Ethics and Research Among Persons With Disabilities in Long-Term Care

Analytical Framework

I chose for my research question a critical theoretical framework informed by a Foucauldian analysis (Foucault, 1982; 1995; Foucault & Gordon, 1980; Hughes, 2005; Price & Shildrick, 2002; Yates, 2005) to account for the contextual aspects of long-term care for adults with disabilities (Guba & Lincoln, 1994, 2005; Tremain, 2002) and the ideologies and discourses related to pertinent social policies and government guidelines (Chadwick, 1996) and persons with disabilities in general (Krogh & Johnson, 2006; Michalko, 2002; Thomas, 2007; Waldschmidt, 2005). I also wanted to explore the power issues often affiliated with care settings, consistent with both a critical perspective (Kincheloe & McLaren, 2005; Pothier & Devlin, 2006) and a Foucauldian analysis. The predominant rationale for choosing a critical standpoint, however, was to give “voice to the unheard” (Ball, 1994, p. 4), to hear personal accounts (Wilson & Beresford, 2002), and to visualize “new possibilities” (Morrow & Brown, 1994, p. 11) for long-term care delivery. Gaining insight into people’s experiences was complementary to a Foucauldian analysis and a qualitative methodology (Denzin & Lincoln, 1994), in which I took an inductive approach to thematic data analysis, engaging in constant comparative analysis of data using NVivo 9 (Ryan & Bernard, 2003).

Data Collection Techniques

Participant observation and in-depth interviews were the two main sources of data collection. Participant observation was beneficial for observing the persons with disabilities in their milieus and providing an emic perspective (Field & Morse, 1992). I spent approximately 160 hours of an average of 40 hours per week at four different adult residential facilities, also referred to as community residences, where care was targeted to persons between the ages of 19 and 64 who had significant disabilities and who required Level-3 and/or -4 care. During participant observation, I assisted staff by doing such activities as washing dishes, baking cookies, sweeping, and vacuuming. Similar to Owens (2007, p. 305), however, “I spent more time observing than participating.” I also conducted 40 in-depth interviews with individuals connected in various ways to adult residential care, including persons with disabilities. The good relationships I established with participants in previous research (Morrisey, 2007) facilitated access to the facilities, where I met 29 care recipients, 8 of whom agreed to be interviewed (6 men and 2 women). The average age of these individuals was 53. I chose an in-depth interview style for its effectiveness in research among vulnerable individuals (Booth & Booth, 1994).

Risks

Review ethics boards (REBs) are well aware of the risks associated with conducting face-to-face interviews and participant observation among individuals with developmental disabilities, which involves periods of personal engagement, raising red flags about expectations of friendship and the difficulties associated with leaving the field. Thompson (2002) was right when he argued, “Extended engagements with participants having developmental disabilities pose very real emotional risks for them” (p. 99; see Booth & Booth, 1996; Swain, Heyman, & Gillman, 1998). There are also concerns about the potential to exploit these participants, given their vulnerability because of their significant disabilities and their situation of living in full time in care, fitting Stalker’s (1998) description of persons with “the greatest degree of impairment and the least autonomy” (p. 8). In addition, there are “the dangers of parachuting in and out” (Stalker, p. 17) of their lives, because some might desire ongoing friendships with researchers (Stalker). Despite the risks, however, there are benefits to conducting research among persons with disabilities living in adult residential facilities.

Benefits

The benefits of conducting research among persons with disabilities include helping others to “better understand their reality” (Holmes, Kennedy, & Perron, 2004, pp. 568-569), and to gain insight into “the experiential nature of ‘quality of life’” (Marquis & Jackson, 2000, p. 413). Attaining “service users’ point of view” (Stalker, 1998, p. 16; see also Minkes, Townsely, & Weston, 1995) is also important for discovering the issues that matter to them. Not including the perspective of care recipients “could actually increase their vulnerability because people will remain in ignorance about their circumstances and treatment” (Owens, 2007, p. 307), thereby missing out on data that is important to envisioning ways of improving care (Holmes et al., 2004; Peel & Wilson, 2008). When care recipients are given the opportunity to relate their own experiences of care rather than have others speak on their behalf, they stand to benefit the most from research into long-term care delivery to this particular population. Perhaps the strongest argument for hearing from these individuals, however, is that the perspectives of this particular population are invisible in academic literature.

