Elsevier

Social Science & Medicine

Volume 63, Issue 8, October 2006, Pages 2091-2104
Social Science & Medicine

Focus group research and “the patient's view”

https://doi.org/10.1016/j.socscimed.2006.05.016Get rights and content

Abstract

A clear emphasis on the patient's view is discernible in the health services research literature of the past decades. Such a switch to patients’ perspectives has been greatly facilitated by a wider acceptance and use of qualitative methods. In particular, focus groups are often used to uncover the range and depth of experiences of health services users and chronically ill individuals. Although this method contributes to a better understanding of patients’ perspectives and practices, a number of analytical considerations have been overlooked. The aim of this paper is to consider how to conceptualise and analyse interactions in focus group research. We argue that focus groups are social spaces in which participants co-construct the “patient's view” by sharing, contesting and acquiring knowledge. Focus groups conducted with home care patients in Quebec, Canada (on antibiotic intravenous therapy, parenteral nutrition, peritoneal dialysis and oxygen therapy) are used to illustrate three interactive processes at work in focus groups: (1) establishing oneself as experienced and knowledgeable; (2) establishing oneself as in search of information and advice; and (3) validating or challenging one another's knowledge claims. We develop an analytical template focused on the subtle dynamics underpinning the various and at times competing claims of patients. This template acknowledges the processes through which participants attribute authority to the claims of others, including the focus group moderator. We find that focus group research does not derive epistemological authority simply from the identity of its participants. Rather, an emerging consensus about what constitutes the patient's view is the result of context-dependent social interactions that need to be scrutinised.

Introduction

A clear emphasis on the patient's view is discernible in the health services research literature of the past decades (Lawton, 2003; Sullivan, 2003). Various factors have contributed to this emphasis, including the need to understand better whether health services, programs and policies are responsive to the needs and concerns of patients. Such a switch to the patient's perspective has been greatly facilitated by a wider acceptance and use of qualitative methods (interviews, focus groups, observation, ethnography, etc.) (Pierret, 2003). These methods are often seen as particularly well suited to analysing the range of subtle dimensions that structure the patient's experience of health and illness and that are not necessarily part of the expert or clinical worldview (Murphy, Dingwall, Greatbatch, Parker, & Watson, 1998).

While focus groups share some assumptions with other qualitative methods, what makes them unique is the social interaction that occurs between participants, and between the moderator and participants. Although this interactive feature has been stressed as pivotal (Kitzinger, 1994), there have been few attempts to derive analytical principles for examining the impact of interactions on the processes and outcomes of focus groups. This gap in the literature is problematic given that focus groups are often used as if they were straightforward and efficient mechanisms for eliciting patients’ perspectives.

We argue that focus groups should be conceptualised as social spaces in which participants co-construct the “patient's view” by sharing, acquiring and contesting knowledge. Focus groups conducted with home care patients (on antibiotic intravenous therapy, parenteral nutrition, peritoneal dialysis and oxygen therapy) illustrate three interactive processes at work in focus groups: (1) establishing oneself as experienced and knowledgeable; (2) establishing oneself as being in search of information and advice; and (3) validating or challenging one another's knowledge claims. These dynamics have received scant attention in the health research literature even though they are key to understanding what emerges as the “patient's view” within the context of focus groups. Our empirical observations together with insights from the literature are used to derive an analytical template that makes explicit the role of interactions in focus group research.

Qualitative methods are usually promoted on the basis of their ability to provide a deeper understanding of the rationales (why), processes (how) and contexts (when) shaping patients’ interactions with health care services (Murphy et al., 1998). Focus groups in particular have become increasingly popular in various research areas. Hydén and Bülow (2003) searched the Psycinfo database in December 2000 using the keyword “focus group.” They found more than 900 articles, one-third of which were published between 1998 and 2000. Repeating the same search in June 2005, we found a total of 2467 papers (1428 published between 2000 and 2005, or 58% of the total). A similar search in the Sociological Abstracts database yielded a total of 606 peer-reviewed papers, 350 of which were published between 2000 and 2005 (58%).

