INTRODUCTION
Bruera (1994) describes the history of
palliative care and hospice programs, which began outside mainstream
medicine in the 1960s in the United Kingdom before expanding into Europe
and North America in the 1970s and 1980s. The emphasis of the palliative
and hospice literature has traditionally been on direct patient and family
care, providing a counterpoint to overly scientific and technological
aspects of medicine including research. However, research has recently
begun to receive more attention. Bruera (1994)
states that these patients usually experience “devastating physical
and psychosocial symptoms” and that families often suffer from
“severe psychosocial distress” (p. 7). He also cites studies
indicating that “80% of patients develop severe cognitive failure
before death,” (p. 8), begging the question as to whether informed
consent can be given by these patients. For informed consent to be
possible, Dunn and Chadwick (2001) explain that
the participant must adequately comprehend the information and participate
voluntarily, without any coercion or undue influence. Provider sensitivity
to informed consent among the terminally ill appears to be very great and
may greatly influence whether hospice staff would support the idea of
research with patients and family members.
In addition, de Raeve (1994) cites
terminally ill patients' vulnerability, dependency, and compromised
autonomy and makes the case that no research can be justified with dying
people. Kristjanson et al. (1994) also found
reason for caution after examining 55 empirical studies involving
palliative care patients published between 1986 and 1993. Although
supporting the need for research and the rights of individuals to
participate, Kristjanson et al. expressed concerns about the possibility
that unethical research is being done, especially considering the
vulnerability of this population as well as the difficulty of obtaining
truly informed consent.
Mount et al. (1995) take strong exception to
de Raeve's statements, finding the stand that the terminally ill
should not be asked/allowed to participate in research
“paternalistic, devaluing and disrespectful” of the patients
(p. 165). Arguing that palliative care research is not simply
curiosity-based experimentation, they make a strong case for the critical
role of research in developing proven therapies that can provide relief
from pain and the many other symptoms with which palliative care patients
suffer greatly (Mount et al., 1995). Other
researchers, too, have noted the inadequacy of research on the many
complexities associated with pain management in various clinical
populations. Vega-Stromberg et al. (2002) cite
data indicating that “pain of all types in all age groups is under
treated despite the existence of effective … treatments” (p.
15). In addition, the question about participation in research can be
considered another area where there is little research about the role of
choice among the terminally ill (Drought & Koenig,
2002). There is a great need for research to better understand the
issues surrounding decision making by patients, family members, and health
care providers at the end of life. However, the decisions about
participation in nonintrusive, noninvasive medical research is less
complicated than the end-of-life decisions about organ donation or
cessation of life-sustaining interventions (Truog,
2003).
Hermann and Looney (2001) state that
although “the hospice movement throughout the years has emphasized
the importance of symptom control” (p. 88), research has lagged
behind and “information related to the effectiveness of
interventions is currently lacking in the literature” (p. 89). Jubb
(2002) finds that there is no justification for
not going forward with improved research in palliative care. The
literature includes passionate as well as contradictory perspectives on
research with hospice and terminally ill patients. Given the ambivalence
in the literature, there is concern that these unresolved beliefs affect
participation in research projects and the ability of science to evaluate
or develop clinical practices for the terminally ill.
The future of palliative medicine and hospice research is affected by
uncertainty about the appropriateness of enrolling terminally ill patients
in research studies. Institutional Review Boards (IRBs) are mandated by
Federal law to protect research participants. Some IRB committees have
raised questions about the appropriateness and safety of involving
seriously ill individuals in research protocols that involve palliative or
other medical intervention surveys, interviews, and even questionnaires as
well as higher risk studies of medical treatments. As noted by Casarett
and Karlawish, (2000) some IRB committees have
suggested that these patients should never be asked to participate in
research, or if they are, only with a very restrictive set of safeguards.
