 |
 |
|
|
|
Research in psychiatry overwhelmingly relies on quantitative methodology (1). Qualitative methodology is employed less frequently, with some regarding it as fundamentally esoteric, if not inferior, to mainstream psychiatry. In this paper, we attempt to dispel this belief by describing qualitative research, its underlying philosophical basis, and its main methods of inquiry. Where appropriate, we support our position with examples of where it has been or may be fruitfully employed. We believe qualitative research can offer interesting insights into the prevention, diagnosis, phenomenology, treatment, management, and understanding of psychiatric disorder. It may also assist resolution of current policy imperatives, such as calls for patient involvement and more thorough evaluations of services (2,3). Overview of Qualitative ResearchQualitative research is a broad umbrella term describing several specific research methods and paradigms that rely on the collection, analysis, and interpretation of nonmathematical data. There are 3 main methods in qualitative research: in-depth interview, focus groups, and participant observation (4). These have a long tradition of use in the social sciences, being prime methods of investigation in sociology and anthropology (5). Documentary analysis is a less commonly employed qualitative method. To limit the scope of this article, we have eschewed discussion of this method, which is described in detail elsewhere (6). That said, interested readers may like to consult several recent studies related to mental health that have employed documentary analysis (1,7,8) Aims of Qualitative ResearchMalinowski stated that qualitative research aims to understand a phenomenon from a “native” point of view that above all emphasizes subjective meaning and experience (9). Others have termed this the documentation of “world views” (10). Thus the respondent’s perspective usually has priority over the researcher’s preconceived framework. To enrich understanding of the issue, the researcher aims to construct a “thick description,” that is, a comprehensive account of the phenomena under investigation (11,12). Unlike quantitative research, which sets a premium on reducing data to easily comprehensible units (for example, averages), qualitative research consciously attempts to document the complexity and multiplicity of experience. This can then be used to construct theory (13). To give an example, a recent qualitative study in psychiatry explored the experience and meaning of stigma, a deeply personal experience, by comparing data from focus groups of schizophrenia patients, their relatives, and health professionals. The “thick description” of stigma allowed the authors to theorize that stigma is dimensional and that appropriate interventions should be targeted to these various dimensions (14). The Philosophical Basis of Qualitative ResearchFor qualitative researchers, the field of inquiry is the social world, usually conceptualized as governed by principles different from the physical world. For example, the social world is not generally seen as governed by universal truths reflecting a single reality that can gradually be uncovered through the advancement of science. In fact, qualitative research has traditionally relied on a strong belief in context-dependent, multiple, and complex realities. This position, which emphasizes the particular over the universal, contrasts with the search for parsimony that dominates the physical sciences (15). Still, one strand of qualitative research suggests that particular and universal perspectives can often complement the investigation of a research question. For example, it has been suggested that isolated case studies can build up “local” theory or knowledge. However, the comparison and integration of many similar qualitative studies may be used to generate more universal theory (13,16). Another orientation of qualitative research that differs somewhat from that of quantitative research is a belief in the importance of documenting and analyzing “folk” beliefs, especially when they may influence health outcomes, for example, regarding medication compliance or service use. An aphorism employed by many qualitative researchers is that the respondent is the expert, whereas the researcher is the layman—a reversal of traditional biomedical epistemological assumptions. For example, a recent study exploring factors that determine continuity of psychiatric care gave equal weight to documenting professional and “folk” knowledge (17). This was done through a combination of participant observation and in-depth interviews with patients and service providers. The results suggested that the approach of mental health professionals, such as a willingness to respond flexibly at times of crisis, is particularly important in determining patients’ perceptions regarding the effectiveness of continuity of care. The Practice of Qualitative ResearchThe design and process of qualitative research and the methods used to analyze consequent data generally differ from the analogous procedures in quantitative research (4). There are guiding rules to this, including appropriate checks on validity and reliability. However, there are no traditional benchmarks in qualitative research equivalent to power calculations, significance tests, or randomized controlled trials. Most quantitative research occurs in a positivist