In health research, focus groups are used to elicit and analyse the range and depth of experiences of health services users and chronically ill individuals. Since groups are typically comprised of 6–10 participants and discussions last 1.5–2.5 h, focus group research is sometimes seen as an economical way to gather the views of several individuals at the same time. Perhaps because of this perceived advantage, focus group research has sometimes neglected the epistemological and sociological underpinnings of group discussions. These are the issues on which this paper focuses. We also delve into what focus groups actually are in practise, rather than what they should be (for a methodological discussion see Krueger, 1998; Morgan, 1996).

Morgan (1996, p. 130) defines the focus group as “a research technique that collects data through group interaction on a topic determined by the researcher.” From this definition arise three issues that need to be unpacked: (1) Do participants’ motivations and expectations necessarily converge with the researcher's purposes? (2) What types of interactions are sought versus what types of interactions actually occur? (3) What is the impact of the moderator on the discussion's dynamics and content?

Hydén and Bülow suggest that focus groups “have often been regarded as a means for obtaining access to people's experiences, attitudes and views, rather than as a communicative and interactive event” (2003, p. 308). By asking “who's talking”, these authors examine more carefully on whose behalf participants are speaking when they express opinions and interact:

Do the participants represent various groups outside the focus group, like professional or social groups; or do they just represent themselves as individuals; or do they act as members of the focus group. Further, do they talk and act as the same type of representatives throughout the entire focus group session, or is it possible that they shift positions. (2003, p. 306)

After reflexively exploring these questions with focus groups they organised, Hydén and Bülow conclude that participants may speak in various capacities—and may shift from one capacity to another—depending on the moderator's instructions and on the contributions of participants. Accessing the “patient's view” is not simply a matter of recruiting individuals who exhibit certain attributes. Rather, it requires paying attention to how participants position themselves and to whether or not they endorse the capacity that was attributed to them in the research design. Hollander (2004) argues that a sociological approach would help better grasp such dynamics as it addresses the socially situated nature of the context-dependent “data” generated through small group discussions. This approach emphasises the larger social structures within which the discussion takes place, as well as the various motivations and expectations of participants and the position(s) they represent as they actively contribute to the ongoing discussion.

Morgan (1996, p. 139) argues that the “ability to observe the extent and nature of interviewees’ agreement and disagreement is a unique strength of focus groups.” Nevertheless, these interactions also create ambiguity and highlight the issue of how to “control” the content of discussions. Focus group methodologists generally advocate seeking homogeneity in group composition (i.e., segmentation). Bringing together participants that share similar characteristics (e.g., age, gender, occupational status, ethnicity) is seen as a kind of social glue that adds fluidity and depth to discussions.

Nevertheless, it is not entirely clear whether the researcher's definition of homogeneity is compatible with that of the participants, and whether it actually facilitates social bonding and disclosure (Madriz, 1998). Hydén and Bülow (2003) indicate two interactive moments that influence how participants relate to one another. The first is when participants establish a “common communicative ground”, which is influenced by the way the moderator initially presents the tasks to the group and invites specific participants to break the ice. The second is when participants add their “contribution to the common ground”, which involves referring to what has been said earlier by another participant and gradually co-constructing a narrative together. A strong, moving story “that is left unanswered and gets almost no support might indicate that it was too personal and thus far too unique to function as a common ground for co-producing narratives in the group” (Hydén & Bülow, 2003, p. 318). Those who feel unable to make such a contribution may also refrain from adding something that would be perceived as trivial compared to the first story shared.

Hollander (2004) describes four types of social context that expand upon the two interactive moments highlighted above. First, there is the associational context—“the common characteristic that brings the participants together [and] influences the group conversation and dynamics” (Hollander, 2004, p. 614). Second, there is the status context—“the relative positions of the participants in local or societal status hierarchies, such as workplace, authority, gender, race, age, sexual identity, or social class” (p. 616). The third type is the conversational context—“the flow of the discussion wherein sharing certain topics or personal experiences may seem incongruous as each contribution reflects one's level of thinking and/or maturity about the issue” (p. 619). Finally there is the relational context—“the level of intimacy already established among the participants” (p. 620) and the degree of rapport established by the moderator and participants.

Associational and status contexts are likely to contribute to establishing the common ground, while conversational and relational contexts may affect what types of contribution the participants feel are appropriate given the specific interactive events that unfold during the focus group.