Rationale for Interviewing Persons With Disabilities

Getting ethics approval to interview persons with disabilities entailed arguing that their involvement was critical to understanding the experiences of people connected to long-term care for such persons with significant disabilities in New Brunswick. I argued that this study would have been incomplete without their input, and that to disregard their voices would have been a direct violation of Tri-Council Policy Statement (CIHR et al., 1998), Article 5.1, wherein researchers are to avoid discriminating against “disadvantaged groups, dependent people” or persons with a developmental disability by not including them in research that relates to them unless there is a valid reason to do so. Not providing a venue for these individuals to share their experiences would have been to victimize them by silencing their voices. The inability to relate incidences of violence or abuse they might be facing and are powerless to resist would render them doubly victimized (Rioux, Cameron, Ticoll, & Bach, 1997, p. 192).

I also argued that their standpoints (Rioux et al., 1997) were crucial to understanding the embodied experience of disability (Pothier & Devlin, 2006); getting “the perspectives of those underrepresented in public policy” (Racino, 1999, pp. 19-20); and discovering positive experiences of care. In addition, I contended that their stories would be valuable for revealing relational power imbalances (Matsuda, 1987), ingrained forms of domination (Parker & Lynn, 2002, p. 11), and enabling me to see the “world through others’ eyes” (Delgado, 1989, p. 2439). Based on Tri-Council Policy Statement (CIHR et al., 1998), Article 2.5(a), that the participation of more vulnerable individuals becomes possible when they are the only ones who can answer the research question, I maintained that only by interviewing care recipients would I be able to discern how they characterized their experiences. Although it was my desire that the persons in care be given the opportunity to share their experiences, I was also aware that I would have to take special efforts to address the potential for harm these individuals might encounter as a result of their participation.

Plans for Protecting the Best Interests of Participants

Attempting to involve persons lacking legal competence also obliged me to explain how I planned to prevent these individuals from being exposed to more than minimal risk (Tri-Council Policy Statement, CIHR et al., 1998) wherein the potential harm to these individuals would be “no greater than those encountered” in everyday life (Article 2.5(c)). This mandate provided the rationale for requesting that third parties well known to care recipients be present at all times during interviews, part of whose role it would be to ensure a secure and safe environment for informants, and to monitor their well-being. Given that the individuals I intended to interview all required adult residential care, having authority figures present at all times was already a normal part of their everyday experience. In addition, I planned to conduct interviews in settings familiar to them, which did not in and of themselves pose risks, to further reduce the measure of risk to these individuals. The questions asked would be entirely topical to their everyday lives, and therefore would pose no greater harm than that of their everyday experience. Finally, participants would not be receiving remuneration for their participation, nor would this research be part of a clinical trial.

To comply with Tri-Council Policy Statement (CIHR et al., 1998), Article 2.6(a), I stated that letters would be sent to the relevant persons in authority over the informants, apprising them of the purpose and nature of the research and the manner in which incompetent persons would be treated. Consent forms signed by these authority figures would give me permission to conduct interviews with the persons in care and to carry out participant observation in the facilities where they lived. I planned to ensure confidentiality by using pseudonyms, not asking questions that could identify them, and not including information that could identify these individuals in publications. I also argued that my own background of having an adult child with disabilities living in long-term care would serve the best interests of clients because my ongoing engagement with care recipients in the facilities where my son has resided over the past 20 years has engendered a keen awareness of how to engage these individuals with dignity and respect, and has given me a high level of comfort interacting with them. Therefore, I posited that my empathy and personal experience would allow participants to be more comfortable during interviews than they might be with a researcher lacking a similar background.

Office of the Public Trustee

As a gesture of transparency on my part, I also sent a letter to the recently established Office of the Public Trustee in New Brunswick to seek advice on how to best proceed in conducting interviews with persons with disabilities in long-term care. I received a written reply, denying me permission to interview persons under their jurisdiction. On contacting the Office of the Public Trustee to reaffirm that my request was for advice rather than consent, an individual speaking on behalf of the Office reiterated their denial, the rationale being twofold: that they were protecting the privacy of clients and that involving clients in this research would be of third-party benefit (to myself as the researcher), with no direct benefit to clients—albeit indirect benefits could ensue at the point of publication if social policies were influenced.

