| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||
Correspondence: Address correspondence to Julie Locher, PhD, University of Alabama at Birmingham, CH19 219, 1530 3rd Ave. S., Birmingham, AL, 35294. E-mail: JLocher{at}uab.edu
 |
Abstract |
|---|
 TOP Abstract Methodological IssuesPractical Strategies for...References |
|---|
Key Words: Research ethics • Homebound elders • Therapeutic misconception • Role conflict
In writing this article, we draw upon more than a decade of both descriptive and intervention research focusing exclusively on the homebound older population in which the authors have collaborated (Locher, Burgio, Yoels, & Ritchie, 1997; Locher, Ritchie, Baker, Bodner, & Allman, 2005; Locher, Robinson, Roth, Ritchie, & Burgio, 2005; Ritchie et al., 1997; Ritchie & Dennis, 1999). Examples from our studies will be used to illustrate or support points to be made.
Therapeutic Misconception and Research Role Conflict
Two ethical issues are particularly salient in studying homebound older adults—therapeutic misconception and role conflict for the researcher. Both involve conflicting roles of the researcher but from different perspectives. In fact, they are essentially different sides of the same coin—therapeutic misconception from the perspective of the study participant and researcher role conflict from the perspective of the researcher. Therapeutic misconception has been well described in the literature and refers to a situation wherein research participants mistakenly believe they are to receive some form of treatment or care (Lidz & Applebaum, 2002).
Therapeutic misconception is a common problem in clinical trials (Lidz & Applebaum, 2002), and researchers disagree over the extent to which therapeutic misconception may be regarded as harmful for study participants. For example, the Declaration of Helsinki (World Medical Association [WMA], 2004) frequently uses such terms as "therapeutic" and "treatment," which encourages the misconception of research as therapeutic. Others have been critical of such language and the misperceptions held by study participants that such language encourages (Annas, 1996). While therapeutic misconception is not unique to homebound study participants for a number of reasons, including how participants are recruited, the homebound status of participants, and the home setting as the interview site, the issue is especially salient in this population.
Researcher role conflict occurs when the researcher is expected to perform two roles with different and sometimes opposing sets of expectations—that of the researcher and that of professional (e.g., physician, sociologist, nutritionist) and, in some instances, even that of the ordinary citizen. Role conflict is an important sociological construct that is rooted in the structural account of roles outlined by Parsons (1951), most notably in his notion of the sick role, and also by Merton (1949) in his construct of role-sets. Each person in a society has a number of statuses (e.g., researcher, physician, citizen) that are associated with corresponding roles to fulfill. In turn, each role has an associated set of individuals or organizations with whom the individual is expected to interact based upon shared expectations for mutual rights and obligations. These expectations are based upon values that may emanate from a variety of sources, including professional (e.g., medical) and organizational (e.g., university). When these role-sets are in disaccord, role conflict occurs.
For example, normative expectations of the researcher presume that he or she engages in an empirical investigation that tests some hypothesis that will ultimately lead to generalizable knowledge. As a scientist, the researcher is expected to remain objective and to not bias study outcomes. In contrast, the obligations of the physician researcher as described in the Declaration of Helsinki (WMA, 2004) state that "The Declaration of Geneva of the World Medical Association binds the physician with the words, ‘The health of my patient will be my first consideration.’" As Annas (1996) observed in his critique of the Declaration, there is a muddling of research and treatment in such accounts. On the one hand, there is the duty to respect autonomy, and on the other there is a duty to protect from harm. Indeed, the Belmont Report (Office of Human Subjects Research, 1979) specifically states that respect for persons (autonomy) and beneficence (do no harm) are the first two ethical principles guiding research involving human subjects. This matter may become complicated, though, when threats to health are not a consequence of the research itself or when the research is observational.
