Patient confidentiality within the context of group medical visits: is there cause for concern?

BACKGROUND: Group medical visits (GMVs), clinical encounters with a medical component delivered to groups of patients, have emerged as an innovative approach to potentially increasing efficiency while enhancing the quality of primary health care (PHC). GMVs have created the need to pay explicit attention to patient confidentiality.
OBJECTIVE: What strategies are used by providers and patients to address issues of confidentiality within GMVs?
DESIGN: In-depth interviews were conducted with 34 PHC providers and 29 patients living in nine rural communities in British Columbia, Canada. Data were analysed using interpretive thematic analysis and a relational autonomy approach.
RESULTS: We found three main themes: (i) choosing to disclose: balancing benefits and drawbacks of GMVs, (ii) maintaining confidentiality in GMVs and (iii) gaining strength from interdependent relationships: patients learning from each other. Confidentiality can be addressed and was not a major concern for patients attending or providers facilitating GMVs in these rural communities. DISCUSSION: Patients adopted strategies to address their own and others' concerns related to confidential health information. Providers used multiple strategies to maintain confidentiality within the group, including renegotiating what information is shared and providing examples of what information ought to be kept confidential.
CONCLUSIONS: Although GMVs are not for all patients, a relational autonomy approach is useful in drawing attention to the context and structures which may influence their patients' ability to act autonomously. Successful delivery of GMVs requires both patients and providers to negotiate between maintaining confidentiality and an appropriate level of disclosure.

Introduction

Information exchanged confidentially between a patient and his/her health‐care provider (e.g. physician, nurse, nurse practitioner and dietician) is central to a therapeutic relationship. With few exceptions, providers are bound by laws and codes of ethics to uphold confidentiality and cannot willingly divulge any information related to a patient without that person's consent.1, 2 Maintaining confidentiality is the responsibility of health‐care providers3 but a right of patients. In maintaining confidentiality, health‐care providers must exercise their fiduciary duty by acting in the patient's best interests to maximize benefits and minimize potentially harmful outcomes. However, in specific instances where public interest overrides the patient's right to confidentiality, providers who ‘fail to report’ may be prosecuted or penalized if a third party has been harmed.4 The disclosed information should not exceed that of what is required to protect others.5

Respect for patient autonomy, where individuals control what information is disclosed and to whom, is a central and practical concern to providers, patients, and others such as bioethicists.6 An implicit assumption occurring in relation to conventional modes (e.g. individual visits) of primary health care (PHC) is that patients control their own information and can reasonably expect that their health information will be kept confidential and protected from unauthorized access. Treatment decisions are made by the provider with informed consent from the patient. That is, patients in an individual visit would express formal agreement with a proposed treatment after receiving clear information regarding all relevant information. Ideally, these decisions, including those about information shared and treatment, are made without coercion and that best express the outcomes of patients' own deliberative processes.6

Yet, with increased efforts to transform the PHC system, group medical visits (GMVs) have emerged as an approach to enhancing the quality of PHC delivery7 and are considered an acceptable way of providing PHC services to groups of patients.8, 9 PHC is where individuals first make contact with the health‐care system to try and resolve short‐term health issues, manage chronic conditions or prevent health problems from developing.10 Within the context of GMVs, the traditional approach to personal patient autonomy must be renegotiated. First, when ill, patients often experience a greater dependency on others, thus have a greater likelihood that they will have their interests and values overridden by others.6 Second, patients being invited to participate in GMVs often raise concerns about whether the information they share will be kept confidential. Providers also wonder about the appropriateness of the information being shared by patients, and using patients' experience, with consent, as a teaching tool to assist others. Although providers are legally and ethically bound to maintain confidentiality, individuals attending GMVs as patients are not legally obligated to do so.11, 12 Finally, living in a small town increases the likelihood of having interactions with individuals who see the same primary care provider. Patients may feel constrained from speaking about their health‐care needs or sharing information for fear that they may be judged by their provider or others.