To the first argument, using pseudonyms and excluding information that could identify participants should address privacy concerns. Besides, once persons are placed in care, any notion of privacy is lost, given that many of these individuals face 24-hours-a-day, 7-days-a-week surveillance for all aspects of their lives because of the number of “experts” overseeing and making determinations about their lives (Carlson, 2005; Foucault, 1995; Hughes, 2005; Krogh & Johnson, 2006; Tremain, 2005). This then begs the question, protecting their privacy from whom? If it is the general public, then we do a disservice to persons with disabilities living in long-term care settings by keeping them invisible.

To the second argument, participants would directly benefit from being able to talk about issues that matter to them, thereby expressing some independence. Wadenstein and Ahlstrom (2009) argued, “Being able to express one’s own opinions and having others show consideration, are prerequisites for achieving independence” (p. 760). Allowing persons with disabilities to participate in research about the care they receive would show the same respect typically granted to persons without disabilities. Although the individual speaking on behalf of the Office of the Public Trustee added that the denial of consent was informed by policies currently under development and the Infirm Persons Act (1973), on perusing this document, as well as the Mental Health Act (1994), the Family Services Act (1983), the Human Rights Act (1985), Public Trustee Act (2008), and the 2008-2009 Annual Report (Department of Justice and Consumer Affairs, 2009), I found little to support disallowing individuals in adult residential care from talking about their experiences so as to protect their privacy.

Perhaps most troubling was that this denial was a basic dismissal of the REB process, suggesting that the rigor of this process and the procedures involved in planning research protocols to meet Tri-Council Policy Statement (CIHR et al., 1998) standards might not be well known in the larger community. Nevertheless, this denial made it necessary for me to ask the directors of each agency if any of the persons living at the facilities were under the jurisdiction of the Office of the Public Trustee, to ensure that none of them took part in interviews during data collection. The fact that the Office of the Public Trustee in New Brunswick is a newer institution might have influenced the inflexibility of their response. Even so, as Yan and Munir (2004) pointed out, “Outright violations involving vulnerable participants are much less likely to occur in today’s research . . . significant advances in investigator training and the media presence in exposing questionable research practices to a conscientious public make adverse outcomes less likely” (p. 45).

Gaining Informed Consent

Gaining informed consent from vulnerable populations, and particularly persons with developmental disabilities in adult residential care, can be problematic, for “there is nothing ‘ordinary’ about informed consent, both conceptually and procedurally—especially for people who have developmental disabilities” (Thompson, 2002, p. 96). Competence refers not only to the ability of potential participants to understand every aspect of both what they can expect and what might occur during the research process, but also the capacity to decide if they want to participate based on that information (Homan, 1991). Having one’s full intellectual capacities, however, does not automatically mean that a participant will have a complete understanding of everything that will take place, notwithstanding the efforts of researchers to provide such information. Nevertheless, the fact remains that people with developmental disabilities requiring a high level of care typically lack legal competence. Therefore, given their incompetence to provide informed consent, I had to take certain measures to address this reality.

I sought permission for their participation from the directors of the agencies of the adult residential facilities, as substitute decision makers, to interview and conduct participant observation among care recipients, in accordance with Tri-Council Policy Statement (CIHR et al., 1998) requirements. Letters were sent to all directors informing them of the purpose and nature of the research, my interest in conducting research with the persons in care, and the manner in which these individuals would be treated. The directors of these agencies signed forms granting me permission to conduct interviews with care recipients and other staff members, and to carry out participant observation at the facilities. On gaining consent from the directors of the agencies, I then sought the assistance of staff members at each facility in asking care recipients if they were interested in being interviewed, which meant I had to depend on them to help recruit participants (Moore & Savage, 2002).

Gatekeeping

Having to negotiate “access to the client group” (Peel & Wilson, 2008, p. 416) revealed the importance of staff members as gatekeepers when it came to conducting interviews with the people living in the various community residences. The directors of the agencies, acting as initial gatekeepers, granted me access to their facilities to conduct participant observation; to ask questions; to carry out interviews with staff and, for the purposes of this article, persons with disabilities in their care. The second tier of gatekeepers were staff members at the facilities on whom I relied to approach care recipients and inquire if they were interested in being interviewed, and to ensure that potential participants understood my explanation of why I wanted to talk to them; how the interviews would proceed; and that they could withdraw at any point during the process. I also needed staff members to remain with clients while being interviewed. The staff members I approached did not question clients’ abilities to understand the process, unlike in other research (see Davis, Watson, & Cunningham-Burley, 1999).

