Opportunities,challenges and ethical issues associated with conducting community-based participatory research in a hospital setting

Abstract

Community-based participatory research (CBPR) is growing in popularity as a research strategy to engage communities affected by health issues. Although much has been written about the benefits of using CBPR with diverse groups, this research has usually taken place in community-based organizations which offer social services and programs. The purpose of this article is to explore the opportunities and challenges encountered during a CBPR project conducted in a small hospital serving people living with HIV and addictions issues. The structure of hospital-based care delivery required the team to account for participants’ health-related limitations, including adopting recruitment strategies built on clinical care, and modifying the original research design to better protect participant confidentiality in a small space. Although CBPR offers an important strategy for researching with members of hospitalized communities, it requires research teams to develop context-relevant approaches to ethically engaging community members that recognize the ways clinical space mediates the research process.

Introduction

In the United Kingdom (UK) (deAlmeida, 2012), United States (The Centers for Disease Control and Prevention, 2011; Community Based Participatory Research Program, National Institutes of Health) and Canada (Canadian Institutes of Health Research, 2009) efforts are under way to increase public and patient involvement in health research. In the UK, the National Institute of Health Research (NIHR) funds INVOLVE (INVOLVE, 2012), a national advisory group tasked with building and monitoring capacity to involve the public in health research. Community-based participatory research (CBPR) has emerged as a model for the equitable involvement of community in all aspects of the research process and for examining health disparities from the perspectives of those most affected (Israel et al., 1998; Minkler, 2005; Wallerstein et al., 2011). The benefits of CBPR have been well documented (Jagosh et al., 2012; Viswanathan et al., 2004), and funders have encouraged community engagement in health research (deAlmeida, 2012; DiGirolamo et al., 2012; Rosenthal et al., 2009; Williams et al., 2010). However, much of the CBPR literature reports on projects conducted with community-based organizations (CBOs) that provide primarily social service programing. Few studies have been conducted with hospitalized individuals and groups. Nevertheless, support for this approach is demonstrated by increased community involvement in practice-based research networks (PBRN) in the United States. Indeed, involvement in PBRNs is expected to lead to asking clinical questions that matter to patients and that are relevant to physicians (Westfall et al., 2006). Evaluation of mandated community engagement within studies funded by the Patient Centred Outcome Research Institute shows that increasing community involvement in research resulted in changes that led to improved study outcomes, goals, methods, interventions and materials (Forsythe et al., 2016). CBPR holds promise for use in hospitals and has been shown to improve medical training (DeHaven et al., 2011; Hunt et al., 2011), medical research (Macaulay and Nutting, 2006; Tapp and Dulin, 2010) and patient care (Boutin-Foster et al., 2008; Macaulay, 2007).

Within CBPR projects, community member involvement can vary from advisory roles to active participation in the project design and implementation. Often, these projects attract community members whose health status enables a certain level of participation. Individuals in hospital are usually the most affected by health issues and are often disadvantaged to participate as a result of health-related limitations (e.g. reduced mobility). Thus current practices in CBPR may present barriers to participation for some. In this article we reflect on issues that arose during a participatory project we conducted to examine how to engage in research projects with multi-morbidity individuals (Fortin et al., 2007) who are currently hospitalized or recently discharged but still receiving hospital care. Our goal is to offer reflections, discussion and examples of the way researchers working in hospitals might adapt their methods to increase participatory and ethical community engagement within these settings. In doing so, we respond to calls for researchers to document and publish their experiences with community engagement (Forsythe et al., 2016).

Project and research setting

Casey House Hospital is a 13-bed facility providing both in-patient and home care services to people living with HIV/AIDS in Toronto, Canada. Individuals with HIV are admitted for sub-acute rehabilitative care, medical and psychiatric symptom control, post-hospitalization discharge support, end-of-life palliative care or respite care. The clinical team and the hospital board responded to the growing numbers of clients who in addition to HIV-related acute illnesses presented with active substance use issues by implementing a ‘harm-reduction’ policy. Harm reduction is an approach that seeks to reduce the risks associated with substance use (Marlatt and Witkiewitz, 2010). The Casey House Hospital clinical management team, in collaboration with Strike and Guta, were keen to evaluate this policy. Strike and Guta had considerable experience conducting CBPR in community settings and were committed to applying the same participatory and flexible techniques in a hospital setting. As a fledgling research group, our first task was to consult with the community and assess interest in possible research collaboration.