Brief description of GMVs

GMVs are typically facilitated by a family physician or nurse practitioner. GMVs offer all components of an individual clinical encounter but are delivered to groups of patients ranging in size from 12 to 20 individuals.13 GMVs are unique in their ability to deliver medical care, health promotion, chronic disease management, health education and group support simultaneously. There are two broad types of GMVs: (i) ‘Homogenous’; (a) co‐operative health‐care clinics, (b) physicals and shared medical appointments, and (ii) ‘Heterogeneous’ or drop‐in group medical appointments.9, 14 All GMVs include an education component that is delivered in a group setting, where information sharing between peers is encouraged. Homogenous GMVs deliver services to patients who share common health‐care needs, such as those acquiring the skills needed to manage their chronic condition or needing an annual physical examination. One of the main goals of the homogenous GMVs is to increase a patient's toolkit in being able to self‐manage their health within their own community, to the best of their abilities. In the co‐operative health‐care clinic model, brief individual medical visits, if necessary, and any physical examination take place either in the group setting or at the end of the GMV. Physicals and shared medical appointments are designed such that brief individual physical examinations take place privately before the group session, while discussion and medical management still take place in the group.15, 16, 17 Heterogeneous GMVs are meant to increase access to PHC.14, 18 Both types of GMVs are different from group sharing or any type of support or health educational group in that they also increase the quality and comprehensiveness of PHC services because professionals of different disciplines meet with patients in a group to evaluate patients, adjust medication, coordinate and deliver preventive services.19, 20

Past work documenting GMVs in PHC suggests their implementation has been mainly with patients who have chronic conditions such as diabetes,19, 21, 22 hypertension, stroke, heart failure 23, 24 and chronic obstructive pulmonary disease.25 GMVs have also been used to deliver prenatal care.26 While delivering health services using a group format is not new, GMVs in the context of PHC are relatively recent in Canada.7, 27

GMVs were introduced in the United States in the 1990s to optimize time and quality of care to a growing number of patients who have one or more chronic conditions.20, 28 Their use has been experimented with in family medicine, managed care, health maintenance and socialized medical systems29 and highlighted in chronic disease management reviews.30 Whether GMVs improve health outcomes remains debatable; some studies provide evidence of improved physiological outcomes such as improved glycemic and lipid control, lower body mass index and higher levels of knowledge about the disease for patients with type 2 diabetes.31, 32 Other studies, however, have failed to document a positive effect.33, 34, 35, 36 Even if GMVs only provided equivalent care as individual visits, the use of time spent with patients could increase equity in access to PHC services and resources needed for self‐management of chronic conditions.37

A relational autonomy approach could mitigate some of these challenges because it can extend the concept of an individualistic conceptualization of autonomy, asking us to take into account the impact of social contexts and political structures on the lives of patients. Relational autonomy considers individuals as social beings, whose identities are developed and maintained within a complex web of social interactions.38 It draws attention to the fact that autonomy and decision making are affected by social forces and that structures (e.g. GMVs) and contexts (e.g. rural community) need to be taken into account in the provision of health care.38 The professional provider–patient relationship, particularly in small communities, is frequently both long‐term and personal.39 The relationship is not limited to one provider and his/her patient but frequently involves a family, mutual friends and overlapping relationships where information sharing may be viewed as an important part of normal networking.40, 41 In such a setting, maintaining confidentiality becomes challenging.39 To date, little work has examined the implications of GMVs on patient confidentiality.42, 43 The purpose of this paper is to examine patient confidentiality within the context of GMVs through a relational autonomy lens.

Methods

Context of the study

The Northern Health Authority of British Columbia (one of six health authorities) approached our research team with a request to examine the effectiveness of a new format for the delivery of PHC, group medical visits (GMVs). Providing PHC in rural communities is impacted by geographical and professional isolation and limited access to resources.3, 44 Northern Health, together with its health‐care providers, has and continues to develop innovative ways to deliver health services and ensure better access to care for its residents.45 One such innovation is the use of GMVs delivered in primary care practices, space provided by Northern Health or community clinics located on a First Nation (FN) reserve. In many rural communities across Canada, the majority of Aboriginal residents live on a FN reserve. A reserve is a small portion of what might have some time been part of a nation's traditional territory, which is protection by legislation for the use and benefit of a FN.46

