after an injury or illness. Perry et al [7]. described the challenges associated with releasing pati ents from the hospital before they are fully recovered. Nurses often become the supervisor of the care rather than the pri mary caregiver. This trend brings to light potential conflicts between supervisor and caregiver. Perry  In this case, ethnography was a useful method in address ing the research question. In order to fully understand the fa mily's perspective regarding premature hospital release as well as the responsibility of such caregiving, the researchers needed to understand the complex nature of these phenomena. Eth nography was chosen since it allows for thick description of these underlying issues.
In general, a deep understanding to a research question is nicely achieved through the use of ethnography. Rice and Ezzy [8] contend that there are several benefits to conducting an ethnography: diverse cultures are better understood through a deep understanding of why people behave in certain ways and have certain beliefs, it is a strategy for the development of grounded theory, and it best addresses the needs of humans.

ETHNOGRAPHIC STUDIES IN HEALTHCARE AND MEDICAL EDUCATION
While cultural understanding is the most cited use of eth nography, there are many specific settings where the method is quite valuable. One of those areas is healthcare. The vast numbers of variables in the clinical setting lend themselves well to be analyzed in a more open ended approach rather than answering questions from a survey or pulling archived data from the hospital and clinic databases [8]. The patient/ nurse relationship is the key to the success of healthcare. Bet ter understanding those dynamics allows decision makers to proceed with more reliable and pertinent information. Get ting to the root issues about patient care rather than tracking behavior leads to real solutions rather than a trial and error process.
Campbell [9] also suggests the use of qualitative research in health care. A large field in itself, health policy would benefit from the complimentary methods of qualitative analysis. Fur ther suggested was metaethnography as a qualitative compli ment to quantitative metaanalysis [9]. Through a study in volving diabetic care, Campbell [9]  pliance was studied with the benefit of seven different studies. Campbell [9] further suggested methodology within the stud ies included needed to be better documented a suggestion for further development of metaethnography.
Garro [10] nicely summarizes the uses of ethnography in health care decisions. It begins with the most basic of decisions as to whether or not to seek health care. Garro [10] goes on to cite several examples of use in medicine. Women deciding to go to the hospital to have their baby generally go too early, hence requiring them to go home and come back again. The Scrimshaw and Souza study cited by Garro [10] goes on to describe a dynamic where healthcare providers assumed pa tients understood the instructions the same way as the health care provider. The result and further action item was to rede sign instructions in two different languages to accommodate the different interpretive tendencies of the patients. Cultural factors continue to be a major factor in the delivery of health care. Patients make decisions through this lens creating a dy namic in the healthcare system with a wide variance of options for the various cultures served.
Garro [10] finally concludes her review with an example leading to an interesting conclusion. Two Mexican towns were studied to determine decision making patterns as they relate to healthcare. The towns were thought to have similar beliefs and the interviewing techniques for these two towns were dif ferent. Again here, we see ethnography can be tailored to the population being studied. The flexibility is very useful. The two techniques yielded similar results in terms of the medical knowledge of the two communities. The difference in the two communities was access. The town having greater access to healthcare sought medical attention at twice the rate of the town lacking access. The Garro study indicated access to health care was more indicative than illness beliefs when determin ing how and when a patient sought care [10].
Van der Geest and Finkler [11] suggest the use of ethnogra phy in the hospital setting. Hospitals are often cultures within themselves. And, while some can be very similar, the commu nity of the hospital is often unique. Because hospitals reflect dominant culture and belief systems, the care in each hospital can be different based on the cultural influences. This is not clear to the naked eye. From the outside, hospitals look and operate similarly. The patient care and decision making pro cesses can vary widely. Rice and Ezzy [8] suggest that, through ethnography, behaviors are understood and used to treat the patient through means that fit the needs of the patient. The benefits brought by the ethnography are understanding of the social and cultural backgrounds of the patients and how health behaviors differ across groups. Savage [3] cites useful ethnog raphies in health care looking at various issues from cultural differences among clinic attendees to the clinical reasoning differences among physician specialists.
Ethnography has been used in medical education for more than 50 years. Two landmark studies conducted in the United States show earlier uses of ethnography. The first, entitled The Student Physician, was a collection of research from several medical schools. Within this collection, Fox's ethnographic project on medical innovations in clinical settings or the en thographic account of uncertainty in medical knowledge has influenced much subsequent research [6,12].
Boys in White, the second ethnographic account, focused on medical student culture. In Atkinson and Pugsley's [6] ac count, this study drew "an explicit parallel between student cultures in universities and shopfloor cultures in workplace settings. Workers and students alike established shared per spectives on their shared problems and collective responses to shared demands" (p. 233). Medical students were found to more effectively manage these demands by using selective neg ligence (only learning vital information). More recent uses of ethnography in medical education have been noted and will be discussed in our series of subsequent articles as exemplars of ethnographic research.

LIMITATIONS OF ETHNOGRAPHY
Sample size is a limitation of the ethnography method. The time required being involved in participant observation and conducting long interviews greatly limits the sample size. Un like a scan sheet used for a common survey, ethnography is laborious and detailed in the collection of data.
It is difficult to generalize with the ethnography method. When researching a certain culture, the results cannot neces sarily be generalized to other populations. Because the results are based on the cultural responses, the outcome of the study cannot be applied beyond where the study was conducted. For instance, if a best practice ethnography was conducted for the Emergency Room in a hospital, the same best practices may not be applied to the hospital on the other side of town. The best practices for hospital number one were developed based on the population using that hospital and the adminis tration running that hospital.
Subjectivity is certainly a limitation of ethnography. The in terpretation of the cultural experience will vary among resear chers. There is not a list of answers from which to choose but rather the use of notes made by the investigator and later in terpreted and categorized by the investigator. The entire proj ect is subject to the processes and interpretations developed by the researcher and the research team. Not so with quantita tive research. With quantitative, there is a measurable response from which to draw conclusions. It is objective.
Funders are often reluctant to fund such projects. Some feel http://jeehp.org this type of research can lack generalisability [3]. Another is sue for funders is the great expense. It is much less expensive to conduct research with a survey than to immerse an investi gator into a culture for a given period of time to extract exten sive information. Finally, the biggest challenge an ethnographer may encoun ter is the acceptance of the culture. Should the researcher choo se ethnography, they must be accepted temporarily within the culture in order to gather accurate information. Hence the need to conduct thorough reviews prior to beginning the eth nography. Additionally, the development of key informants is essential.

CONCLUSION
Qualitative research, while time consuming, is an excellent tool to investigate differences among cultures, genders, profes sions, and geographic regions, to name a few. It provides a rich collection of information for the investigator. Oftentimes, it is the source of determining variables for further research and can therefore serve as the launching pad for larger original studies. While we have seen some use of qualitative research within healthcare and medical education, it is underrepresent ed compared to its quantitative counterpart. Of these qualita tive approaches, ethnography is a method quite amenable to medicine, and the application of ethnography to healthcare is widely supported. Health behaviors and differences in health care delivery are not necessarily detected in a quantitative study. The use of ethnography allows the decision maker to have a better understanding of the patient and the healthcare deliv ery team.