paradigm relying on the hypothetico-deductive method: data are collected to test a theoretically driven, predefined hypothesis, that may support or refute the said hypothesis (18). Qualitative research can also proceed according to this Popperian model, where researchers enter the field with a prestructured design aiming to test a hypothesis. However, a lot of qualitative research also moves from data to theory, rather than vice versa. Such inductive studies usually begin with a loose research question rather than with a strict hypothesis. Data are then gathered and analyzed through inductive rather than deductive principles. Though reliance on induction is often considered one of the defining factors of qualitative research, it may be better to picture individual qualitative research studies as occurring somewhere on a deductive–inductive spectrum (19). Deductive qualitative research usually involves more front-end work, whereas an inductive study will involve more back-end work (16). A major component of this back-end work will usually be the development of “grounded theory.” In grounded theory, the respondents’ predominant themes and concepts are documented and analyzed through inductive content analysis. The astute qualitative researcher will then use this analysis to construct a local theory, grounded in the data, to explain local occurrences and phenomena (13). A more universal theory can then be developed as the researcher dialectically fuses his or her expert knowledge of preexisting literature with the local theory developed from his or her specific study. To give an example, qualitative methodology has been used to investigate the applicability of Western psychiatric nosology in ethnographic fieldwork in China (20,21). Through this loose research question, perhaps best visualized as lying midway along the inductive–deductive spectrum, Western researchers have combined their own expert knowledge with themes and concepts gathered from Chinese respondents. These data have been used to develop a new theory—that Western systems of psychiatric nosology are not universally applicable but are, in fact, loaded with tacit cultural biases that may be invalid in other cultures. Through inductive analysis, it was found that other cultures have their own nosological systems that may not overlap with Western systems of categorization. This finding led to the development of the important concept of “category fallacy,” that is, the projection of one culture’s diagnostic categories onto another culture that may in fact perceive, experience, and express mental illness quite differently (22). The Temporal Procedure of Qualitative ResearchThe temporal procedure of quantitative research usually relies on a linear development of 3 distinct and ordered categorical divisions: 1) formulation of an immutable design, 2) data collection, and 3) data analysis. Although deductively orientated qualitative research can legitimately progress along these lines, inductive qualitative research is often more flexible and iterative, with less distinction between design, data collection, and data analysis. An inductively orientated qualitative study may therefore be loosely designed to answer a broad research question. However, a main aim in this inductive research is to get into the field as quickly as possible to gather generic data that can be analyzed simultaneously, the results being used to guide further, more precise data collection. The complete process can thus be envisioned as circular rather than linear, with feedback loops affecting the ongoing research development. This form of inductive, circular research may be most appropriate when studying an underresearched group or such complex realities as the experience of a new group of refugees in an urban context (23). SamplingIn qualitative research, purposive (or theoretical) sampling is more common than random sampling. Purposive sampling describes a process that deliberately recruits individuals or groups with the requisite demographic or clinical characteristics into the research, allowing the study to be grounded in a local context. It is generally not necessary to predefine an exact or desirable number of respondents before the research begins (24). Data initially collected from diverse individuals can be analyzed simultaneously. This analysis can then direct numbers and groups to be involved in subsequent data collection. Again, an iterative relation exists, this time between data analysis, further sampling, ongoing data collection, and theoretical development (25). Sampling and data collection may be terminated when the researcher judges that new respondents are not adding anything significant to the database—a situation known as “theoretical saturation” and only reached once the researcher has consciously tried to sample groups or individuals that might stretch the diversity of the data. This is known as extreme case sampling; it is a check on validity, as it ensures adequate representation and precludes premature conclusions (26). For example, a qualitative study evaluating patient satisfaction with a psychiatric outpatient service may find, after initial sampling and analysis, that one ethnocultural group’s responses differ radically from the responses of everyone else. A flexible design and sampling strategy will allow researchers to shift the focus to this subgroup for in-depth sampling and exploration. Methods