Participants

This mixed‐method study included in‐depth interviews with patients attending and providers facilitating GMVs, and direct observation of GMVs. Rural practices delivering PHC and FN communities were identified by our Northern Health Authority partner as potential participants. Among those who were identified, rural practices and FN communities were recruited if they had offered GMVs for more than 2 years. Nine communities, consisting of five family physician‐led primary care practices and four FN communities, agreed to participate in our study. The communities ranged in size from 200 to 76 000 people. Participating communities in the study were anywhere from 100 to 1000 km away from Prince George, a community known as British Columbia's northern capital.47

Procedures

Patients and PHC providers were recruited through our community partners. Our partners were either PHC coordinators, a contact person in the practice, or a community health representative from the FN communities. PHC coordinators are health authority employees hired to facilitate access and service delivery of PHC in specified geographical areas. Community health representatives are members of FN communities who play a key role in on‐reserve health‐care teams in the areas of health promotion, protection and injury prevention and are often change agents within their communities.48 In using PHC coordinators and community health representatives to help in the recruitment process, care was taken to adhere to ethical principles as outlined in the Tri‐Council Policy Statement: Ethical Conduct for Research Involving Humans, 2010.49

Inclusion criteria for patient participants included (i) adults aged 19 years or older; (ii) living in a rural community in Northern Health; and (iii) having no significant cognitive impairment. Providers were recruited if they had either provided or taken part in GMVs in the past year. Basic demographic data were collected for both patients and providers. Research team members conducted interviews either in‐person or over the telephone. All procedures were approved by the University of British Columbia, University of Northern British Columbia, Northern Health ethics boards and PHC practices and FN communities. Memoranda of understanding were negotiated and agreed upon with each of the FN communities. Our engagement with FN communities was guided by the 4Rs (respect, relevance, reciprocity and responsibility) framework adopted by the Network Environments for Aboriginal Research British Columbia. These procedures are also consistent with Tri‐Council Guidelines for Conducting Research in Aboriginal Communities.50

Purposive sampling51 was used to explore patients' and providers' experiences in attending or facilitating a GMV, respectively. This type of sampling was used to identify participants who could speak to particular issues (e.g. gendered perspectives) or underdeveloped areas identified in the iterative analysis of the data. After obtaining signed informed consent, in‐depth individual interviews with PHC providers and patients and one group interview with two patients were conducted by one of three research team members (STW, JL and the research assistant LF). Open‐ended questions were asked to explore (a) participant's experiences; (b) what was critical for providers to the successful deliver of GMVs; (c) whether there were any barriers to delivering or receiving PHC in a group format; and (d) what participants would recommend for improvement of GMVs. Patient participants were specifically asked whether they were concerned with sharing personal information and how they negotiated confidentiality within the GMV. Providers were asked what they did to maintain confidentiality within the GMV. Interviews were audio‐recorded, and field notes were written immediately following the interviews. All participants were given $15 each in appreciation of their time. In addition, field notes were written after members of the research team directly observed GMVs.

Data analysis

We drew on a relational autonomy approach52 to critically analyse the data because: (a) patients disclose information to fellow group members with no guarantee that others will keep information private53 and; (b) issues of confidentiality cannot be considered in isolation of the context of living in a rural area where people are more likely to have overlapping social networks. Individual choices in the rural context, compared to an urban setting, can be constrained by geography, familiarity with others in their community, transportation, economic considerations, weather and limited numbers of providers.39

Interview recordings were transcribed and anonymized. The transcripts were then compared with the audio recordings for accuracy. The research team independently read the transcripts and conducted interpretive thematic analysis according to procedures for qualitatively derived data.54 A coding scheme was iteratively developed for the data (interviews and field notes), reviewed through peer evaluation and organized using Atlas TI, a qualitative software package. The need to clarify and refine emerging conceptualizations or patterns informed subsequent sampling. Thematic interpretations were discussed with the research team, along with the consideration of alternative interpretations, which led to a more conceptual representation of the themes reflected in the data. An audit trail of the analytic decisions was maintained. Trustworthiness of the analysis2 was discussed among the research team and evaluated through discussions with PHC experts and feedback from a selection of participants on the aggregated results.