Researcher role conflict is especially likely to occur in the home environment, when in the course of investigation the researcher becomes aware of something that is immediately or potentially harmful to the study participant. On the one hand there is the obligation of the researcher to remain objective and not do anything that could interfere with the data or with the results of the study, while on the other hand there may be an obligation as a health care provider or simply as a citizen to act in response to perceived abuse, neglect, or exploitation of the study participant. Under such circumstances, there may be codified professional and legal requirements to deal with the situation. For example, the American Sociological Association's Code of Ethics (1997) observes: "Sociologists may confront unanticipated circumstances where they become aware of information that is clearly health- or life-threatening to research participants, students, employees, clients, or others. In these cases, sociologists balance the importance of guarantees of confidentiality with other principles in this Code of Ethics, standards of conduct, and applicable law."
|
Methodological Issues |
|---|
|
TOPAbstractMethodological IssuesPractical Strategies for...References |
|---|
Recruitment of Participants
It is not unusual for researchers who exclusively study homebound older adults to rely on service providers such as home health agencies for recruitment of participants. Under such circumstances, service providers typically ask clients if they are agreeable to being referred to the study. This linkage can unintentionally extend the therapeutic orientation of the client toward the service provider outward to the researcher. First, individuals may be more inclined to participate in a study if it is recommended to them by their health care provider, in whom they have instilled trust (Stevens & Pletsch, 2002). Those who are homebound may be less likely to carefully weigh the costs and benefits of participating in a study because they have been unduly influenced by the recommendation of a nurse on whom they depend (Johnson, 1997). Persons may want to please their health care provider by agreeing to participate, or they may believe they will get better care from their health care provider by participating. Conversely, persons may be concerned that their care may be threatened if they refuse to participate (Steinke, 2004). Others may agree to participate because they are lonely and lack social contact. Lastly, home health patients who are socially isolated may, more than other patients, see research as a means to get easier access to health care.
Particular care must be made in recruiting homebound participants to ensure that coercion is not used in recruiting or in obtaining consent as voluntary informed participation by research subjects is a cornerstone of contemporary research ethics (Office of Human Subjects Research, 1979). Several authors have noted this is especially challenging when patients are in a compromised health state and assume a submissive role (Bosk, 2002; Casarett, 2000; Emanuel, Wendler, & Grady, 2000; Steinke, 2004) that includes the expectation to follow the recommendations of their health care providers (Parsons, 1951).
Beyond linking and extending the expectations directed toward service providers outward to researchers, potential participants may actually confuse research participation with service provision. When an individual is receiving home health services, there are often many care providers who come into the home to perform different tasks related to the patient's care. The home health nurse is typically the gatekeeper and the one who provides an explanation to the patient regarding who will be coming into the home. While most of our studies do not involve clinical trials, there is often a mistaken attribution that the research interviewers are there to provide some sort of therapy or other service. For example, participants have asked us for help in obtaining such things as a power lift chair, diapers, and even food. Because of this, it is especially important that the agency recruiter and the investigator make clear during the consent process that no treatment will be given. On the other hand, there may be times when it is reasonable to share information regarding available resources that may be available to participants without jeopardizing the goals of the research. For example, in the case of the man who requested diapers, we were able to refer him to an agency that provided diapers free of charge.
‘Homebound’ Status of Participants
Another issue related to participant recruitment is the particularly vulnerable status of the homebound population. Many researchers define homebound status using the same definition that Medicare uses to establish eligibility criteria for receipt of home health services. According to Medicare's criteria, a person is considered "homebound, or unable to leave the home unassisted ... (if) leaving home takes considerable and taxing effort." (Centers for Medicare & Medicaid Services, 2004, p. 11) Additionally, persons who are receiving home health services are often either experiencing an acute illness or an exacerbation of a chronic condition. This means that they are both sicker than the general population of older adults and more socially isolated. Both of these may influence a person's understanding or decision-making capacity and willingness to participate in a research study for a number of reasons, which include their capacity to fully understand the consent process because of the illness and their vulnerability to loneliness.