Defining community in a hospital-based CBPR research project

In CBPR, the ‘community’ refers to a group with bonds based on geographic proximity and/or shared experience of illness (Israel et al., 1998; Minkler and Wallerstein, 2008). This differs from medicine where ‘community’ commonly refers to individuals not hospitalized. For our purposes, we began our consultations defining ‘community’ in terms of a shared experience of living with HIV and accessing specialized care at Casey House Hospital. Our first consultation involved a series of informal meetings where community members (i.e. current inpatients and those discharged but still receiving care from Casey House Hospital clinicians) were invited to meet with Strike and Guta. With two groups of community members, Strike and Guta discussed the level of interest (if any) in collaborating on a research project about harm reduction, the model of research to be used (i.e. CBPR or not), feasibility issues related to participation and engagement at various stages of the research process, and how to define ‘community’ for the future projects.

These discussions led to several important observations. First, clients defined themselves as a ‘community’ whose experiences of receiving care in hospital set them apart from other people living with HIV (PHAs). Some of the clients had been receiving care at Casey House Hospital for many years and referred to it as a ‘home’. Second, clients shared an interest in evaluating the harm-reduction policy as some were supportive and others opposed to this policy. Third, clients were clear that their roles and opportunities for participation on research teams and as study participants would need to be adapted to reflect their current abilities and to avoid interference with their healthcare routines. In particular, the group advocated that the traditional approaches used to embed community voices in governance (e.g. membership on time-consuming advisory committees or naming clients as members of the research team) be abandoned in favour of a flexible approach where advice and direction is solicited during group meetings throughout the project and where any who wished to advise could attend (or not). Clients also expressed an interest in commenting on but not being involved in detailed analyses, leading us to use a community dissemination event to solicit feedback on analyses. Finally, clients expressed an interest in using arts-based research approaches because they were perceived to be more engaging than typical interview-based methods and are accessible for people with varied abilities.

Reflecting on these consultations, we formed a research team comprised of academic researchers and clinicians at Casey House Hospital, and invited current and recently discharged Casey House Hospital clients to attend flexible advisory group sessions. At the next flexible advisory group session, attendees were asked to participate in decisions about what type of arts-based methods (i.e. photovoice, collaborative muralling, creative writing and/or mixed media arts) to use in future assessments of the harm-reduction policy. Each approach was explained with examples. Preferred methods were selected by voting. The group also participated in offering suggestions and making decisions about implementation issues such as preferred time of day for research events, single- vs multi-session data collection methods, honoraria, methods and venues to disseminate findings, and also usefulness of supports such as appointment reminders. Based on voting, we moved forward with photovoice (Baker and Wang, 2006; Hergenrather et al., 2006; Vaughn et al., 2008), wherein participants took photographs to represent their experience of the harm-reduction policy. At the end of the photo project, we invited current and recently discharged Casey House Hospital clients to a community-friendly presentation of study findings to offer public feedback or anonymous feedback in writing.

Ethical challenges when recruiting within hospital settings

CBPR requires the same ethical protections as other forms of research but also invites thinking about the impact of research on whole communities (Flicker et al., 2007; Minkler, 2004). The ethical challenges we encountered at the intersections of community and hospital occurred in relation to recruitment, confidentiality and informed consent. Research teams are required to use the least intrusive and non-coercive methods to recruit participants. However, the members of the community were living with advanced HIV disease and related multi-morbidities, including cognitive decline and substance use, making it difficult for them to remember appointments without assistance and support. Clinicians (e.g. nurses and social workers) advised the research team they had developed systems for reminding clients about Casey House Hospital appointments and events. Using trusted persons to promote research activities is common in CBPR but raises questions about boundaries when clinicians take on this role. During the consultation process, we asked those who attended the flexible advisory group meeting how they wanted to be notified of future research activities and whether they wanted clinicians to be involved. Based on their recommendations and with client’s permission, the research coordinator provided regular reminders to clients by telephone and in person about research-related appointments. For clients who did not have access to or could not be reached by telephone or email, we asked in advance if reminders could be sent through clinicians or other on-site programs. The research coordinator balanced reminders with opportunities to withdraw from the study, and reminded participants that the research was separate from their other hospital-related appointments. This strategy was time-consuming but proved effective in engaging clients with multi-morbidities in recruitment activities. Community members expressed satisfaction with this approach.