The moderator is obviously pivotal in facilitating open and detailed exchange among participants. Whereas some seek a neutral moderator who “fits” with the group's culture (e.g., same gender and/or culture), others recognise the moderator's influence on the group and make this interaction a subject of analysis (Duggleby, 2005; Smithson, 2000). Crossley points out that a moderator may sometimes deliberately try to trigger a strong reaction:

Bringing on the health leaflets at this point allows me to ‘get around’ R's stubborn stance by making myself into a slightly ridiculous figure who is obviously ‘fighting a losing battle’ in trying to persuade R to change her behaviour [she has no intention to quit cigarette smoking]. In effect, I make myself a ‘stooge’ for the group, who collectively turn against me, but in doing so, provide valuable insights into how people may actually respond to health promotion. (2002, p. 1479)

Another aspect that will influence emerging narratives is the extent to which the moderator succeeds in intervening when one or more group members dominate the discussion. Smithson (2000, p. 112) argues that “the group is collectively ‘powerful’ in that it has access to shared knowledge of which the moderator is ignorant.” Sometimes a subset of participants will take control of the discussion, challenging each other and raising issues they perceive as relevant and preventing others from presenting their arguments fully. It is therefore necessary to understand the types of exchanges that participants are actively seeking, tacitly avoiding or explicitly rejecting during discussions.

Less obvious are the implicit ways in which the moderator structures the conversational expectations and climate and validates the legitimacy of the contributions of participants (Madriz, 1998). For instance, by defining the role of the moderator and the relevant themes to be explored, the moderator indicates the kinds of responses expected (Hydén & Bülow, 2003). Acknowledging that focus groups unfold according to their own context-dependent social characteristics helps reframe the moderator as a participant who also contributes to the construction of the narratives.

The knowledge construction processes that lie at the heart of focus group interactions represent a powerful analytical tool for rethinking exactly what kind of “patient's view” is generated in focus groups. This is because patients’ views are a product of their social histories—grounded in the context of everyday situations and reflecting, among other things, experiential knowledge about illness (Lawton, Peel, Parry, Araoza, & Douglas, 2005). Experiential knowledge refers to:

The ultimate source of patient-specific knowledge—the often implicit, lived experiences of individual patients with their bodies and their illnesses as well as with care and cure. Experiential knowledge arises when these experiences are converted, consciously or unconsciously, into a personal insight that enables a patient to cope with individual illness and disability. (Caron-Flinterman, Broerse, & Bunders, 2005, p. 2576)

Indeed, scholarly work in the sociology of health and illness emphasises the “lay” perspective of illness, including its experience, explanation and management. Chronic illness in particular has been a fertile ground for exploring the ways in which patients “actively employ strategies in order to mitigate the potentially disabling impact” of illness on everyday life (Gabe, Bury, & Ramsay, 2002, p. 1621). More broadly, Bury describes three perspectives that tend to highlight different aspects of chronic illness:

a structural perspective in which the occurrence of chronic disorders is seen to be socially patterned and related to social divisions; an interactionist perspective in which chronic illness is seen as an emergent and disruptive experience shaped by cultural expectations and interactional difficulties; and a postmodern perspective in which personal experience and chronic illness narratives are characterised by discourses which emphasise opportunities for personal growth and change. (2002, p. 7)

Although scholars may disagree about the relative merits of those approaches, most recognise that socioeconomic, cultural and phenomenological factors shape the illness experience. What remains a matter of debate, is the exact nature of the knowledge acquired, generated or mobilised by patients. First, locating the patient's knowledge within the boundaries of individual experiences may be problematic: “When patients share experiential knowledge, the communal body of knowledge exceeds the boundaries of individual experience” (Caron-Flinterman et al., 2005, p. 2576). Second, the expertise researchers attribute to (chronic) patients may be more than just experiential knowledge; it may be an amalgamation of different kinds of knowledge. Fox et al, examining the use of a weight-loss drug by participants in an Internet forum, observed a form of “expert reflexivity” that “emerged from a community that reinforced the normative messages of the medical model in its ethos and its commitments” (Fox, Ward, & O’Rourke, 2005, p. 1307).

Because of the above-mentioned social dynamics that pervade focus group research and the fact that patients’ knowledge may prove heterogeneous and permeable, we believe that focus groups should be conceptualised as social spaces in which participants themselves construct the “patient's view.” Within this perspective, the ways in which knowledge is articulated and information censured become paramount (Kitzinger, 1994). During discussion, participants may align themselves with or reject previous members’ statements (Hydén & Bülow, 2003). They may try to persuade each other of the rightness of their claim (Kitzinger, 1994). They may also exaggerate their experiences or opinions in order to create symbolic effects on group members or to reinforce certain aspects of their projected identity (Smithson, 2000).