of InquiryQualitative research has 3 main methods of inquiry: in-depth interviews, focus groups, and participant observation. These can be used in isolation or in conjunction. The use of different methods in the same research project and the subsequent comparison of results is known as triangulation and is considered a highly desirable check on validity. It strengthens findings by increasing the breadth of results allowing cross-comparison, so that overlap and discrepancy can be assessed (27). Common Themes in Interviews and Focus GroupsIn psychiatry, interviews and focus groups are probably the most common methods of qualitative research. They both involve the elucidation of subjective meaning, experience, beliefs, and attitudes, either through one-on-one interviews or small, facilitator-led, group discussion. They can last between 1 and 2 hours and are usually recorded and then transcribed for later analysis. Interviews and focus groups are generally directed by a series of written prompts that act as aides-memoir, reminding the researcher of key points to be covered. These are sometimes known as topic guides, and their level of development will depend on how far the researcher is taking an inductive or deductive orientation. The key points in the topic guide may be identified through literature review and brainstorming with colleagues and relevant lay people. A well-designed topic guide is essential to ensure a smooth research interaction and the collection of valid and reliable data. For example, the rank ordering of questions should be designed carefully to ease the respondent gently into the encounter. As rapport builds, initial nervousness should dissipate, leading to self-disclosure of more private and controversial issues. Leading and closed questions, which may be useful in clinical interviews, should be avoided (for example, the open and neutral question, “Why did you stop taking your medication?” is superior to the suggestive question, “Did you stop taking your medication because of unpleasant side effects?”) Topic guides may be liberally followed because some respondents may forestall topics and in effect rearrange the guide, or they may raise interesting issues not on the guide that should be followed up. Focus GroupsFocus groups differ from one-on-one interviews in that they are a collective act that can access group norms, collective opinions, and shared “knowledge,” rather than individual views (28). Focus groups provide an environment wherein respondents can share experiences with like-minded people, giving a feeling of safety in numbers. They have thus been used, for example, to access the opinions of family physicians regarding the detection and management of mental health problems in the primary care setting and perceived barriers to the optimal delivery of mental health care (29). They are also advantageous in that, unlike interviews, they do not resemble the hierarchical doctor–patient relationship that may inhibit some participants. Focus groups are thus ideal for obtaining data from groups of people (for example, prisoners) who may feel intimidated or uncomfortable in one-on-one situations (30). Focus groups can also usefully assist the identification of the active ingredients of multifaceted interventions. The success or failure of these interventions can be difficult to unpack quantitatively because they consist of various initiatives involving complex interactions. One study used this approach by convening 5 focus groups (2 for clinicians and 3 for patients) to help elucidate why psychiatric patients assigned to an integrated primary care clinic had greater improvements than those assigned to care as usual in a regular medical clinic. Clinicians and patients reported that the integrated clinic offered greater flexibility and smoother communication than did general medical clinics, which led to improved outcomes (31). InterviewsQualitative interviews differ from standard clinical interviews, as questions are generally open and structure loose; moreover, in inductively orientated studies, the interviewee is as much in control of the agenda as is the interviewer (32). The intimate and open interaction of an interview can usefully explore why people act in certain health-promoting or health-damaging ways—sensitive issues that may be less accessible in focus groups. Interviews may also be advantageous where data are gathered predominantly from key informants such as health managers, physicians, or other senior health staff. They may prefer the confidential atmosphere of the interview, or it may be impossible to practically organize focus groups with busy individuals. This approach was used with some success in a recent study of key informants’ opinions about the impact of policy changes on the health of recent immigrants and refugees living in the inner city (33). A further example of the intelligent use of qualitative interviews in psychiatry is a study exploring aspects of the doctor– patient relationship through interviews with 30 family practitioners and 30 people with depression (34). The study’s aim was to explore the effect of perceptions of entitlement to time in consultations for depression. This perception is important because it may determine the patient’s level of symptom disclosure, increasing the risk that the doctor will overlook the mental illness or misdiagnose it as a somatic complaint. The study in