The challenges of maintaining confidentiality

Maintaining confidentiality and working within hospital protocols required modification of our research design. CBPR projects promote the sharing of lived experience with the goal of fostering support and connection and mobilizing community members in response to their shared health needs. However, not all members of the community were supportive of the harm-reduction policy. Following the preliminary consultations, it was apparent that the sensitive nature of the issues being investigated (drug use and harm reduction) caused some clients to become upset, and impacted personal relationships in the social environment of this small facility. Although all participants experienced multi-morbidity, with some form of substance use being common, there were clear social risks associated with discussing substance use openly.

We initially planned separate photovoice sessions for active substance users and non-users to avoid conflict. However, most participants selected the photovoice session for non-substance users. Several clients disclosed to the research coordinator that they did not sign up for the other group because attending would identify them as a substance user. In response, we altered the design of the project from photovoice (which is often premised on a group analysis of a shared community concern) to photo-elicitation, where clients were asked to take photos and later discuss them during a one-on-one interview (Radley and Taylor, 2003). We retained a group process for the information, photo training workshops and community exhibit, and kept discussions task-focused to reduce the potential for conflict (Switzer et al., 2015).

In CBRP conducted in CBOs, participants could plausibly be there for any number of reasons (i.e. counselling, picking up resources etc.). However, in a hospital clients could not be as anonymous because clinicians and staff needed to know where they were for their safety and to provide care in the form of medications, specialized meals etc. As well, clients living in the community had to sign in at the front door and were expected to advise staff of who they were seeing. This issue was difficult to mitigate because we conducted the project at Casey House Hospital to ensure that location was not a barrier for participation. In this case, the need for confidentiality was balanced with clients’ desire to participate in research and also their trust in the integrity of the clinical staff. As a further protection we advised participants that anyone on the research team in a clinical role at Casey House Hospital would only have access to aggregated data.

Ensuring consent for multi-session research activities

Although we agree with the CBPR literature that advances the active participation of community members in research, there has been little discussion in the CBPR literature about obtaining informed consent with individuals managing multiple illnesses and related cognition issues. Marshall et al.’s (2012) case study of researching with youth labelled with intellectual disabilities is an important exception in the literature. We complemented standard consent approaches with a multi-step process that involved discussing the project during group sessions, then individually over the phone or in person, obtaining verbal and written consent before data collection, checking in with clients during data collection if they seemed confused or distracted, and following up afterwards to make sure they were still comfortable with their experience. Our primary concern was to ensure that clients experiencing HIV-associated neurocognitive disorders and/or cognitive impairment related to substance use would be adequately supported. During the consent process, we needed to explain confidentiality of the interview data but also discuss confidentiality and future use of the photos that would be taken. Individual sessions where consent was obtained allowed participants as much time as they needed to learn about the procedures, risks and benefits, ask questions and to determine, without feeling pressured, if indeed they wanted to participate. The research coordinator worked with each client to assess their capacity to consent, and cancelled or rescheduled interviews as needed.

Conclusion

We believe our approach to team composition, methods and ethics added considerable richness to the data we collected at Casey House Hospital. Had we involved participants who were in better health, we might not have learned about the opportunities and challenges created by harm reduction in a hospital (Strike et al., 2014). As a team we were committed to having regular meetings and keeping the lines of communication open with the clinical staff and clients who attended the advisory group sessions. This enabled us to solicit advice to make timely and reasoned decisions about how to proceed when we encountered challenges. Overall our team was committed to maximizing the participation of the community from project design to implementation to dissemination while also adhering to high standards of ethical practice and methodological rigor. We were also supported by the HIV Research Ethics Board at the University of Toronto that expedited reviews for amendments and enabled us to quickly obtain approval and implement changes. Although our setting is unique, we believe that many of the issues we discussed are transferrable to other healthcare settings seeking to use CBPR methods to understand and improve the healthcare experience of patients with multi-morbidity and on-going health service utilization.