More often IRBs, clinicians, and researchers themselves are uncomfortable
with potentially vulnerable patients, yet have no evidence to either
support or refute the source of the discomfort. Information on the source
of these concerns is mainly developed from experts in research ethics,
but, ironically, is not grounded by any research or even qualitative
information. The inevitable result of this approach is that information
about the needs and treatment responses of the terminally ill is
exclusively filtered through expert opinion or care providers who may have
vested interests in sustaining current practices. On the other hand, many
palliative care researchers think that the opportunity to participate in
research protocols can be a positive and meaningful experience for
patients (Janssens & Gordijn, 2000; Addington-Hall, 2002; Hudson,
2003). Based on feelings voiced to them by those patients who have
been involved in studies, the chance to meaningfully contribute to the
care of others meets deep-seated needs for generativity and altruism for
some patients. This perspective suggests that instead of research being an
unwanted intrusion into patients' lives it might be a way for
patients to contribute to others and to see some meaning in their
suffering.
Hospice care patients offer unique challenges to health care
professionals due to the advanced nature of their illnesses and the
interplay of physical, psychological, social, and spiritual problems. Care
providers often have concern about protecting patients and their family
members from what may be perceived as “intrusive” research
activities (Ross & Cornbleet, 2003). These
concerns, which have been raised by IRBs as well as practitioners,
researchers, and ethicists, appear to center around the vulnerability of
dying patients and difficulty in determining the patient's competency
to give informed consent, as well as ethical concerns for practitioners in
dual roles. However, concerns about patients' competence to give
truly informed consent to treatment have been addressed by showing that
special algorithms or multimedia approaches for assessing competence and
comprehension among those with cognitive difficulties can allay worry
about an individual patient's capacity to give informed consent
(Grisso & Applebaum, 1995; Daugherty et al., 1997; Wirshing et
al., 1998; Berg et al., 2001). And,
conversely, there is reason to worry about overly rigid and literal
interpretations of consent processes that could impede any advance of
medical science (Pellegrino, 1998).
In fact, there is considerable concern regarding the consequences of
not doing research, which results in a serious lack of
evidence-based interventions, which to some, is a patient right (Thyer & Myers, 1998). To date, little
investigation has been done to assess the attitudes of hospice health care
professionals regarding their willingness to participate in research. The
purpose of this pilot study is to assess the attitudes and beliefs about
research with hospice patients and their family members.
As a step toward developing research in hospice and palliative care
settings, this project was designed to explore the prevalence of positive
and negative beliefs about research among hospice providers. Although a
recent study attempted to undertake a similar tact (Ross & Cornbleet, 2003), the researchers focused
only on inpatient care and sampled a mix of both patients and providers.
However, this study's findings were limited by a very small number of
nurses (n = 13) as the provider sample. Physicians, social
workers, and other bereavement staff were not included in the study. The
overall goal of this study was to survey a large number of provider staff
across multiple disciplines to better understand hospice staff attitudes,
beliefs, and values about research with hospice patients and family
members. The specific objectives were to survey staff beliefs, attitudes,
and values about research at Hospice of the Bluegrass, located in
Lexington, Kentucky, and to examine differences in those issues among
different professions and clinical and nonclinical staff.
METHODS
Participants
Hospice of the Bluegrass has over 400 staff in 23 counties in central,
northern, and eastern Kentucky with eight different offices or treatment
sites. The staff members include a mix of nurses, certified nursing
assistants, social workers, chaplains, bereavement counselors, physicians,
and administrative and clerical staff. Four hundred members of the Hospice
of the Bluegrass were approached with the opportunity to fill out the
anonymous survey in this IRB-approved study.
Instruments
Hospice Survey
The survey is a 14-item instrument that asks key questions about
providers' thoughts about research (see the Appendix). This included
areas such as informed consent, decision making about participation, and
the perceived benefits and risks of participation in research. The survey
included a cover page with a script that introduced the study.
Procedure
This study utilized a cross-sectional survey design to measure hospice
staff members' beliefs, attitudes, and values using an anonymous
survey instrument. The survey contains questions derived from the
literature including literature on informed consent by hospice patients
and the terminally ill. The survey contains 14 questions and staff were
able to complete it in less than 5 min.
Staff members who attended team meetings were invited to participate
in the survey during the meeting or immediately following the conclusion
of the meeting. The survey was anonymous and included no identifying
information. Staff members who did not wish to participate were under no
pressure to respond to the survey. Staff members who did not attend the
staff meetings (such as those on leave) had an opportunity to respond to
the survey. Additional copies of the survey were left at each hospice
office for distribution to office mailboxes of staff who were not at the
meeting. Self-addressed envelopes were attached to each instrument so that
staff members could seal the survey and mail it to the researchers.