fact found that patients self-censored information to avoid taking up too much time, which they perceived as being overly precious to the physician, leading to a suboptimal interaction. Physicians reported that their time was indeed precious but that longer appointments could easily be arranged for people who needed them (for example, people with psychological complaints). Patients were generally unaware of this possibility. This research thus identified a simple intervention: raising awareness that longer appointments are possible can assist those seeking help for psychological reasons. Participant ObservationParticipant observation involves the systematic description and analysis of behaviour and talk in such real-world settings as a clinic, a day centre, or a hospital ward. It is theoretically driven, as relevant times and places are selected to explore a research question. Researchers may record everyday occurrences, speech, dress, acute events, interpersonal interaction, and unwritten rules of behaviour. These are usually recorded as a collection of field notes that provide a basis for later analysis (9,11). Participant observation often occurs as part of an in-depth case study. It progresses along the same lines as the ethnographic reports of anthropologists and has been used occasionally in psychiatry. The in-depth case study relies on the researcher’s becoming intimately acquainted with a small and relatively homogeneous setting. Scope is deliberately limited and the intensity of assessment high, allowing maximum undivided attention to the setting (35). As the researcher becomes a familiar figure, the likelihood of valid and reliable responses increases. This familiarity and the vast amount of data generated allow the development of a thick description (36). Goffman used participant observation in a case study of a mental hospital to discover and document the intricacies of life therein from an insider’s point of view (37). This study resulted in his formulation of the valuable concept of the “total institution,” a place such as a mental hospital, army barracks, or boarding school, wherein a high degree of demoralizing rules and regulations cover individuals’ daily lives. It led to a degree of reflection about the wisdom of total institutions and assisted those arguing for deinstitutionalization in psychiatry (38). Most case studies will have a limited applicability beyond the specific setting; extensions drawn by the researcher should generally be modest and grounded in the setting’s particularities (39). However, detailed description of the particularities allows readers to judge for themselves the generalizability of the results and conclusions. Analysis of ResultsQualitative data analysis is usually ongoing during data collection, with the one usually informing the development of the other. However, at the end of data collection and following a period of withdrawal from the field for reflection, many researchers will engage in what could be termed a grand analysis. This grand analysis attempts to synthesize all the data into a thick description. If the research is at the deductive end of the spectrum, this thick description can be used to test the specific hypothesis. At the more inductive end of the spectrum, it can be used to generate, first, local grounded theory that may be situation-specific and then, more generic theory. Various methods are commonly used to assist these processes, most of which are variants on the same theme, content analysis. Content analysis has been described as the “systematic examination of text (field notes) by identifying and grouping themes and coding, classifying, and developing categories” (4). Content analysis involves systematically distilling the massive amounts of raw data into a comprehensible description without losing the complexity inherent in the original responses. This may best be conceptualized as a process of “mapping,” whereby themes are identified, appropriately weighted, and then related (16). This is a lengthy and wordy process, which perhaps explains why many qualitative studies are published as books rather than as journal papers (for example, 20,37). A commonly employed form of content analysis is known as constant comparative analysis (13). It relies on theoretically informed constant comparison between subgroups, for example, between doctors and patients, from whom differential thematic responses are extracted; or between coding frames within the data, for example, between positive or negative opinions about a subject (40). As a result of this analysis and comparison, the researcher should be able to document a theory grounded in the respondents and the setting. Several computer software packages have been designed to aid the analysis of qualitative data (for example, ethnograph [www.qualisresearch.com/], atlas-ti [www.atlasti.de/] and nud*ist/Nvivo [www.qsrinternational.com/]). These are useful for the systematic indexing, coding, storing, and management of the large amounts of data generated by a qualitative study. However, they do not obviate the need for close manual analysis, because they cannot formulate the thick description or theoretical framework in themselves; they are simply tools to assist this process (41). Maintaining RigourA traditional critique of qualitative research is that it is subjective, anecdotal, and highly prone to investigator bias. Nonhypothesis-driven