Perhaps more subtle is the recognition that while everyone may initially agree with a certain claim, dissent may set in when participants are asked to explain what they mean. Often closing “placatory” comments or jokes are used to end a topic when a “final” resolution is unlikely to be achieved (Crossley, 2002). Nevertheless, participants who are seen as possessing experience tend to dominate discussions and are deferred to by other group members who see them as the experts (Smithson, 2000). Depending on group composition, there may be several different types of competing expertise (Crossley, 2002). Although participants’ narratives may form an apparent consensus, it is also true that:

Opinions stated in the groups should be viewed not as previously formed, static things which people brought to the focus group, but as constructed in social situations. Neither should these opinions be treated as ‘belonging’ to individuals with the group, or as opinions held by the whole group, but as discourses which emerge in this context. (Smithson, 2000, p. 116)

Indeed, some opinions and ideas may not be already there, located inside the individuals who were recruited, but rather mutually formed through the interactive process (Hollander, 2004). Similarly, when participants express their thoughts and attitudes, they are not “simply ‘reflecting’ something that already existed in the heads, but performing certain social actions (such as allocating blame, making excuses, accepting responsibility, etc.)” (Crossley, 2002, p. 1472).

Hence, focus groups are social spaces in which a set of co-constructed narratives are debated or ignored as several social contexts interact through and structure the discussions. Our empirical examples will illustrate how analysing the interactive features that pervade knowledge-based deliberations exposes subtle dynamics at play throughout the entire process.

Section snippets

How patients construct the patient's view in focus groups on high-tech home care

In 2002, we conducted five focus groups with patients using four high-tech home care interventions: antibiotic intravenous therapy (IV), parenteral nutrition (PN), peritoneal dialysis (PD), and oxygen therapy (02). These interventions were chosen because they are common, and because they vary in their technical and clinical features and so likely differentially influence how patients perceive their value and limitations.

An analytical template for patient focus groups

By summarising insights from the literature and exploring empirical material, this paper so far showed how focus group research requirements and assumptions often conflict with participants’ everyday interactional routines, and how social dynamics shape the knowledge construction processes that unfold in focus groups. We now turn our attention to developing an analytical template that systematises this knowledge in a form researchers can use to identify and address the interactive features of

Conclusion

In our attempts to rethink “what focus groups are actually teaching us” (Hollander, 2004, p. 627), we have synthesised and elaborated on reflexive work by scholars who have underscored the sociological and communicative underpinnings of focus groups. We have also used empirical data obtained from our study with patient focus groups. Our aim was to develop an analytical tool that could facilitate the identification and analysis of the collective dynamics through which multiple, competing

Acknowledgement

This research was funded by an operating grant from the Canadian Institutes of Health Research (CIHR; #15472). When the study was conducted, the first author benefited from a career award from the CIHR Institute for Health Policy and Services Research (IHPSR, 2003–2008; #SNI-65571). We are grateful to the patients who participated in the study. A research assistant, Murielle Vergnhes, collected and provided a preliminary analysis of the data presented in this paper. We would like to thank the

References (24)

  • J.A. Hollander

    The social contexts of focus groups

    Journal of Contemporary Ethnography

    (2004)
  • L.-C. Hydén et al.

    Who's talking: Drawing conclusions from focus groups—some methodological considerations

    International Journal of Social Research Methodology

    (2003)
  • Cited by (171)

    • Individual interviews versus focus groups for evaluations of international development programs: Systematic testing of method performance to elicit sensitive information in a justice study in Haiti

      2023, Evaluation and Program Planning
      Citation Excerpt :

      What might explain this difference between findings in studies in health versus resilience domains? Drawing on the classic work of Lee and Renzetti (1990) on threats and costs to participants as the hallmarks of sensitive topics, AUTHOR⁎⁎ et al. (2010, p.89) suggest that sharing sensitive information in a FGD may depend upon participant perceptions of “opportunity to exchange important information or solve a pressing problem”. The costs and benefits of disclosure in FGDs have rarely been discussed outside of health and environmental research, and represent new terrain for evaluation research in international development.

    View all citing articles on Scopus
    View full text