Although these staff members did not have the power to deny me permission to interview care recipients, because directors had already given me that authority, it was definitely “in their power to block access” (Stalker, 1998, p. 8) to clients. They could have declined to ask the persons in care if they wanted to participate in the research or refused to remain with them during interviews, either of which would have meant I could not proceed. The power of gatekeepers was also explicit in their ability to make determinations about which individuals might be more eligible to participate in the research than others (Wadenstein & Ahlstrom, 2009). “Eligible” care recipients who did agree to participate appeared to do so because they wanted to, not because they were pressured or coerced in any way (Moore & Savage, 2002), which was important for following the mandate of Tri-Council Policy Statement (CIHR et al., 1998), Article 2.2, and avoiding acquiescence.

Acquiescence

Addressing acquiescence is an important ethical consideration when conducting research among persons with developmental disabilities (Flynn, 1986; Lindsey, 1994), given its reported prevalence among persons with “lower education and lower intelligence” (Matikka & Vesala, 1997, pp. 75-76). Acquiescence occurs when incompetent individuals, out of “a strong motive to give socially desirable answers” (Sigelman et al., 1983, p. 159), or to “please” researchers (Biklen & Moseley, 1988, p. 159), “go along with the wishes, commands, or requests of the (often) able-bodied researcher or practitioner” (Thompson, 2002, p. 103). High rates of acquiescence are well established in quantitative survey research, particularly for yes or no type questions, but less the case for open-ended style questions (Sigelman, Budd, Spenhel, & Schoenrock, 1981; Sigelman, Budd, Winer, & Martin, 1982; Sigelman et al., 1983). Consequently, my choice of an in-depth questioning approach was more amenable, enabling care recipients not only to talk about their experiences of care, but also about whatever else they wanted to.

Waiting until at least the third day of participant observations was also beneficial for helping to guard against acquiescence. By that time, I had a sense of the individuals who might be interested in being interviewed because they had already engaged me in numerous conversations on previous days. I also made it clear to staff that care recipients were not to be coerced in any manner to participate. To that end, I was careful to be present when each individual was approached to see if he or she was interested in participating because I wanted to be certain that participants were able to exercise their freedom of choice. As Orb, Eisenhauer, and Wynaden (2001) put it, “The desire to participate in a study depends on participant’s willingness to share his or her experience” (p. 93). It was also my opinion that if participants freely offered to talk about their experiences without duress, the credibility of the data would be enhanced. Therefore, although keenly interested in getting the perspectives of the persons with disabilities in care, I was not willing to compromise their freedom of choice or sense of well-being to do so.

Staff members, because of their long-term association with care recipients, were beneficial to ensuring that participants understood my explanation of the interview process and that they had complete freedom to withdraw from participating at any point. This freedom to decline from participating was substantiated when one individual stated he was not interested, shrugged, and simply walked away. He later approached me, after spending some time alone in his room, to see if I would tape him singing one of his favorite songs, which I did, because I wanted participants to be “in the driving seat” (Owens, 2007, p. 303) during interviews. Despite the steps I took to reduce acquiescence, to get permission from the directors of the agencies to conduct participant observation and interviews at the residences, and to have staff assist with recruiting participants, I also set out to gain the assent of the persons with disabilities themselves to ensure I was following appropriate ethical protocol.

Gaining Assent

Seeking the assent of the persons in care was consistent with Tri-Council Policy Statement (CIHR et al., 1998), Article 2.7, regarding incompetent persons, wherein researchers are to “seek to ascertain the wishes of the individual concerning participation” and where “dissent” precludes their participation, in situations where consent for their participation has already been attained from authorized representatives acting on their behalf. In addition, gaining consent from authorized representatives did not automatically mean that the persons in care would want to participate (Stalker, 1998, p. 8). Despite being aware that proxy consent by care providers could help to protect me from potential litigation issues (Dewing, 2007; Vass et al., 2003), I was still not satisfied because I was not interested in perpetuating the notion of care recipients as “eternal children” (Stalker, 1998, p. 9). By not including care recipients in at least some aspect of the informed-consent process I would have been guilty of “reinscribing . . . disabling regimes” (Thompson, 2002, p. 96). Seeking the assent of the persons in care without first establishing that these individuals had the capacity to provide that assent, however, would also have been inappropriate.