Results

Sixty‐three participants discussed their experiences with GMVs. Providers included family physicians (n = 10), nurses (n = 7), nurse practitioners (n = 2), PHC coordinators (n = 4), other allied health workers (n = 11), such as nutritionists, social workers, medical office assistants and community health representatives. Different types of GMVs were delivered to groups of patients ranging in size from 12 to 20 individuals (Table 1). Table S1 shows that patient participants (n = 29) were an average age of 62 years, mostly female, and married. Patients reported being either Caucasian (55%) or of indigenous ancestry (41%). Almost half of patient participants reported a household income of less than $30 000 CDN. Patients had attended an average of four GMVs in the past year. Most participants (83%) attended homogenous GMVs. Of those who attended homogenous GMVs, one participant had attended a men's group for diabetes and three participants had attended a women's only group for smoking cessation, depression or diabetes, respectively. All remaining GMVs, including prenatal groups, were of mixed gender. Most patients (55%) reported having three or more chronic conditions, with the three most common conditions being diabetes (59%), high blood pressure (52%) and arthritis (48%).

Table 1

Types of group medical visits delivereda

Types of GMVs delivered+
Cooperative Health Clinic model/Homogenous	88.6
Drop‐in Group Medical appointments/ Heterogenous	34.3
Mean (SD)	1.4 (1.9)
Range	1‐6

Providers were asked to list all types of GMVs delivered.

Our analysis resulted in three main themes: (i) choosing to disclose: balancing benefits and drawbacks of GMVs, (ii) maintaining confidentiality in GMVs and (iii) gaining strength from interdependent relationships: patients learning from each other (Table S2). Perspectives about confidentiality were similar across the sample, regardless of type of GMV attended, gender, age or ethnicity.

Choosing to disclose: balancing benefits and drawbacks of the GMV

Group medical visits can foster relational autonomy because patients are able to draw upon more informational resources and social support from attendees and often feel more empowered to pose questions to their providers than they might otherwise in individual encounters. The data from our study suggest that GMVs created an environment where patients shared personal information within the context of their relationships with peers and their provider. Patients can consider their treatment plans and options within their context and social relationships. Sharing information with other GMV attendees and being able to gain access to their PHC provider at times convenient to them was considered a benefit by most patients.

Some patients from providers' practices were invited to participate in a GMV declined to attend. Although none of those patients were interviewed as part of this study, providers (e.g. Provider #31; female) reported that the most common reason for not attending a GMV was patients' concerns about confidentiality and hence a preference for individual visits. Clearly, for some patients, attending a GMV raised discomfort in sharing personal information or concerns about breaches in confidentiality. However, patients who did attend a GMV consciously selected which information they were comfortable sharing in a group situation. One patient (Patient # 8; 59 years, male) expressed sensitivity about information he would share in a mixed gender group. Having to ‘filter’ what information is shared in a GMV could be seen as a drawback. Yet, patients voiced that there was strength in numbers when discussing health‐care needs and that the health information shared was intended to be useful to others living with the same chronic condition (Patient # 15; 77 years, female).

The majority of patients (e.g. Patient # 17; 66 years, female) pointed out that they were not concerned if the content of what they shared did not remain confidential (and some felt that this was inevitable, especially in smaller rural communities). Patients reported learning from others' experiences, gaining additional information from their provider based on his/her responses to other attendees' questions and ability to schedule an individual follow‐up appointment, if necessary. Both patients and providers also reported that patients felt less intimidated and more secure interacting with PHC providers in a group, thus sharing more health information. Few participants discussed breaches in confidentiality.

Most providers had considered the issue of patient confidentiality prior to implementing GMVs. Direct observation of GMVs revealed that the sharing of personal health information and the individual's contextual circumstances required frequent renegotiation. Rather than having an implicit assumption that all information exchanged needed to remain confidential, providers reported that it was important to consciously attending to confidentiality. Providers often asked patients to give verbal consent prior to sharing specific information such as test results. Despite the amount of attention to confidentiality and other processes (e.g. scheduling, having charts available, group dynamics) that were needed to implement GMVs, providers perceived the benefits of the GMV format as greater than the drawbacks for patients. Some providers (e.g. Provider # 10; female and Provider # 21; female) shared the view that they were ‘surprised’ by what was shared: attendees were willing to disclose personal information that providers considered highly sensitive (e.g. the medical diagnoses, types of prescription medications).