The Home as an Interview Site
Research conducted in the home allows the investigator to be privy to much information that he or she would not have access to if the research was conducted in another setting or over the phone. As a consequence, the researcher may be more likely to be a witness to something that is threatening to the participant's health or life, including elder abuse. This is not an unusual occurrence because homebound older adults may be especially vulnerable to elder abuse due to their increased social isolation, physical impairment, cognitive impairment, and dependence on others for care (Lachs & Pillemer, 2004; National Center on Elder Abuse, 1998). Elder abuse is a broad term that includes physical abuse, sexual abuse, emotional or psychological abuse, financial or material exploitation, neglect by a caretaker, and self-neglect (Administration on Aging, 2005). The fact that researchers have access to the home setting intensifies researcher role conflict in several ways.
The first of these involves the researcher's obligation to report elder abuse. The role of the researcher is to remain objective and to not intervene in participants' lives but observe, document, and report findings in aggregate. Intervention alters the natural course of events that research is intended to describe, and consequently may be considered flawed science. On the other hand, there are legal obligations to report incidents of abuse and neglect. In research ethics there is an obligation to protect research participants from risks generated by the study; however, there is no absolute requirement to protect participants from risks that might be incurred unrelated to the research. At the same time, researchers should seek to maximize the benefits accruing to participants involved in the study, and reporting and attempting to ameliorate risks is an important way that participants may benefit from their decision to provide data for the study.
There are several examples of this from our research. In perhaps the most disturbing instance of elder abuse we encountered, one of the researchers became aware during the interview that an older female participant who was bed-bound had been sexually assaulted by one of her caregivers. The researcher immediately contacted the local adult protective service agency. The agency investigated the matter (including formally interviewing the researcher), a formal protective restraining order was issued against the assailant, and a criminal investigation ensued. In another situation, we interviewed a man who had especially severe pressure sores that were not healing, and we discovered that he was not consuming any protein in his diet. We promptly reported this to the home health nurse responsible for his care. Clearly, we were aware that any intervention could influence his eating habits as well as future health outcomes, which we were studying.
The value placed on research and its outcomes must be weighed against the value placed on protecting others from harm—even if the harm is not caused by the research. Additionally, the value placed on participant autonomy must be taken into account as well. For example, in one instance when we interviewed a blind woman, the researcher noted that, unbeknownst to the participant, she had broken windows in her home and had been brushing her teeth with hydrocortisone cream. The investigator informed her of both of these concerns and asked her if she was aware of services for the blind. The woman told her that she did not want to get outsiders involved because it would upset her daughter, who, along with the daughter's husband, was one of her primary caretakers. The woman also expressed a desire to remain living in her own home. In this case, the investigator had to consider the autonomy of the participant and her wishes to not have outside intervention.
Respect for participant autonomy requires that action to ameliorate risks be taken only at the participant's request. However, if the participant is afraid to seek help for fear of retribution, then one must consider whether the participant is truly autonomous or competent to make the decision. Additionally, the risks of the harm observed may need to be balanced against the risks that may result if an intervention is made. In the case of the blind woman, one possible future scenario if the researchers had intervened was nursing home placement, which was contrary to the woman's desire to remain at home. Therefore, the researchers made the decision not to intervene.
Another matter that presents an ethical challenge to research investigators involves determining what exactly constitutes abuse. One particularly vexing issue that has come up repeatedly in our research involves possible financial exploitation. For example, we discovered on several occasions during data gathering that participants had granted power of attorney to their children, including the handling of their financial matters. These participants mentioned that their children would not give them money to donate to their church or had used their money in ways that the participant thought was inappropriate. While some of these activities may be immoral or thoughtless, they are not illegal. Second, the ethics of intervening may interfere with other ethics held in high regard in our society, particularly that of autonomy. The researchers' training and orientation may incline them to avoid altering the situation or moving beyond the research process to effect some change for the purpose of bettering the participant's situation. Additionally, because the researcher is in a nontherapeutic relationship with the participant and has a predisposition toward not intervening, he or she may set the threshold for determining abuse at a higher level than others who come into contact with the participant.