To determine the capacity of these individuals to freely provide assent, I focused on their ability to engage in everyday decision making (Thompson, 2002). Paying attention to this aspect of their lives was important because people living in formal care settings sometimes take on a “learned helplessness” (Lindsey, 1994, p. 161), which can be typical of individuals living under constant supervision where choices are conceded to “caregivers and/or other able-bodied persons” (Thompson, 2002, p. 102), although less so in smaller community-type residences (Ellis, 1992). Lloyd, Gatherer, and Kalsy (2006) pointed out, however, that “respect must be given to those areas of functioning where individuals retain competency and . . . the ability to consent should not be simply assumed” (p. 1399). I also focused on indicators of the general level of “well-being” of clients, including facial expressions (Dewing, 2007), to assist in determining their capacity to assent. Conducting participant observation proved beneficial for gaining a sense of their level of both well-being and engagement in everyday decision making.

By the third day at each facility, I had a general impression of what well-being looked like for each client, as well as insight into the degree to which these individuals took advantage of opportunities to make decisions, which was informative when it came to determining their potential to express a measure of control in their lives (Thompson, 2002, p. 102). Although both the opportunities to make decisions and the areas of their lives where choices were possible varied between community residences, the most common decisions were related to clothing preferences; participation in the particular activities offered to them; and to a lesser extent, food options during snack and meal times, revealing that care recipients did experience a measure of inner control in their lives. Not surprisingly, as the severity of disabilities for persons increased, opportunities for and engagement in decision making decreased. Increased levels of disability, however, did not appear to be associated with a proportional decrease in the well-being of the persons in care. On the contrary, some of the individuals appeared to be quite content and happy, despite displaying little engagement in decision making.

Nevertheless, the decision making exercised by the persons I ended up interviewing led me to conclude that the individuals who did participate did so freely, their capacity to assent embedded in the knowledge that assent was provided during a state of well-being in which they exhibited some control over their lives. In an additional attempt to ensure that the persons in care participated of their own free will, as mandated in Tri-Council Policy Statement (CIHR et al., 1998), Article 2.7, I presented each participant with a brief assent form purposely devised to take into account their “perceived level of comprehension” (Stalker, 1998, p. 8). I then proceeded to explain these assent forms to them in the presence of staff members. Only once the individuals had agreed to be interviewed, and staff and I had determined that the care recipients understood what they were agreeing to, did I ask them to sign the assent forms, which they did, either by printing or writing their name, or by making a mark to symbolize their assent.

In addition, at various points during the interviews, I stopped to reconfirm each participant’s wish to continue, as I considered their initial agreement to participate “insufficient” (Sigelman et al., 1981, p. 57), given that assent is not a one-time event (Dewing, 2007; Gilbert, 2004; Knox, Mok, & Parmenter, 2000; Lloyd et al., 2006; Orb et al., 2001; Thompson, 2002; Tymchuk, 1997; Usher & Arthur, 1998). Having staff members well known to participants present at all times during interviews, acting as third parties or “validators” (Dewing, 2007, p. 20), was also important for recognizing signs of “discomfort or stress” (Lloyd et al., 2006, p. 1398) in clients should they occur. To that end, I argue that I took an ethically responsible approach because the persons in care who did agree to be interviewed had the capacity to provide assent, which they did of their own volition and without coercion (Holmes et al., 2004). Nevertheless, other ethical dilemmas remained.

The Context of Interviews

Another problem was where to conduct interviews with care recipients. I set out initially to interview these individuals at locations familiar to them other than the facilities where they lived, and with individuals not necessarily affiliated with these facilities who would stay with them during interviews, but this proved impossible. First, I had difficulty coming up with locations different from the adult residences where they lived. The only other locations where clients spent significant amounts of time were vocational centers/workshops, or day centers, where they participated in scheduled activities, but these settings do not lend themselves to private one-on-one interviews. A considerable number of individuals generally frequent these centers, and the greater the number of adherents to these programs, the less freedom the persons who work there have to sit in on interviews because of the time involved in providing leadership, monitoring activities, and mediating relationship issues between those able to take advantage of these programs.