After completing the survey, participants were instructed to fold the
survey and place it in the addressed envelope and seal it. The sealed
envelops were collected and mailed to the research team at the University.
Supervisory staff did not see unsealed and completed instruments. Data
were entered by University staff into SPSS for analysis purposes.
The limited geographic data that were collected were not
cross-referenced with individual interviews to protect confidentiality and
anonymity. Participants completed the survey either at their work location
or by taking the form home and returning it by mail. The locations for
distribution of the survey were the offices of the Hospice of the
Bluegrass located throughout central, northern, and eastern Kentucky.
RESULTS
A total of 225 employees (56.25% return rate) completed the survey and
returned it for inclusion in this study. The respondents had an average of
4.7 years of hospice-related service (SD = 4.0, range = 1 month
to 20 years). The most common hospice roles occupied by the respondents
were nurse or certified nursing assistant (n = 67, 30%),
nonclinical staff (n = 62, 28%), social work (n = 40,
18%), home care aide (n = 34, 15%), bereavement counselor
(n = 11, 5%), and chaplain (n = 9, 4%).
The first seven items of the survey asked respondents to rate their
views toward research on a 1–5 scale, with 1 representing
strongly disagree and 5 representing strongly agree. The
items covered the perceived importance of conducting research with family
members, surveys of staff members, offering patients the chance to
participate in research, offering patients specific drug trials, offering
psychosocial treatment trials, and research on family members to evaluate
the effectiveness of new services and programs. There was a strong
agreement for all the items. Means ranged from 4.08 to 4.44 (SD
ranges = 0.88–1.07), indicating agreement or strong agreement to the
importance of research in the aforementioned areas.
Along these same lines, one of the later items asked respondents
“In general, my view of research with hospice patients is positive
if patients are free from coercion to participate,” utilizing the
same 1–5 scale. A total of 86% of the respondents agreed or strongly
agreed to this statement, with an additional 7% answering in the neutral
position. Only 5% disagreed or strongly disagreed with the statement.
It was also of interest to determine attitudes about who might be the
most appropriate contact person to approach hospice patients about
research. The majority (n = 117, 52%) felt that the hospice
clinician should approach the patients, whereas 37% (n = 87) felt
that it should be done by the researcher. A minority (n = 22,
10%) felt that the caregiver should be the one to present the research.
Given this strong desire to have hospice staff act as gatekeepers for
research, it was important to note how much time staff felt they could
devote to research. Twenty-five percent (n = 56) reported that
they had no additional time for research, and an additional 34%
(n = 77) reported they could spend up to 10 min on research
activity with a patient. Few reported that they could spend 20 min
(n = 16, 7%), 30 min (n = 11, 5%), or more (n =
2, 0.9%) on research. Thus, there is interest and a desire to control
access, but little perceived time for engaging in research activity.
Respondents were also asked to rank perceived barriers to conducting
hospice research. The number one perceived barrier to conducting research
was reported as research being “too intrusive to privacy”
(n = 110, 49%), followed closely by a sense that it “takes
too much time” (n = 108, 48%). The next tier of perceived
barriers included fears that research “interferes with patient
care” (n = 53, 24%), that “staff are not
consulted” (n = 46, 20%), and that “patients can not
give informed consent” (n = 36, 16%). Finally, although
infrequent, some felt that “hospice does not gain anything from
research” (n = 5, 2%) and that “research is not that
important” (n = 3, 1%). To counteract perceived barriers,
respondents also listed factors that could increase staff participation in
research activities as well as identifying what research topics were of
most importance and interest. The top five findings of each query are
listed in Table 1.
Top five factors to increase research participation and the top five
topics of interest
As a final quantitative analysis, it was of interest to determine if
the profession of the respondent or the time spent working for hospice had
an impact on the survey items and overall attitude toward research. To
this end, a series of correlations were conducted. However, neither
profession nor length of time in employment by hospice was found to be
significantly related to any of the items covering the perceived
importance of research.