|
Practical Strategies for Conducting Studies With Homebound Older Adults |
|---|
|
TOPAbstractMethodological IssuesPractical Strategies for...References |
|---|
In regard to recruitment of participants from service providers, we recommend that the study team develop a close relationship with those providers with direct patient contact who will be making referrals. A close relationship entails one in which regular (e.g., weekly) face-to-face interaction takes place between those service providers making the referrals and the study staff receiving the referrals. Service providers must understand what the study is about in order to convey such information accurately to potential participants. This is enhanced by regular interaction.
Additionally, we employ a two-step consent process to enhance informed consent. In the first step, the home health nurse obtains written informed consent from a potential participant prior to making a referral. During this process, the home health nurse briefly describes the study to participants and asks them if they are willing to be referred to the study. The patient is not consenting to participate in the study—only to having his or her name referred. This step helps to ensure that the study description provided to potential participants by service providers distinguishes between treatment and study participation. This is in contrast to what we believe is the unacceptable practice of passive consent used in some studies in which patients are told if they do not contact the nurse in a given period of time that they will be referred to the project.
In the second step of our informed consent process, which takes place in the home, the researcher obtains consent from the patient who was referred to the study to participate in the research. The consent form is the means by which autonomy is respected and satisfied in the research process. Any action that could come to pass should be described in this document. In most cases individuals are consenting to the research and entry into their home and not to action beyond what is described therein. Care must be taken to describe the course of action that will be taken if threats to health or life are observed during the course of investigation. For example, in our consent forms, we clearly inform participants that if we observe any circumstance that poses a threat to their health or well-being, we will inform either their primary physician or home health nurse. We further inform potential participants that if we observe any sign of abuse or neglect we will notify adult protective services. This up-front approach was a specific requirement of our internal review board (IRB) and one that other IRBs might consider adopting as well.
To the extent possible, it is wise for researchers to decide in advance what constitutes a threat or abuse. In our studies, we establish guidelines for when intervention is required. For example, if a person is consuming fewer than 800 calories per day or if we observe a 5% or greater loss in body weight from one interview to the next, we will notify the patient's physician or home health nurse. Obviously, not all possible threats can be anticipated in advance, and many fall within a very gray area. To deal with this, we hold biweekly research team meetings at which we discuss any problems that may have been encountered with participants. When there is disagreement among researchers regarding what constitutes a problem, our approach is to consider differing perspectives and arrive at a group consensus. When we do observe a threatening situation, we report these to our IRB as adverse events.
Last, in regard to the consent process, researchers must be careful to clarify the purpose of the study in order to especially address participants' desires for therapeutic outcome, social contact, or practical help. Reading the consent form to participants, asking them if they have any questions, and asking them to describe their understanding of the study are ways to ensure that "informed consent" is taking place.
Finally, if the researcher did intervene in some way in participants' lives, he or she has an obligation to the scientific community to correct for the introduction of any potential bias. One way this can be accomplished is to statistically adjust for any bias that could affect study outcomes. Another approach is to describe in the methods section of a paper any interventions that occurred.
|
Footnotes |
|---|
1 Department of Medicine, Division of Gerontology and Geriatric Medicine, University of Alabama at Birmingham.
2 Center for Aging, University of Alabama at Birmingham.
3 Department of Sociology, University of Alabama at Birmingham.
4 Department of Health Care Organization and Policy, University of Alabama at Birmingham.
5 Lister Hill Center for Health Policy, University of Alabama at Birmingham.
6 Birmingham VA Geriatric Research, Education, and Clinical Center (GRECC) and Birmingham VA Deep South Center on Effectiveness, Alabama.
Decision Editor: Linda S. Noelker, PhD
Received for publication June 28, 2005. Accepted for publication December 12, 2005.
|
References |
|---|
|
TOPAbstractMethodological IssuesPractical Strategies for...References |
|---|
| ||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