In addition, I was wrong in assuming that I could conduct interviews in the clients’ homes of origin, with parents or family members who would remain with their loved ones during interviews. I had difficulty establishing contact with relatives for several reasons. The research was carried out in areas of the province other than where I live, which meant that I did not know the names of family members I could contact. Increased privacy regulations exacerbated this situation. In addition, according to staff who worked at the facilities, fewer relatives maintained contact with their adult loved ones once placed in care than did relatives of children who entered the long-term care system. One reason staff members believed this to be the case was that some of the parents of those in care were in their senior years and facing health issues and other life complications. Other relatives remained fearful of the aggressive behaviors their adult children with disabilities displayed from time to time. Still other family members had passed away.

Another explanation for the limited contact is that before living in the adult residences, some care recipients lived in one of the larger institutions in the province at a time when parents and family members were discouraged from visiting because it might upset their loved ones. I eventually did find a number of parents and family members to interview, but they were not relatives of the persons with disabilities who assented to be interviewed. Conducting interviews at the adult community residences where clients lived, with staff members acting as third parties, turned out to be the most familiar context for clients. The particular staff members who agreed to serve as third parties were individuals with whom the care recipients appeared to be most comfortable, probably because of their long-term relationships with them. Interviews with interested clients took place either in their bedrooms or at kitchen tables, with one staff person present, and at opportune times when other staff members were busy with other clients in different areas of the residences. Although the involvement of staff as third parties was beneficial in a number of ways, it did create problems related to anonymity and confidentiality for participants.

Confidentiality

The steps I took to ensure confidentiality for participants with disabilities, using pseudonyms and not asking questions or using information that could identify participants, were insufficient for ensuring complete anonymity and confidentiality for these individuals. Although the involvement of staff enabled the inclusion of persons with disabilities (Goodley, 1996), protected their best interests by ensuring their well-being (Dewing, 2007) during interviews, protected me from potential accusations of inappropriate activity toward clients, and added validity to the data by their confirmation of participants’ responses, their presence greatly compromised anonymity and confidentiality for participants. This was a problem, because confidentiality is a key tenet of ethical research. Moreover, several staff members tried to exercise influence during interviews. In one case an individual made suggestions about other potential questions I could ask, and in another, the person with disabilities was encouraged to reframe a negative description of a particular staff member.

It is crucial for both persons with disabilities and researchers that third parties remain as a constant presence during all aspects of research that involve these vulnerable individuals, but it should be duly noted that the trade-off is weakened anonymity and confidentiality. It is no surprise that internal confidentiality (Phtiaka, 1994) is difficult and in some cases impossible when research is carried out in institutional-type settings (Snyder, 2002; Stratton, 2002; Weinberg, 2002). This awareness obliges researchers to state at the outset their “limitations in ensuring the anonymity of research participants and the confidentiality of their comments” (Snyder, p. 78). Van den Hoonaard’s (2002) contention that “REB’s insistence on confidentiality is legally wasted in the social sciences” (p. 179) is particularly appropriate given this reality.

Incomplete Disclosure

Prior to data collection, I had to decide whether to reveal my personal background of having a son with significant disabilities living in a long-term care facility. I opted to refrain from revealing this information unless asked about my personal background. I reasoned that such knowledge might engender questions about the ulterior motives behind my research focus. More specifically, the wrong assumption could have been made that I was using the data collection process to discover problems associated with care or to scout out alternative residences for my son, which could have resulted in staff being less forthcoming in the information they shared. As it turned out, various staff members did express opinions about parents and, more often mothers, that I am certain would not have been shared were they aware of my background. For instance, one staff member expressed his belief that if one looked carefully at the parents of care recipients, one would discover that at least one of them had borderline mental issues.

Another reason for incomplete disclosure was that I did not want to be treated differently than another researcher without a similar background. A final rationale for not sharing this information was the concern that I might be less successful in establishing rapport with individuals at the facilities, which would have implications for the quality of the data I would be able to elicit. My goal was to observe routine daily activities at the facilities, the interactions between staff members and the persons in care, and between staff members themselves as they would naturally occur (as much as it is possible to be natural when someone is conducting participant observation). Nevertheless, this incomplete disclosure created within me a sense of unease as to whether this was a breach of acceptable ethical conduct, and more typical of covert research practices. Cognizant of other studies denigrated for their covert approaches (Clark, 1996; Humphreys, 1970; Milgram, 1974), I eventually relied on Lugosi (2006), who explained covert research as that which obscures motives and purposely deceives informants, where “researchers disguise their identities” (p. 544).