Providers reported that reflection and discussion about how to maintain confidentiality was necessary with all office staff and other health‐care providers. Office staff needed to familiarize themselves with the GMV concept in order to speak comfortably when addressing patients' concerns about confidentiality. Coordinating groups of 12–20 patients required an understanding how services were delivered in a group setting (e.g. blood pressure and flu shots but examinations requiring privacy would take place in a separate room), how confidentiality is maintained, and a clear articulation of the GMV process. As one provider (Provider # 4; female) explained, changing the flow of a practice to incorporate GMVs required work by all providers and staff. Given that other providers or community organizations (e.g. dietician, recreation centre manager and public health nurse) were involved, support from office staff was especially critical.

Maintaining confidentiality in GMVs

While providers re‐confirmed cautions about sharing personal information, attendees had to sort out what information to share during a GMV and how they would respect others' confidential information within their small community. GMV participants also determined how they would react if they ran into someone from the GMV in the community. As one patient (Patient # 19; 72 years, male) pointed out, when group members saw each other in the community and information was shared outside the GMV context, it created discomfort. Living in a small community can seem like a ‘fishbowl’. Yet, in this study, concerns about confidentiality quickly became less prominent as the group became more familiar with the group's purpose and dynamics, as indicated by one patient who said that he initially held back but eventually became comfortable in sharing his health and personal information (Patient # 17; 66 years, female). Participants reported having more opportunities to foster supportive bonds with each other because they shared personal and geographical contexts. This perspective was echoed by providers.

Prior to attending their first GMV, the office staff (e.g. medical office assistant, nurse) would call potential participants and explain this alternative form of medical appointment. During this interaction, patients were provided with some information about confidentiality and signing a confidentiality agreement. Immediately before starting the GMV, the provider would reiterate the content on confidentiality; Participants did not need to share their own health information if they did not feel comfortable, they could leave at any time during the visit and that they should not discuss other people's health information outside the GMV (Provider #22, male). Providers asked patients to choose what they share especially as it was likely that they might see each other in different contexts because they all lived in a small community.

Most providers discussed following a process where patients signed a confidentiality form at the start of their first GMV and reminding them at each subsequent GMV that discussion was confidential. Providers also discussed what confidentiality meant with GMV attendees by providing examples of what information was acceptable for discussion outside the session. For example, one provider explained to the group that if he/she learned something about thyroid disease, then this information could be shared with others. The provider went on to tell the group that what was to remain confidential was who ‘Mrs. Jones, our neighbor’ was the person who has a thyroid condition (Provider #3, male). This example reflects the fact that in small towns, people attending the GMVs are likely to know each other but that information still needed to be kept confidential Providers also explained that patients were free to share their personal information or simply listen to other GMV attendees. Some providers (e.g. Provider # 2; female) emphasized that although they could not guarantee confidentiality and that sharing personal information was within the control of each GMV participant.

Strategies were actively used to maintain confidentiality during GMVs including: providers asking permission of patients to share certain types of information (e.g. laboratory values), patients specifically being given the opportunity to decline sharing information, but still participate in the GMV; and letting patients know they could always follow up with an individual visit. Providers were observed to explicitly addressed confidentiality at the beginning of each GMV.

Participants reported that conversations about confidentiality at the beginning of every GMV and setting the ‘ground rules' for keeping information private helped people choose what personal information was shared. Patients spoke of maintaining confidentiality through not divulging any information or gaining mutual consent outside the GMV. They were observed to ask permission from each other before discussing their challenges of day‐to‐day management of their health. Patients also described how they conveyed respect for other group members through their actions. For example, patients (e.g. Patient # 6; 71 years, female) described how the phrase, ‘How are you doing?’ took on a new meaning. If they met someone in the community who had attended the same GMV, this phrase was described as being respectful and supportive to the person but as not breaching confidentiality.