qualitative research may be particularly at risk of these accusations, as the researcher may be tempted to selectively interpret data to fit into conscious or unconscious preconceptions (9,11). This critique is anchored in the awareness that some well-known, mostly anthropological, qualitative studies—for example, Margaret Mead’s portrayal of the Samoans (42)—have been considerably criticized for painting a distorted picture of other people’s reality. (43). Most qualitative researchers accept that the risk of investigator bias is high; however, several checks and balances have been developed in response, which can be employed to reduce this risk. We have already mentioned some of these, such as triangulation, extreme case sampling, and use of a theoretically informed topic guide. Two other commonly employed methods of adding rigour to qualitative studies are respondent validation and multiple coding. Respondent validation involves taking key themes and grounded theory back to respondents near the end of the research to discern levels of congruence between researchers’ and respondents’ theories (44). This reduces the likelihood of misrepresenting respondents’ views. Multiple coding involves 2 or more researchers analyzing the same data set and then comparing and discussing findings. Again, this diminishes investigator bias and can be seen as a qualitative form of interrater reliability (27). Combining these developing epistemological canons should add rigour to studies employing qualitative research. Presentation of ResultsFor the presentation of results, Miles and Huberman provide a widely used 3-stage model akin to that used in literary criticism (16). This model involves reducing data through thematic extraction, displaying selective examples of data to support arguments, and drawing conclusions embedded in the data. In sum, the aim should be to provide a convincing set of results and conclusions that adequately summarize advances of the research. It has been argued that an explicit and reproducible description of the method of analysis (sometimes known as an “audit trail”) should accompany the results (45). Readers can then judge for themselves whether the design, analysis, and interpretation are appropriate and rigorous with regard to the research question under discussion. ConclusionThis paper is a necessarily brief overview of qualitative research in psychiatry. As such, we have not treated certain general qualitative research issues with the attention they deserve. Though we have outlined some qualitative research procedures, we have avoided detailed discussion of some everyday practical aspects of conducting qualitative studies, such as assembling participants or groups—these are described in general qualitative textbooks (5,11,15). Additionally, we have not discussed the unique ethical issues raised by qualitative research carried out among people who may have mental health problems or within identifiable localities. Fortunately, a large developing literature is devoted to qualitative research ethics, to which we can direct interested readers (46,47). Despite these limitations, we hope that this article will increase the ability to conduct, assess, and understand qualitative research in psychiatry. At root, research methodology should be judged on its suitability to answer the question under discussion. In many cases, quantitative research is well suited to answer typical questions posed in psychiatry. However qualitative research may be better suited to answer some of the innovative questions arising out of contemporary psychiatry, especially where it is intelligently combined with parallel quantitative research (48–50). This is especially so in a changing policy climate, where demands are increasing for evidence-based medicine, patient involvement, and evaluation of services in the desirable shift toward health promotion and a “new public health” (3,51,52). Funding and SupportWe received no specific funding or support for writing and preparing this paper. Rob Whitley is currently supported by a grant from the Leverhulme Trust, UK. References1. Crawford MJ, Ghosh P, Keen R. Use of qualitative research methods in general medicine and psychiatry: publication trends in medical journals 1990–2000. Int J Soc Psychiatry 2003;49:308–11. 2. Jones R. Why do qualitative research? BMJ 1995;311:2. 3. Ashton J, Seymour H. The new public health. Milton Keynes: Open University Press; 1988. 4. Pope C, Mays N. Reaching the parts other methods cannot reach: an introduction to qualitative methods in health and health services research. BMJ 1995;311:42–5. 5. Pope C, Mays N. Qualitative research in health care. London: BMJ Books; 2000. 6. Silverman D. Interpreting qualitative data. London: Sage; 2001. 7. Landeen J, Pawlick J, Rolfe S, Cottee I, Holmes M. Delineating the population served by a mobile crisis team. Can J Psychiatry 2004;49:45–50. 8. Whitley R, Prince M. Are inner-cities bad for your health? Comparisons of residents’ and third-parties’ perceptions of the urban neighbourhood of Gospel Oak, London. Sociol Health Illn 2005;27:44–67. 9. Malinowski B. A scientific theory of culture and other areas. Raleigh (NC): University of North Carolina Press; 1990. 10. Kuhn T. The structure of scientific revolutions. Chicago (IL): University of Chicago Press; 1962. 11. Spradley JP. Participant observation. London: Thomas Learning; 1980. 