Qualitative Data
We included open-ended questions on the survey tool for gathering
qualitative data. Specifically, the last item was an open-ended call for
any other pertinent comments that respondents might want to share.
Highlights of these comments are listed in Table
2.
Positive and negative quotes about conducting research with hospice
patients, family members, and staff
DISCUSSION
This survey provides an opportunity to revisit some of the issues that
affect how research will be done with hospice patients and the terminally
ill. Barriers to research with medically fragile populations have
generally included negative beliefs by the provider professions about
doing research with hospice patients. However, this study suggests that
there are important factors that can influence acceptance of research in
hospice programs.
There are several limitations to this study. There are methodological
issues to be addressed when conducting surveys. For instance, it is
generally known that those individuals who choose not to respond to
surveys are different than those who choose to participate (Groves, 1989; Groves & Couper,
1998). Therefore, there is certainly some degree of nonresponder
bias evident in these findings. However, although a response rate closer
to 100% in this survey would have improved its generalizability, the
return rate was viable for analyses and may have at least cautious
generalizability. For instance, Clark et al. (2001) found survey response rates from medical
professionals to be between 22 and 29% depending on the type of paper
stock that was used to print the surveys. In other medical populations,
researchers have achieved response rates of 69% (Sunshine & Bansal, 1995). Closer to our findings,
Fowler et al. (2002) achieved a 46% response
rate to mail surveys among health plan subscribers, which was increased to
66% when paired with an aggressive telephone campaign. Thus, our response
rate of 56.25% at least mirrors or improves on previous studies. Another
limitation is that the survey did not examine attitudes about the role of
private industry funding of research and the degree to which the goal of
commercial gain might erode trust in the research process (Michels, 2004).
Given the shortcomings of survey methodology, there nonetheless remain
some important lessons from these findings. For instance, it is apparent
that most staff members in hospice are at least open to the
“idea” of research being conducted with patients and family
members. However, we uncovered a rather intriguing dynamic regarding this
issue. On the one hand, staff felt that they should be largely responsible
for presenting research projects to patients and families. This
potentially reflects a certain paternalistic desire to protect patients
and to be “gatekeepers” from potential harm and abuse. On the
other hand, however, there was a clear trend toward reports of having very
little to no time to actually engage in the process of research with
patients. This likely reflects a sense of apprehension about adding even
more requirements and duties to an already stressful job. Thus, a form of
“catch-22” occurs in which staff are interested in research
and want to control access to it but won't get involved and actually
let it happen because they do not have time to absorb the extra burden it
would create. This could be a fruitful dichotomy to explore further in
larger trials.
On a final note, the survey also was useful for identifying the need
to engage in staff education about the purposes and real-world functioning
of research. The finding that concerns over protection of privacy ranked
second on their list of barriers is indicative of this need. Clearly,
staff members with this concern are unaware of the modern guidelines and
rigor imposed by IRBs and human protections concerns in our country.
Although these issues still certainly exist in places and some variability
exists between institutions and sites that review protocols (Maloney, 2003; McWilliams et al.,
2003), IRBs are in place and are challenged to apply federal
standards in a consistent manner (Wagner, 2003).
Therefore, education might be a good first step to build on this
study's findings. Indeed, it might be of interest to study the impact
of research training on hospice staff to determine if this changes
perceptions and desire to conduct and/or allow research to occur with
hospice patients and family members. Also, it is unclear form this survey
whether staff would feel an equal need to introduce research participation
to their patients in every study. Perhaps, were staff brought into the
study early in the process and given an opportunity to question
confidentiality and other protections, they may have their protective
interests satisfied. In addition, researchers might approach
administrators to examine ways in which research activities can be done in
the least intrusive way for staff. In conclusion, this study clearly shows
support from Hospice provider staff for conducting research with their
patients and family members. It also suggests the barriers that need to be
further examined in future studies.
ACKNOWLEDGMENTS
We thank the Hospice of the Bluegrass Research Committee members:
Janet Snapp, Kathleen Rutledge, Susan Byers, Lori Poe, Terry Gutgsell,
Cindy Cummings, Debra Kelly, Eugenia Smither, Marleen Burns, and Michelle
Staton Tindall.
APPENDIX: HOSPICE SURVEY USED IN THE
STUDY