In the case of motives, all participants with the exception of care recipients were made aware from the outset that I would be conducting interviews and participant observation to get at the experiences of people connected in various ways to the care that takes place in adult residential care, thereby fully disclosing my intentions to collect data while in their midst (Patton, 1990). On the first day at each facility, I was introduced to care recipients by staff members as one who was there to conduct research. Several days into participant observation, a number of the persons in care asked me again why I was there, at which time I reiterated that I was doing research. When staff members approached some of the persons with disabilities to see if they were interested in being interviewed, the purpose of these interviews and my time among them was relayed yet again. Therefore, no one was deceived as to why I was in their midst, nor did I make any attempts to disguise my role as a researcher.

Taking a covert approach would have been impossible anyway, because I had to explain my research intentions to the directors of the various agencies to get permission to collect data at the facilities. At every facility, by the second or third day of participant observation, staff members or persons in care asked specific questions about my personal background, at which time I was completely honest, sharing that I have a son living in a different adult community residence in another area of the province. Rather than this being perceived as a problem, several staff members shared with me that they could then understand why I appeared to be so comfortable with clients, which had generally not been the case with other visitors in the past. Consequently, I argue that sharing my personal background information only when asked turned out to be the right approach for this particular study because I was able to establish rapport with participants and elicit more credible data without transgressing acceptable ethical research practices.

The Risk Factor

I took specific precautions to keep from exposing the persons in care to more than minimal risk. Early in the research, I clarified “the boundaries of the researcher-researched relationship” (Thompson, 2002, pp. 104-105), cognizant that withdrawing from the field after conducting research among persons with disabilities can be fraught with ethical implications. I told care recipients at the outset that I would be with them for only 5 days and would return to my home in a different area of the province, thereby eliminating false expectations as to the length of my stay. Consequently, these individuals knew from the start that an extended friendship with me would be impossible because of the limited amount of time I had with them, and because I lived a significant distance from the facilities. By the fourth day at each facility, I reminded clients that the following day would be my last day there, to keep them briefed of my status and to prepare them for my departure.

Nevertheless, some expressions of attachment did begin to surface. For instance, one of the women in care gave me a copy of a poem she had written for me, and a man who lived at a different residence made up some questions to ask me once I finished interviewing him, which I was careful to answer. Another woman with disabilities at a third facility indicated that she had something she wanted to talk to me about. She shared with me that she and the other care recipients had been talking, and wondered if I might come and work for them full time. I was aware that persons with developmental disabilities in long-term care tend to have “fewer friends and more restricted social contacts” (Stalker, 1998, p. 10), so I asked clients about the status of their friendships during interviews. Some described fellow care recipients as friends, one citing his roommates as the reason he liked where he lived. Three spoke fondly of fellow care recipients as their “girlfriend,” “boyfriend,” or “sweet.” Conducting participant observation allowed me to observe these “special” friendships.

Other persons with disabilities described staff members as friends, which was obvious during the time spent in participant observation because of the laughter and camaraderie they shared. Still others spoke of friends at the day centers or downtown in addition to their friends where they lived. Although clients, in general, did appear to have fewer friends and more restricted social contact, particularly in the case of persons with more significant disabilities, they still experienced active and ongoing friendships. The smaller spaces afforded by adult community residences, which are operated much like typical family dwellings, appeared to facilitate closer and more familial relationships between staff and clients, and between clients themselves, than generally possible in the larger institutions where some of these individuals had formerly lived.

Time spent at day centers as well as locations such as coffee shops and malls in the larger community also provided opportunities for the persons in care to build friendships with individuals other than with just staff and fellow clients. Based on this knowledge, the accounts of the persons in care I interviewed, and my personal observations at the facilities, I concluded that my leaving the field would have a limited detrimental impact because of the friendships these individuals already enjoyed. I failed to anticipate, however, the difficulty I would experience personally withdrawing from the field. I continue to think about these folks and wonder how they are doing. I am intentionally revealing this aspect of the research process for two specific reasons. First, it is important to acknowledge that emotion work can be a very real part of the experience for qualitative researchers who become involved in fieldwork (Gilbert, 2001; Holland, 2007; Hubbard, Backett-Milburn, & Kemmer, 2001). Second, it highlights the value of a reflexive, transparent approach to analysis.