Gaining strength from interdependent relationships: patients learning from each other

A relational autonomy lens sheds light on the multiple layers of interdependence that creates a welcoming and emotionally safe health‐care environment for those attending GMVs. For patients who attended GMVs, inherent interdependencies such as sharing the same rural context, community, health condition and PHC provider were seen as strengths. While keeping personal information confidential was a concern expressed by both providers and patients, many GMV patients reported benefitting from this familiarity with each other. Some participants knew each other, whiles others suggested that being from the same community and having to manage their health condition(s) provided enough familiarity in order to gain new knowledge, empathy and support from each other. One patient (Patient # 9; 59 years, female) noted that the information shared in the group was less important than being part of a group, in addition to knowing that they were not alone in their day‐to‐day management of their health. Providers reported (e.g. Provider # 2; female, Provider # 4; female) that peer support between attendees was a strong benefit of GMVs.

Within the GMV, interdependence was also created through sharing stories and exchanging information such as laboratory or diagnostic test results. For example, providers, after agreement from each patient, would write anonymized haemoglobin A1Cs on a white board and discuss the meaning of the results with the group. Many patients reported this type of discussion as a strong motivator for their own self‐management (Patients # 5; 31 years, female and #16; 74 years, male). Participants' enthusiasm to learn lifestyle strategies overshadowed concerns about ‘who’ knew their personal information. Patients reported feeling more of a sense of individual control, in part, because they had the opportunity to more thoroughly address the complexities of day‐to‐day management of their health. Interaction with others (e.g. seeing new perspectives) along with obtaining important medical and health information that would not otherwise be available was welcomed. Others (e.g. Patient # 8; 59 years, male) emphasized the fact that both patients and providers learned more in a group session than from individual visits.

Finally, strength was gained from patients being in a group and knowing that their PHC provider would be actively listening to their perspectives. Patients (e.g. Patient # 16; 74 years, male, Patient # 6; 71 years, female) reported being less intimidated by the PHC providers and more secure in divulging personal information than in an individual visit. Health‐care providers also pointed out that the inherent power differential between providers and patients was minimized because of the group format. Providers (e.g. Provider # 22; female) reasoned that patients may feel more comfortable in a group where they could engage with providers in ways that were empowering. For some patients, their PHC needs may be better met when care is provided in the context of relationships. Patients and providers discussed the importance of respecting patients' preferences in selecting care options (GMVs and/or individual visits), peoples' confidence participating in a group and the importance of having enough time for sharing information.

Discussion and conclusion

Examining the implications of the GMV format for patient confidentiality in rural Canadian settings is unique to this study. Implementing GMVs in PHC has created a space to promote the relational autonomy of patients who may feel they are not able to express their health concerns or needs adequately in an individual visit. A relational autonomy lens points out that, for some patients, the structure of GMVs provides a welcoming and safe place to gain social support from others who live in a similar context with the same health condition, decide how to manage the symptoms of their condition based on their provider's suggestions and others' information and share with others their experiences. In these GMVs, individuals exercise their autonomy in choosing what health information they share and committing to keeping others' health information confidential. This autonomy takes place in relation to living in a community where it is more likely that people know each other and where resources to support successful management of his/her medical condition is limited. Moreover, the creation of interdependent relationships among patients and patients with their provider enables both parties to participate in GMVs knowing that confidentiality of their health information could be compromised. Strategies for maintaining patient confidentiality can be successfully implemented by providers. Patients reported benefits from attending GMVs. Their positive experiences promoted trust and openness, allowing for a greater degree of self‐disclosure and confidence in posing questions to their providers.53 Consistent with previous reports,55, 56 this study shows that potential breaches in confidentiality were not a major concern for GMV attendees or those delivering GMVs.

Navigating around issues of confidentiality in GMVs is built on a foundation of trust between patients, their provider and office staff.57 Our analysis suggests that a relational autonomy lens extends the concepts of confidentiality and individualistic notions of autonomy because it requires providers to recognize that patients are influenced by social and political circumstances which can affect their ability to act in different situations.52, 58 A relational autonomy approach prompts providers to take into account the impact of social contexts and political structures on the lives of patients.52, 58 It draws attention to the fact that autonomy and decision making are affected by social forces and that structures (e.g. GMVs) and contexts (e.g. rural community) need to be taken into account in the provision of health care.59 In contrast, an individualistic perspective of patient autonomy overlooks power dynamics and may not account for the pressures and vulnerabilities, or social contexts, and patients may experience when seeking care or how such pressures may affect their capacity to exercise choice including with whom to share information.60 A more rounded consideration of peoples' social context and the adverse effects this may have on their health condition, and the choices they make with respect to confidentiality, is thus warranted.