12. Geertz C. The interpretation of cultures. New York: Basic Books; 1973. 13. Glaser B, Strauss A. The discovery of grounded theory. Chicago (IL): Aldine; 1967. 14. Schulze B, Angermeyer MC. Subjective experience of stigma. A focus group study of schizophrenic patients, their relatives and mental health professionals. Soc Sci Med 2003;56:299–312. 15. Lincoln Y, Guba E. Naturalist inquiry. London: Sage; 1985. 16. Miles M, Huberman A. Qualitative data analysis. Thousand Oaks (CA): Sage; 1994. 17. Ware N, Tugenberg T, Dickey B, McHorney C. An ethnographic study of the meaning of continuity of care in mental health services. Psychiatr Serv 1999;50:395–400. 18. Popper K. The logic of scientific discovery. London: Hutchinson; 1959. 19. Weisner TS. The ecocultural project of human development: why ethnography and its findings matter. Ethos 1997;25:177–90. 20. Kleinman A. Social origins of distress and disease: depression, neurasthenia and pain in modern China. New Haven (CT): Yale University Press; 1986. 21. Kleinman A, Kleinman J. Somatisation: the interconnections in Chinese society among culture, depressive experience, and the meaning of pain. In: Kleinman A, Good B, editors. Culture and depression. Berkley (CA): University of California Press; 1985. 22. Kleinman A. Rethinking psychiatry. New York: The Free Press; 1988. 23. Rousseau C, Mekki-Berrada A, Moreau S. Trauma and extended separation from family among Latin American and African refugees in Montreal. Psychiatry 2001;64:40–59. 24. Edgerton RB. Methods in psychological anthropology. In: Naroll R, Cohen R, editors. New York: Columbia University Press; 1970. p 338–52. 25. Maxwell J. Qualitative research design: an interactive approach. London: Sage; 1996. 26. Crabtree BF, Miller WL. Doing qualitative research. London: Sage; 1992. 27. Mays N, Pope C. Rigour and qualitative research. BMJ 1995;311:109–12. 28. Kitzinger J. Introducing focus groups. BMJ 1995;311:299–302. 29. Craven MA, Cohen M, Campbell D, Williams J, Kates N. Mental health practices of Ontario family physicians. Can J Psychiatry 1997:42:943–9. 30. Nurse J, Woodcock P, Ormsby J. Influence of environmental factors on mental health in prisons: focus group study. BMJ 2003;327:480–3. 31. Miller CL, Druss BG, Rohrbaugh RM. Using qualitative methods to distill the active ingredients of a multifaceted intervention. Psychiatr Serv 2003;54:568–71. 32. Britten N. Qualitative interviews in medical research. BMJ 1995;311:251–3. 33. Steele LS, Lemieux-Charles L, Clark JP, Glazier RH. The impact of policy changes on the health of recent immigrants and refugees in the inner city. A qualitative study of service providers’ perspectives. Can J Public Health 2002;93:118–22. 34. Pollock K, Grime J. Patient’s perceptions of entitlement to time in general practice consultations for depression: qualitative study. BMJ 2002;325:687–90. 35. Hammersley M. What’s wrong with ethnography? London: Routledge; 1992. 36. Denzin NK, Lincoln YS. Handbook of qualitative research. London: Sage; 1994. 37. Goffman E. Asylums. New York: Doubleday; 1961. 38. Leifer R. In the name of psychiatry: the social functions of psychiatry. New York: Science House; 1969. 39. Mays N, Pope C. Observational methods in health care settings. BMJ 1995;311:182–4. 40. Green J. Commentary: grounded theory and the constant comparative method. BMJ 1998;316:1064–5. 41. Fisher M. Qualitative computing: using software for qualitative data analysis. Aldershot (UK): Ashgate; 1997. 42. Mead M. Coming of age in Samoa. Magnolia (MA): Peter Smith; 1928. 43. Freeman D. Margaret Mead and Samoa: the making and unmaking of an anthropological myth. Cambridge (MA): Harvard University Press; 1983. 44. Bloor M. On the analysis of observational data: a discussion of the worth and uses of inductive techniques and respondent validation. Sociology 1978;12:545–52. 45. Greenhalgh T, Taylor R. How to read a paper: papers that go beyond numbers (qualitative research). BMJ 1997;315:740–3. 46. Macauley AC, Delormier T, McComber AM. Participatory research with Native community of Kahnawake creates innovative code of research ethics. Can J Public Health 1998;89:105–8. 47. Usher K, Holmes C. Ethical aspects of phenomenological research with mentally ill people. Nursing Ethics 1997;4:49–56. 48. Baum F. Researching public health: behind the qualitative-quantitative methodological debate. Soc Sci Med 1995;40:459–68. 49. Weaver T, Renton A, Tyrer P, Ritchie J. Combining qualitative studies with randomised controlled trials is often useful. BMJ 1996;313:629. 50. Crawford MJ, Weaver T, Rutter D, Sensky T, Tyrer P. Evaluating new treatments in psychiatry: the potential value of combining qualitative and quantitative research methods. Int Rev Psychiatry 2002;14:6–11. 51. Black N. Why we need qualitative research? J Epidemiol Community Health 1994;48:425–6. 52. Ottawa Charter for Health Promotion. Health Promotion 1992;1:1–5. Manuscript received March 2004, revised, and accepted June 2004. Author(s)1. Research Fellow, Division of Social and Transcultural Psychiatry, Department of Psychiatry, McGill University, Montreal, Quebec. 2. Senior Lecturer in Psychiatry, Department of Psychological Medicine, Imperial College, London, UK. Address for correspondence: Dr R Whitley, Division of Social and Transcultural Psychiatry, Department of Psychiatry, McGill University, 1033 Pine Avenue West, Montreal, QC H3A 1A1 e-mail: robert.whitley@mail.mcgill.ca
1 | 2
|
||||||