While it has been reported elsewhere that providers of GMVs use strategies to address concerns about confidentiality,61, 62 these results illuminate how and when these strategies can be used by both providers and patients. In addition, our results show that a relational autonomy approach helps draw attention to the interdependence of social relationships and how this can create opportunities for fostering greater involvement from patients in managing their health.

Providing GMVs, particularly in rural communities, may be a feasible and an effective addition to the traditional individual visit. Providing PHC in these communities is unique due to a number of factors such as limited access to resources (e.g. limited choice of PHC providers, no specialists) and overlapping social connections extending throughout the community.3, 44 With incorporation of explicit discussions about confidentiality and having attendees sign confidentiality agreements, hence foregrounding the importance of non‐disclosure outside the group, GMVs can be successfully implemented.

Limitations of this work must be considered. Data were only collected from providers delivering GMVs or GMV attendees in rural communities. Observational data were similarly limited to GMVs involving patients and providers who were comfortable in group settings. Future work should include patients who have declined to attend a GMV and with those living in urban areas. Another limitation is that our research team did not explore to any depth which information patients and providers consider confidential. Future work could examine what information these individuals consider confidential and the shifting role of patient and community responsibilities in maintaining confidentiality. Nonetheless, our results provide a window into the issues of confidentiality within PHC from patient and provider perspectives. With appropriate strategies to address confidentiality, our research, along with others' on international experience with GMVs in other health‐care settings, shows that patients attending GMVs can safely and confidently discuss their health and health‐care needs.28, 63

Although GMVs are not suitable for all patients, they do offer an opportunity for some patients to access quality PHC. We acknowledge that issues of confidentiality, however, remain complex, particularly within the context of a setting where personal health information is exchanged and medical treatment is being provided. Providers have little control over the amount and nature of information disclosed by group members within and outside the group.

This study illuminates the complex and evolving nature of confidentiality in a group setting. Some of these same issues surrounding confidentiality may occur in other kinds of group work such as support groups. However, GMVs are a specific type of visit because they involve a group of patients and their PHC provider. The provider has legal responsibilities, is able to order medical tests and is required to document the contents of the visit in the patient's chart. Our study aligns with past research on GMVs shows that providers are able to diagnose and medically treat patients8 and interact with patients in a more synergistic and productive relationship.64 Further work is needed to understand if patients and providers have similar ideas of what information is considered confidential, if this is similar across patients who have different types of chronic conditions (e.g. HIV positive vs. diabetes) or multiple conditions and particularly given the rapid expansion of technologies that permit sharing of any personal information through the use of social media such as Facebook, Twitter and YouTube.

As discussed in this paper, a relational autonomy approach applied to the context of GMV is useful in drawing attention to the context and structures which may influence their patients' ability to act autonomously when deciding whether or not to participation in GMVs. GMVs may not be amenable to all PHC delivery or all patients' preferences, yet many patients can participate in GMVs. Both providers and GMV attendees use strategies to maintain confidentiality within the GMV and in the community. Concerns about breaching confidentiality should not prevent wider implementation of GMVs into PHC given the significant role they can play in expanding access to timely care and their capacity for managing health issues.

Authors' contributions

STW and JL conceptualized the study, conducted the in‐depth interviews, led the data analysis and led the drafting of this article. AJB and MLPM were involved in conceptualizing the study, provided feedback on the interview guide, confirmed themes and subthemes and provided feedback on the article. MC assisted in drafting the article, coding data and provided input into the analysis. All authors read, edited and approved the final manuscript.

Competing interests

The authors declare that they have no competing interest.

Table S1. Characteristics of patients (n =29) attending Group Medical Visits.

Table S2. Confidentiality themes and illustrative quotes.

Click here for additional data file.